7.1.0.11

* fix tests on ghc 8.10.7

* fix test

* fix schema dump test

---------

Co-authored-by: Evgeny @ SimpleX Chat <[email protected]>

.github/workflows/build.yml

@@ -9,6 +9,10 @@ on:
       - "v*"
   pull_request:
 
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+
 jobs:
 
 # =============================
@@ -62,8 +66,19 @@ jobs:
   build:
     name: "ubuntu-${{ matrix.os }}-${{ matrix.arch }}, GHC: ${{ matrix.ghc }}"
     needs: maybe-release
+    permissions:
+      contents: write
+      packages: write
     env:
       apps: "smp-server xftp-server ntf-server xftp"
+      opt: ${{ github.event_name == 'pull_request' && 'O0' || 'O2' }}
+      ghc_options: ${{ github.event_name == 'pull_request' && '--ghc-options=-O0' || '' }}
+      release: ${{ startsWith(github.ref, 'refs/tags/v') }}
+      postgres_dist: ${{ startsWith(github.ref, 'refs/tags/v') && 'dist-newstyle' || 'dist-postgres' }}
+      branch: ${{ github.base_ref || github.ref_name }}
+      cache_o0: ${{ github.event_name == 'push' && (github.ref == 'refs/heads/master' || github.ref == 'refs/heads/stable') }}
+      image_cache: ghcr.io/simplex-chat/simplexmq-build-cache:${{ matrix.os }}-${{ matrix.arch }}-ghc${{ matrix.ghc }}
+      source_hashes: dist-newstyle/sources.sha256
     runs-on: ${{ matrix.runner }}
     services:
       postgres:
@@ -142,6 +157,18 @@ jobs:
           sudo apt update
           sudo apt -y install postgresql-client-15
 
+      - name: Log in to GitHub Container Registry
+        if: matrix.should_run == true && env.release != 'true'
+        shell: bash
+        env:
+          GHCR_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: echo "$GHCR_TOKEN" | docker login ghcr.io -u "${{ github.actor }}" --password-stdin
+
+      - name: Compute image cache week
+        id: week
+        if: matrix.should_run == true && env.release != 'true'
+        run: echo "week=$(date -u +%G-W%V)" >> "$GITHUB_OUTPUT"
+
       - name: Build and cache Docker image
         if: matrix.should_run == true
         uses: simplex-chat/docker-build-push-action@v6
@@ -153,20 +180,14 @@ jobs:
           build-args: |
             TAG=${{ matrix.os }}
             GHC=${{ matrix.ghc }}
-
-      - name: Cache dependencies
-        if: matrix.should_run == true
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cabal/store
-            dist-newstyle
-          key: ubuntu-${{ matrix.os }}-${{ matrix.arch }}-ghc${{ matrix.ghc }}-${{ hashFiles('cabal.project', 'simplexmq.cabal') }}
+          cache-from: ${{ env.release != 'true' && format('type=registry,ref={0}-{1}', env.image_cache, steps.week.outputs.week) || '' }}
+          cache-to: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' && format('type=registry,ref={0}-{1},mode=max,ignore-error=true', env.image_cache, steps.week.outputs.week) || '' }}
 
       - name: Start container
         if: matrix.should_run == true
         shell: bash
         run: |
+          mkdir -p ~/.cabal
           docker run -t -d \
             --device /dev/fuse \
             --cap-add SYS_ADMIN \
@@ -177,21 +198,66 @@ jobs:
             -v ${{ github.workspace }}:/project \
             build/${{ matrix.os }}:latest
 
+      - name: Compute cache key prefixes
+        id: keys
+        if: matrix.should_run == true && env.release != 'true'
+        run: |
+          toolchain=$(docker exec builder dpkg-query -W -f '${db:Status-Status} ${binary:Package} ${Version}\n' 'binutils*' 'cpp*' 'gcc*' '*-dev' | grep '^installed ' | sha256sum | cut -c1-16)
+          target=${{ matrix.os }}-${{ matrix.arch }}-ghc${{ matrix.ghc }}-$toolchain
+          echo "deps=deps-$target" >> "$GITHUB_OUTPUT"
+          echo "dist=dist-$target" >> "$GITHUB_OUTPUT"
+
+      - name: Restore dependencies from cache
+        id: restore-deps
+        if: matrix.should_run == true && env.release != 'true'
+        uses: actions/cache/restore@v4
+        with:
+          path: ~/.cabal/store
+          key: ${{ steps.keys.outputs.deps }}-${{ env.branch }}-${{ github.sha }}
+          restore-keys: |
+            ${{ steps.keys.outputs.deps }}-${{ env.branch }}-
+            ${{ steps.keys.outputs.deps }}-
+
+      - name: Warn if dependencies not in cache
+        if: matrix.should_run == true && env.release != 'true' && steps.restore-deps.outputs.cache-matched-key == ''
+        run: echo "::warning title=Cache miss::${{ steps.keys.outputs.deps }}-${{ env.branch }}"
+
+      - name: Restore build directories from cache
+        id: restore-build
+        if: matrix.should_run == true && env.release != 'true'
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            dist-newstyle
+            dist-postgres
+          key: ${{ steps.keys.outputs.dist }}-${{ env.opt }}-${{ env.branch }}-${{ github.sha }}
+          restore-keys: |
+            ${{ steps.keys.outputs.dist }}-${{ env.opt }}-${{ env.branch }}-
+            ${{ steps.keys.outputs.dist }}-${{ env.opt }}-
+
+      - name: Warn if build directories not in cache
+        if: matrix.should_run == true && env.release != 'true' && steps.restore-build.outputs.cache-matched-key == ''
+        run: echo "::warning title=Cache miss::${{ steps.keys.outputs.dist }}-${{ env.opt }}-${{ env.branch }}"
+
+      - name: Set old timestamps on unchanged sources (GHC 8.10.7)
+        if: matrix.ghc == '8.10.7' && steps.restore-build.outputs.cache-matched-key != '' && hashFiles(env.source_hashes) != ''
+        run: scripts/ci/source-timestamps.sh restore ${{ env.source_hashes }}
+
       - name: Build smp-server, xftp-server (postgresql) and tests
         if: matrix.should_run == true
         shell: docker exec -t builder sh -eu {0}
         run: |
-          chmod -fR 777 ~/.cabal ./dist-newstyle || :; git config --global --add safe.directory '*'
-          cabal clean
+          chmod -fR 777 ~/.cabal ./dist-newstyle ./dist-postgres || :; git config --global --add safe.directory '*'
+          scripts/ci/prune-old-builds.sh dist-newstyle dist-postgres
           cabal update
-          cabal build --jobs=$(nproc) --enable-tests -fserver_postgres
+          cabal build --builddir=${{ env.postgres_dist }} --jobs=$(nproc) --enable-tests -fserver_postgres ${{ env.ghc_options }}
           mkdir -p /out
-          for i in smp-server xftp-server simplexmq-test; do
-            bin=$(find /project/dist-newstyle -name "$i" -type f -executable)
+          for i in smp-server xftp-server ntf-server simplexmq-test; do
+            bin=$(find /project/${{ env.postgres_dist }} -name "$i" -type f -executable)
             chmod +x "$bin"
-            mv "$bin" /out/
+            cp "$bin" /out/
           done
-          strip /out/smp-server /out/xftp-server
+          strip /out/smp-server /out/xftp-server /out/ntf-server
 
       - name: Copy simplexmq-test from container
         if: matrix.should_run == true
@@ -224,16 +290,17 @@ jobs:
           cat hashes.output >> "$GITHUB_OUTPUT"
 
       - name: Build everything else (standard)
+        id: build-standard
         if: matrix.should_run == true
         shell: docker exec -t builder sh -eu {0}
         run: |
-          cabal build --jobs=$(nproc)
+          cabal build --jobs=$(nproc) ${{ env.ghc_options }}
           mkdir -p /out
-          for i in ${{ env.apps }}; do
+          for i in smp-server xftp-server xftp; do
             bin=$(find /project/dist-newstyle -name "$i" -type f -executable)
             strip "$bin"
             chmod +x "$bin"
-            mv "$bin" /out/
+            cp "$bin" /out/
           done
 
       - name: Copy binaries from container and prepare them
@@ -278,6 +345,33 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
+      - name: Record source hashes (GHC 8.10.7)
+        if: matrix.should_run == true && matrix.ghc == '8.10.7'
+        shell: docker exec -t builder sh -eu {0}
+        run: scripts/ci/source-timestamps.sh record ${{ env.source_hashes }}
+
+      - name: Compute dependency cache key
+        id: hash-deps
+        if: matrix.should_run == true && env.release != 'true'
+        shell: bash
+        run: echo "hash=$(ls ~/.cabal/store/ghc-*/ | sha256sum | cut -c1-16)" >> "$GITHUB_OUTPUT"
+
+      - name: Save dependencies to cache
+        if: matrix.should_run == true && env.release != 'true' && steps.restore-deps.outputs.cache-matched-key != format('{0}-{1}-{2}', steps.keys.outputs.deps, env.branch, steps.hash-deps.outputs.hash)
+        uses: actions/cache/save@v4
+        with:
+          path: ~/.cabal/store
+          key: ${{ steps.keys.outputs.deps }}-${{ env.branch }}-${{ steps.hash-deps.outputs.hash }}
+
+      - name: Save build directories to cache
+        if: matrix.should_run == true && env.release != 'true' && steps.restore-build.outputs.cache-hit != 'true'
+        uses: actions/cache/save@v4
+        with:
+          path: |
+            dist-newstyle
+            dist-postgres
+          key: ${{ steps.keys.outputs.dist }}-${{ env.opt }}-${{ env.branch }}-${{ github.sha }}
+
       - name: Test
         if: matrix.should_run == true && matrix.arch == 'x86-64'
         timeout-minutes: 120
@@ -302,6 +396,52 @@ jobs:
             exit 1
           fi
 
+      - name: Remove -O2 build directories
+        id: remove-o2
+        if: env.cache_o0 == 'true' && !cancelled() && steps.build-standard.outcome == 'success'
+        shell: docker exec -t builder sh -eu {0}
+        run: rm -rf dist-newstyle dist-postgres
+
+      - name: Restore -O0 build directories from cache
+        id: restore-build-o0
+        if: ${{ !cancelled() && steps.remove-o2.outcome == 'success' }}
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            dist-newstyle
+            dist-postgres
+          key: ${{ steps.keys.outputs.dist }}-O0-${{ env.branch }}-${{ github.sha }}
+          restore-keys: |
+            ${{ steps.keys.outputs.dist }}-O0-${{ env.branch }}-
+            ${{ steps.keys.outputs.dist }}-O0-
+
+      - name: Set old timestamps on unchanged sources for -O0 build (GHC 8.10.7)
+        if: ${{ !cancelled() && matrix.ghc == '8.10.7' && steps.restore-build-o0.outputs.cache-matched-key != '' && hashFiles(env.source_hashes) != '' }}
+        run: scripts/ci/source-timestamps.sh restore ${{ env.source_hashes }}
+
+      - name: Build at -O0 for pull request caches
+        id: build-o0
+        if: ${{ !cancelled() && steps.remove-o2.outcome == 'success' }}
+        shell: docker exec -t builder sh -eu {0}
+        run: |
+          scripts/ci/prune-old-builds.sh dist-newstyle dist-postgres
+          cabal build --builddir=dist-postgres --jobs=$(nproc) --enable-tests -fserver_postgres --ghc-options=-O0
+          cabal build --jobs=$(nproc) --ghc-options=-O0
+
+      - name: Record source hashes for -O0 build (GHC 8.10.7)
+        if: ${{ !cancelled() && matrix.ghc == '8.10.7' && steps.build-o0.outcome == 'success' }}
+        shell: docker exec -t builder sh -eu {0}
+        run: scripts/ci/source-timestamps.sh record ${{ env.source_hashes }}
+
+      - name: Save -O0 build directories to cache
+        if: ${{ !cancelled() && steps.build-o0.outcome == 'success' && steps.restore-build-o0.outputs.cache-hit != 'true' }}
+        uses: actions/cache/save@v4
+        with:
+          path: |
+            dist-newstyle
+            dist-postgres
+          key: ${{ steps.keys.outputs.dist }}-O0-${{ env.branch }}-${{ github.sha }}
+
 # =============================
 #      Resolver test job
 # =============================
@@ -316,15 +456,14 @@ jobs:
       - name: Clone project
         uses: actions/checkout@v3
 
-      - name: Set up Python
-        # Matches the runtime stage of scripts/resolver/service/Dockerfile.
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.13"
+      - name: Set up uv
+        # uv takes Python from .python-version, the version of the Dockerfile's runtime stage.
+        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
 
-      - name: Install resolver dependencies
-        # Must match scripts/resolver/service/pyproject.toml.
-        run: python -m pip install "eth-hash[pycryptodome]>=0.7"
+      - name: Lint
+        working-directory: scripts/resolver/service
+        run: uv run --frozen ruff check .
 
       - name: Test
-        run: python -m unittest discover -s scripts/resolver/service -v
+        working-directory: scripts/resolver/service
+        run: uv run --frozen pytest -v

.gitignore

@@ -1,4 +1,8 @@
 *.lock
+# the resolver's Docker image and CI install exactly these
+!/scripts/resolver/service/uv.lock
+# local overrides of scripts/resolver/.env.example
+/scripts/resolver/.env
 *.db
 *.db.bak
 *.session.sql

cabal.project

@@ -31,7 +31,7 @@ source-repository-package
 source-repository-package
     type: git
     location: https://github.com/simplex-chat/direct-sqlcipher.git
-    tag: f814ee68b16a9447fbb467ccc8f29bdd3546bfd9
+    tag: 2330df3dc8c4b660674d3c02d76bb86f66c2abee
 
 source-repository-package
     type: git

cbits/sntrup761.c

@@ -719,23 +719,32 @@ Small_random (small * out, void *random_ctx, sntrup761_random_func * random)
 
 /* ----- Streamlined NTRU Prime Core */
 
+/* x^p-x-1 has a degree-19 factor mod 3, so a random g is not invertible in R3
+   with probability about 3^-19; KeyGen_attempts failures in a row mean a broken RNG */
+#define KeyGen_attempts 10
+
 /* h,(f,ginv) = KeyGen() */
-static void
+/* returns 0 if KeyGen succeeded; else -1 */
+static int
 KeyGen (Fq * h, small * f, small * ginv, void *random_ctx,
         sntrup761_random_func * random)
 {
   small g[p];
   Fq finv[p];
+  int i;
 
-  for (;;)
+  for (i = 0; i < KeyGen_attempts; ++i)
     {
       Small_random (g, random_ctx, random);
       if (R3_recip (ginv, g) == 0)
         break;
     }
+  if (i == KeyGen_attempts)
+    return -1;
   Short_random (f, random_ctx, random);
   Rq_recip3 (finv, f);          /* always works */
   Rq_mult_small (h, finv, g);
+  return 0;
 }
 
 /* c = Encrypt(r,h) */
@@ -884,18 +893,21 @@ typedef small Inputs[p];        /* passed by reference */
 #define PublicKeys_bytes Rq_bytes
 
 /* pk,sk = ZKeyGen() */
-static void
+/* returns 0 if KeyGen succeeded; else -1 */
+static int
 ZKeyGen (unsigned char *pk, unsigned char *sk, void *random_ctx,
          sntrup761_random_func * random)
 {
   Fq h[p];
   small f[p], v[p];
 
-  KeyGen (h, f, v, random_ctx, random);
+  if (KeyGen (h, f, v, random_ctx, random) != 0)
+    return -1;
   Rq_encode (pk, h);
   Small_encode (sk, f);
   sk += Small_bytes;
   Small_encode (sk, v);
+  return 0;
 }
 
 /* C = ZEncrypt(r,pk) */
@@ -960,19 +972,21 @@ HashSession (unsigned char *k, int b, const unsigned char *y,
 /* ----- Streamlined NTRU Prime */
 
 /* pk,sk = KEM_KeyGen() */
-void
+int
 sntrup761_keypair (unsigned char *pk, unsigned char *sk, void *random_ctx,
                    sntrup761_random_func * random)
 {
   int i;
 
-  ZKeyGen (pk, sk, random_ctx, random);
+  if (ZKeyGen (pk, sk, random_ctx, random) != 0)
+    return -1;
   sk += SecretKeys_bytes;
   for (i = 0; i < PublicKeys_bytes; ++i)
     *sk++ = pk[i];
   random (random_ctx, Inputs_bytes, sk);
   sk += Inputs_bytes;
   Hash_prefix (sk, 4, pk, PublicKeys_bytes);
+  return 0;
 }
 
 /* c,r_enc = Hide(r,pk,cache); cache is Hash4(pk) */

cbits/sntrup761.h

@@ -19,7 +19,8 @@
 
 typedef void sntrup761_random_func (void *ctx, size_t length, uint8_t *dst);
 
-void
+/* returns 0 on success, -1 if the RNG never produced an invertible polynomial */
+int
 sntrup761_keypair (uint8_t *pk, uint8_t *sk,
                    void *random_ctx, sntrup761_random_func *random);
 

contributing/CODE.md

@@ -7,6 +7,7 @@ This file provides guidance on coding style and approaches and on building the c
 When designing code and planning implementations:
 - Apply adversarial thinking, and consider what may happen if one of the communicating parties is malicious.
 - Formulate an explicit threat model for each change - who can do which undesirable things and under which circumstances.
+- Use the default cryptographic primitive for each purpose from the [primitive registry](../protocol/crypto-registry.md); any exception requires review, and the registry must be updated with every new primitive use or domain-separation string.
 
 ## Code Quality Standards
 

plans/2026-10-08-fast-rotation-old-queue-drain.md

@@ -0,0 +1,62 @@
+# Fast rotation: draining the old receive queue
+
+Recipient only. The sender is unchanged.
+
+## Decisions
+
+- New `RcvSwitchStatus` value:
+
+  | constructor      | encoding        |
+  |------------------|-----------------|
+  | `RSReceivedQEND` | `received_qend` |
+
+  - Stored in `rcv_queues.switch_status`.
+  - `canAbortRcvSwitch` returns `False` for it.
+- `qEndMsg`, by the queue that delivered `QEND`:
+  - the queue named in `QEND`: the named queue is marked deleted, `ICDeleteRcvQueue` is enqueued.
+  - the new queue: the named queue is set to `RSReceivedQEND` and stays subscribed.
+  - `NSCCreate` and `SWITCH QDRcv SPCompleted`: sent when the named queue had no `RSReceivedQEND`.
+- Deletion of an `RSReceivedQEND` queue:
+  - trigger: `OK` response to an `ACK` on that queue, without a `GET` lock on it.
+  - action: the queue is marked deleted, `ICDeleteRcvQueue` is enqueued.
+- Bound, a `cleanupManager` step, for each connection with an `RSReceivedQEND` queue:
+  - condition: another receive queue of the connection has an active subscription.
+  - `delete_errors + 1 < deleteErrorCount`: `delete_errors` is incremented.
+  - otherwise: the queue is deleted as above; the first `DEL` error of `ICDeleteRcvQueue` removes it locally.
+- `ackSMPMessage` result, `type QueueDrained = Bool`:
+  - `OK`: `True`.
+  - `MSG`: `False`; the message is written to `msgQ`.
+- Apps: `received_qend` is added to `RcvSwitchStatus`.
+- Migration `20261008_rcv_switch_status`, SQLite and Postgres:
+  - up: index `idx_rcv_queues_switch_status` on `rcv_queues(switch_status)`.
+  - down: `received_qend` is set to `NULL`, the index is dropped.
+
+## Mechanics
+
+- `Agent/Protocol.hs`: `RSReceivedQEND`.
+- `Agent/Store.hs`: `canAbortRcvSwitch`.
+- `Client.hs`: `QueueDrained`; `ackSMPMessage` returns it.
+- `Agent/Client.hs`: `sendAck` returns `QueueDrained`.
+- `Agent/Store/AgentStore.hs`: `getEndedRcvQueueConnIds`:
+
+  ```sql
+  SELECT conn_id FROM rcv_queues WHERE switch_status = ? AND deleted = 0
+  ```
+
+- `Agent.hs`:
+  - `deleteRcvQueueAsync`: `setRcvQueueDeleted`, then `ICDeleteRcvQueue`.
+  - `qEndMsg`.
+  - `ackQueueMessage`.
+  - `cleanupManager`: `expireEndedRcvQueues`.
+- Apps:
+  - `apps/ios/SimpleXChat/APITypes.swift`
+  - `apps/multiplatform/common/src/commonMain/kotlin/chat/simplex/common/model/SimpleXAPI.kt`
+
+## Tests
+
+- `QEND` on the new queue while a message waits on the old queue:
+  - the message is delivered;
+  - the old queue is deleted after the `ACK` that returns `OK`.
+- The same with a recipient restart before the `ACK`.
+- Old server stopped: the old queue is deleted by the bound.
+- Downgrade: `received_qend` is cleared, other switch statuses are kept.

plans/2026-10-09-auth-error-delay.md

@@ -0,0 +1,52 @@
+# Delayed AUTH error responses
+
+## Decisions
+
+- Responses that include `ERR AUTH` are sent at the next multiple of the delay after the start time.
+- Delayed response sets:
+
+  | Source | Responses sent together | Start time |
+  |---|---|---|
+  | `receive`, failed verification | verification and parsing errors of one received block | block receipt |
+  | `client`, command processing | responses and messages of one batch | batch read from `rcvQ` |
+  | `RFWD`, forwarded command | the `RRES` of that command | start of forwarded command processing |
+
+- A thread created by `forkClient` writes the responses to the send queue.
+- Delay 0: the responses are written to the send queue directly.
+- `authDelayExceeded`: incremented when processing took longer than the delay.
+  - Prometheus counter `simplex_smp_auth_delay_exceeded`.
+  - Not saved in the stats backup.
+  - Not in the daily stats log.
+- Configuration:
+
+  | Item | Value |
+  |---|---|
+  | INI section | `TRANSPORT` |
+  | INI key | `auth_error_delay_ms` |
+  | default | 50 |
+  | `ServerConfig` field | `authErrorDelay`, microseconds |
+  | test servers | 20 ms |
+
+- AUTH timing test: elapsed time of 5 requests, test servers' delay.
+- Out of scope: NTF and XFTP servers.
+
+## Mechanics
+
+- `Server.hs`:
+  - `isAuthError`: response is `ERR AUTH`.
+  - `sendResponses`: responses with `ERR AUTH` to `sendAfterDelay`, others to the send queue.
+  - `sendAfterDelay`: writes responses to the send queue at the next multiple of the delay.
+  - `receive`: block receipt time from the `rcvActiveAt` reading.
+  - `client`: batch time.
+  - `processForwardedCommand`: start time.
+- `Server/Env/STM.hs`: `authErrorDelay` field, `defaultAuthErrorDelayMs`.
+- `Server/Stats.hs`: `authDelayExceeded` in `ServerStats` and `ServerStatsData`.
+- `Server/Prometheus.hs`: `simplex_smp_auth_delay_exceeded`.
+- `Server/Main.hs`: INI value.
+- `Server/Main/Init.hs`: INI template line.
+- `simplexmq.cabal`: `timeit` removed from the test suite.
+- `tests/SMPClient.hs`: `authErrorDelay = 20000`.
+- `tests/ServerTests.hs`:
+  - delayed AUTH for wrong key, absent queue, absent link and suspended queue; immediate `SOK`.
+  - AUTH timing test measures elapsed time.
+- `tests/SMPProxyTests.hs`: delayed AUTH to forwarded `SEND` and `LGET`.

protocol/agent-protocol.md

@@ -85,7 +85,7 @@ SMP agent protocol has 2 main parts:
 
 ![Duplex connection procedure](./diagrams/duplex-messaging/duplex-creating.svg)
 
-The procedure of establishing a duplex connection is explained on the example of Alice and Bob creating a bi-directional connection consisting of two unidirectional (simplex) queues, using SMP agents (A and B) to facilitate it, and two different SMP routers (which could be the same router). It is shown on the diagram above and has these steps:
+The procedure of establishing a duplex connection is explained on the example of Alice (the initiating party) and Bob (the joining party) creating a bi-directional connection consisting of two unidirectional (simplex) queues, using SMP agents (A and B) to facilitate it, and two different SMP routers (which could be the same router). It is shown on the diagram above and has these steps:
 
 1. Alice requests the new connection from the SMP agent A using agent `createConnection` api function.
 2. Agent A creates an SMP queue on the router (using [SMP protocol](./simplex-messaging.md) `NEW` command) and responds to Alice with the invitation that contains queue information and the encryption keys Bob's agent B should use. The invitation format is described in [Connection link](connection-link-1-time-invitation-and-contact-address).
@@ -208,7 +208,7 @@ This syntax of decrypted SMP client message body is defined by `decryptedAgentMe
 Decrypted SMP message client body can be one of 4 types:
 - `agentConnInfo` - used by the initiating party when confirming reply queue - sent in `agentConfirmation` envelope.
 - `agentConnInfoReply` - used by accepting party, includes reply queue(s) in the initial confirmation - sent in `agentConfirmation` envelope.
-- `agentRatchetInfo` - used to pass additional information when renegotiating double ratchet encryption - sent in `agentRatchetKey` envelope.
+- `agentRatchetInfo` - used to pass additional information when renegotiating double ratchet encryption - sent in `agentRatchetKey` envelope. A key sent in reply to another key includes the hash of that key; agents do not reply to such keys.
 - `agentMessage` - all other agent messages.
 
 `agentMessage` contains these parts:
@@ -233,7 +233,8 @@ connInfo = *OCTET
 agentConnInfoReply = %s"D" smpQueues connInfo
 smpQueues = length 1*newQueueInfo ; NonEmpty list of reply queues
 agentRatchetInfo = %s"R" ratchetInfo
-ratchetInfo = *OCTET
+ratchetInfo = [answeredKeyHash *OCTET] ; bytes after answeredKeyHash are ignored
+answeredKeyHash = %s"0" / (%s"1" shortString) ; "0" in a key that starts renegotiation, otherwise SHA-256 of the two raw public keys of the answered key
 
 agentMessage = %s"M" agentMsgHeader aMessage
 agentMsgHeader = agentMsgId prevMsgHash

protocol/crypto-registry.md

@@ -0,0 +1,189 @@
+Revision 1, 2026-10-01
+
+# SimpleX Network: cryptographic primitive registry
+
+The cryptographic primitives, constructions and domain-separation strings used by this repository, and which of them new code may use. For the threat model see [Security](./security.md).
+
+Module aliases follow the codebase: `C` = [Simplex.Messaging.Crypto](../src/Simplex/Messaging/Crypto.hs), `LC` = [Crypto.Lazy](../src/Simplex/Messaging/Crypto/Lazy.hs), `CR` = [Crypto.Ratchet](../src/Simplex/Messaging/Crypto/Ratchet.hs), `SL` = [Crypto.ShortLink](../src/Simplex/Messaging/Crypto/ShortLink.hs), `KEM` = [Crypto.SNTRUP761](../src/Simplex/Messaging/Crypto/SNTRUP761.hs) and [its bindings](../src/Simplex/Messaging/Crypto/SNTRUP761/Bindings.hs), `BBS` = [Crypto.BBS](../src/Simplex/Messaging/Crypto/BBS.hs). Lengths are in bytes.
+
+## Table of contents
+
+- [Policy](#policy)
+- [Defaults for new code](#defaults-for-new-code)
+- [Primitives](#primitives)
+- [SimpleX box constructions](#simplex-box-constructions)
+- [Domain separation and KDF inputs](#domain-separation-and-kdf-inputs)
+- [Nonces and IVs](#nonces-and-ivs)
+- [Implementation sources](#implementation-sources)
+- [xftp-web (TypeScript)](#xftp-web-typescript)
+- [Tests](#tests)
+
+## Policy
+
+- New code uses the default primitive for its purpose from [Defaults for new code](#defaults-for-new-code).
+- Using any other primitive, or a primitive marked *restricted* outside its listed purpose, requires a written justification in the pull request and review by a maintainer responsible for cryptography.
+- *Legacy-only* primitives stay only for the existing wire formats and stored data listed below. Do not add call sites.
+- Every new HKDF info string or hash prefix must be unique, start with `SimpleX`, and be added to [Domain separation and KDF inputs](#domain-separation-and-kdf-inputs).
+- Changing a primitive, key length, KDF input, info string or nonce construction changes the wire format. It requires a protocol version and an update of this registry and of the affected `protocol/` specification.
+
+Status values used below:
+
+| Status | Meaning |
+|---|---|
+| Default | Approved and preferred for new code for this purpose |
+| Approved | Approved for new code for this purpose when the default does not fit |
+| Restricted | Approved only for the listed purpose (external interoperability or a single protocol) |
+| Legacy-only | Kept for existing wire formats or stored data; no new call sites |
+
+## Defaults for new code
+
+| Purpose | Default | Functions |
+|---|---|---|
+| Signature | Ed25519 | `C.sign'`, `C.verify'` with `C.SEd25519` |
+| Key agreement | X25519 | `C.generateKeyPair`, `C.dh'` |
+| Post-quantum KEM | sntrup761 | `KEM.sntrup761Keypair`, `sntrup761Enc`, `sntrup761Dec` |
+| Hybrid DH + KEM secret | HKDF-SHA512 over `dh \|\| kemSecret` with a new info string, as in `CR.rootKdf` | `C.hkdf` |
+| Authenticated encryption | XSalsa20-Poly1305 [SimpleX secretbox](#simplex-box-constructions) | `C.sbEncrypt`/`C.sbDecrypt`, `LC.sbEncryptTailTag` for streams |
+| Authenticated encryption with a DH secret | XSalsa20-Poly1305 [SimpleX crypto_box](#simplex-box-constructions) | `C.cbEncrypt`/`C.cbDecrypt` |
+| Forward-secret symmetric session | HKDF-SHA512 key chain | `C.sbcInit`, `C.sbcHkdf` |
+| KDF | HKDF-SHA512 | `C.hkdf` |
+| Hash commitment, derived identifier | SHA3-256 | `C.sha3_256` |
+| Certificate fingerprint | SHA-256 X.509 fingerprint | `C.signedFingerprint`, `C.certificateFingerprint` |
+| Randomness | ChaChaDRG seeded from system entropy | `C.newRandom`, `C.randomBytes` and the `random*` helpers |
+
+## Primitives
+
+| Primitive | Status | Purpose and protocol | Implementation | Lengths |
+|---|---|---|---|---|
+| Ed25519 | Default | SMP/NTF/XFTP queue and file keys (agent default `rcvAuthAlg`, `sndAuthAlg` in [Agent/Env/SQLite.hs](../src/Simplex/Messaging/Agent/Env/SQLite.hs)), short-link signatures and owner auth (`SL.encodeSign`, `SL.newOwnerAuth`), XRCP identity and session keys ([RemoteControl/Invitation.hs](../src/Simplex/RemoteControl/Invitation.hs)), agent service request signatures, client/service TLS certificates ([Transport/Credentials.hs](../src/Simplex/Messaging/Transport/Credentials.hs)) | crypton `Crypto.PubKey.Ed25519` via `C.sign'`/`C.verify'` | pub 32, priv 32, sig 64; X.509 SPKI 44 |
+| Ed448 | Approved | Server CA and online certificates (default `signAlgorithm = ED448` in [Server/CLI.hs](../src/Simplex/Messaging/Server/CLI.hs)); accepted for SMP command signatures and TLS | crypton `Crypto.PubKey.Ed448`; server key generation by the `openssl genpkey` CLI | pub 57, priv 57, sig 114; SPKI 69 |
+| X25519 | Default | Per-queue e2e and server-to-recipient `crypto_box` keys, SMP session DH, command authenticator (SMP v7+), proxy (PRXY/PFWD), notification tokens, XFTP chunk transport, XRCP hello and announcements | crypton `Crypto.PubKey.Curve25519` via `C.dh'` | pub 32, priv 32, secret 32; SPKI 44 |
+| X448 | Restricted | E2E double ratchet only: X3DH and DH ratchet (`CR.RatchetX448`, E2E v3) | crypton `Crypto.PubKey.Curve448` | pub 56, priv 56, secret 56 |
+| sntrup761 | Default | PQ KEM in double ratchet (E2E v3, agent v5+) and XRCP v1 hello | Vendored C [cbits/sntrup761.c](../cbits/sntrup761.c) via FFI; randomness from ChaChaDRG through `haskell_rng_func` ([Bindings/RNG.hs](../src/Simplex/Messaging/Crypto/SNTRUP761/Bindings/RNG.hs)) | pk 1158, sk 1763, ct 1039, shared 32 |
+| XSalsa20-Poly1305, SimpleX crypto_box | Default | SMP message bodies (server to recipient), per-queue e2e envelope, confirmations, proxy forwarding, notifications, XRCP hello, XFTP chunk transport | crypton `Crypto.Cipher.XSalsa`, `Crypto.MAC.Poly1305`; `C.cryptoBox`, `LC.cbInit` | key 32 (DH secret), nonce 24, tag 16 |
+| XSalsa20-Poly1305, SimpleX secretbox | Default | SMP transport block encryption (SMP v11+), short-link data (SMP v15+), XFTP file encryption, local file encryption ([Crypto/File.hs](../src/Simplex/Messaging/Crypto/File.hs)), XRCP session | same; `C.sbEncrypt`, `LC.sbEncryptTailTag`, `LC.sbInit` | key 32, nonce 24, tag 16 |
+| crypto_box authenticator | Approved | Deniable sender command authorization with X25519 queue keys (SMP v7+); agent currently creates Ed25519 keys | `C.cbAuthenticate`, `C.cbVerify`: `crypto_box(sha512(msg))` | 80 = 64 + 16 |
+| AES-256-GCM, 16-byte IV | Legacy-only | Double ratchet header and body (E2E v3, [pqdr.md](./pqdr.md)) | crypton `Crypto.Cipher.AES`, `Crypto.Cipher.Types`; `C.encryptAEAD`, `C.decryptAEAD`, `C.initAEAD`; J0 = GHASH(IV) as NIST SP 800-38D defines for non-96-bit IVs | key 32, IV 16, tag 16; padded header 88 (PQ off) or 2310 (PQ on), `CR.paddedHeaderLen` |
+| AES-256-GCM, 12-byte IV | Restricted | WebRTC frame encryption in simplex-chat (`Simplex.Chat.Mobile.WebRTC`), for WebCrypto interoperability | `C.encryptAESNoPad`, `C.decryptAESNoPad`, `C.initAEADGCM`, `C.GCMIV` | key 32, IV 12, tag 16 |
+| HKDF-SHA512 | Default | All KDFs listed in [Domain separation](#domain-separation-and-kdf-inputs) | crypton `Crypto.KDF.HKDF` with `SHA512`; `C.hkdf` (extract + expand) | output per use, at most 255 * 64 |
+| SHA-256 | Default for X.509 fingerprints, otherwise Restricted | X.509 fingerprints (`C.KeyHash`, server identity, XRCP CA, service certificate hash, SMP v16+), XFTP chunk digests, agent message hashes, `requestCode`, ratchet key dedup hash, security code `codeAD = sha256(rcAD)` | crypton; `C.sha256Hash`, `LC.sha256Hash`, `getFingerprint ... HashSHA256` | 32 |
+| SHA-512 | Restricted | XFTP file digests and file description hash, input of the crypto_box authenticator | crypton; `C.sha512Hash`, `LC.sha512Hash` | 64 |
+| SHA-512 (inside sntrup761) | Restricted | sntrup761 internal hash | [cbits/sha512.c](../cbits/sha512.c) calls OpenSSL `SHA512()` from the system `libcrypto` (`extra-libraries: crypto`) | 64 |
+| SHA3-256 | Default | Short-link key `sha3_256(fixedData)` (SMP v15+), XRCP hybrid secret, service request binding (agent v8) | crypton; `C.sha3_256`, `KEM.kemHybridSecret` | 32 |
+| SHA3-384 | Restricted | Client-supplied sender ID: first 24 bytes of `sha3_384(corrId)`, computed by the agent (`prepareConnectionLink'`, `newRcvConnSrv`) and the server (`createQueue`) | crypton; `C.sha3_384` | 48, truncated to 24 |
+| Keccak-256 | Restricted | Label hash for SMP name queries (`NameQuery NQHash`), must match the on-chain registry | crypton `Keccak_256`; `labelHash` in [SimplexName.hs](../src/Simplex/Messaging/SimplexName.hs) | 32 |
+| MD5 | Legacy-only, non-security | XOR-aggregated queue ID hash `IdsHash` for service subscription reconciliation (SUBS, NSUBS, SOKS, ENDS) | crypton (`C.md5Hash`, `Protocol.queueIdHash`); SQLite UDF `simplex_xor_md5_combine` ([Agent/Store/SQLite.hs](../src/Simplex/Messaging/Agent/Store/SQLite.hs)); PostgreSQL pgcrypto `digest(..., 'md5')` in server and NTF schemas | 16 |
+| ChaChaDRG | Default | Keys, nonces, correlation IDs, random IDs, sntrup761 randomness | crypton `Crypto.Random`; `C.newRandom` seeds with `drgNew` from system entropy | n/a |
+| BBS+ over BLS12-381, SHA-256 suite | Restricted | Badge entitlement credentials and proofs ([Crypto/Entitlement.hs](../src/Simplex/Messaging/Crypto/Entitlement.hs)), XFTP storage-time extension | Vendored submodules libbbs and blst (C and assembly) via FFI; randomness from libbbs `getentropy`, `SecRandomCopyBytes` with the `commoncrypto` flag, `BCryptGenRandom` on Windows ([cbits/getentropy_win.c](../cbits/getentropy_win.c)) | sk 32, pk 96, sig 80, proof 272 + 32 per undisclosed message |
+| ECDSA with SHA-256 (ES256) | Restricted | APNS provider JWT ([Push/APNS.hs](../src/Simplex/Messaging/Notifications/Server/Push/APNS.hs)) | crypton `Crypto.PubKey.ECC.ECDSA.sign`; key read by cryptostore `Crypto.Store.PKCS8`; curve taken from the key file | signature DER `SEQUENCE {r, s}`, base64url |
+| TLS 1.3 / 1.2 | Restricted | SMP, XFTP, NTF, XRCP transport: `TLS_CHACHA20_POLY1305_SHA256` (1.3), `ECDHE_ECDSA_CHACHA20POLY1305_SHA256` (1.2), groups X448 and X25519, Ed448 and Ed25519 signatures (`defaultSupportedParams`) | `tls` 1.9 | n/a |
+| TLS for browsers | Restricted | XFTP server with HTTPS credentials (`defaultSupportedParamsHTTPS`: `ciphersuite_strong`, FFDHE, P-521, ECDSA and RSA signatures); SMP server HTTPS requires RSA-4096 (`checkHTTPSCredentials`) | `tls`, `warp-tls` | n/a |
+| X.509 | Restricted | Certificate chains, fingerprints, signed session keys (`C.signX509`, `C.verifyX509`) | `crypton-x509`, `crypton-x509-validation`, `cryptostore` | n/a |
+| SQLCipher | Restricted | Agent SQLite database at rest (`PRAGMA key`); cipher parameters are SQLCipher defaults, not set in this repository | `direct-sqlcipher` (git dependency, [cabal.project](../cabal.project)) | n/a |
+
+## SimpleX box constructions
+
+`C.cryptoBox` and `LC.sbInit_` compute, for a 32-byte key `k` and 24-byte nonce `n`:
+
+```
+k1 = HSalsa20(k, 0^16)
+subkey = HSalsa20(k1, n[0..16])
+stream = Salsa20(subkey, n[16..24])
+polyKey = stream[0..32]
+ct = msg XOR stream[32..]
+tag = Poly1305(polyKey, ct)
+```
+
+| Construction | Key `k` | Equivalent libsodium call | Layout |
+|---|---|---|---|
+| SimpleX crypto_box (`C.cbEncrypt`, `LC.cbInit`) | X25519 shared secret | `crypto_box_easy` (`crypto_box_beforenm` is `HSalsa20(dh, 0^16)`) | strict: `tag \|\| ct`; lazy tail-tag: `ct \|\| tag` |
+| SimpleX secretbox (`C.sbEncrypt`, `LC.sbInit`, `KEM.kcb*`) | 32-byte symmetric key | `crypto_secretbox_easy` with key `crypto_core_hsalsa20(0^16, k)`, not with `k` | same |
+
+Padding before encryption: `C.pad` prefixes a 2-byte big-endian length and fills with `#` (message at most 65533 bytes); `LC.pad` prefixes an 8-byte length. `*NoPad` variants skip padding.
+
+## Domain separation and KDF inputs
+
+HKDF is `C.hkdf salt ikm info len` (HKDF-SHA512).
+
+| Info string | Salt | IKM | Output (split) | Function | Use |
+|---|---|---|---|---|---|
+| `"SimpleXX3DH"` | 64 zero bytes | `dh1 \|\| dh2 \|\| dh3 [\|\| kemShared]` | 96: `hk`, `nhk`, root key (32 each) | `CR.pqX3dh` | Ratchet initialization, added in E2E v2; KEM secret from E2E v3 (current minimum) |
+| `"SimpleXVerifyCode"` | 64 zero bytes | same as `SimpleXX3DH` | 32: `rcVCPQ` | `CR.pqX3dh` | Verification code covering all handshake keys, new ratchets |
+| `"SimpleXRootRatchet"` | root key | `dh [\|\| kemShared]` | 96: root key, chain key, next header key | `CR.rootKdf` | DH/PQ ratchet step |
+| `"SimpleXChainRatchet"` | empty | chain key | 96: chain key, message key (32 each), message IV (16), header IV (16) | `CR.chainKdf` | Per-message keys |
+| `"SimpleXSbChainInit"` | SMP session ID | X25519 session secret | 64: two 32-byte chain keys | `C.sbcInit` from `Transport.blockEncryption` | SMP transport block encryption, SMP v11+ |
+| `"SimpleXSbChainInit"` | empty | XRCP hybrid secret (below) | 64: two 32-byte chain keys | `C.sbcInit` in [RemoteControl/Client.hs](../src/Simplex/RemoteControl/Client.hs) | XRCP v1 session |
+| `"SimpleXSbChain"` | empty | chain key | 88: chain key (32), secretbox key (32), nonce (24) | `C.sbcHkdf` | Each SMP block and XRCP message |
+| `"SimpleXContactLink"` | empty | link key (32) | 56: link ID (24), secretbox key (32) | `SL.contactShortLinkKdf` | Contact short links, SMP v15+ |
+| `"SimpleXInvLink"` | empty | link key (32) | 32: secretbox key | `SL.invShortLinkKdf` | Invitation short links, SMP v15+ |
+
+Other derivations:
+
+| Value | Construction | Function | Use |
+|---|---|---|---|
+| Service request binding | `sha3_256("SimpleXService" \|\| rcAD)` | `serviceReqBinding` in [Agent.hs](../src/Simplex/Messaging/Agent.hs) | Agent RPC, agent v8 |
+| BBS header | `"SimpleX badges v1"` | `entitlementBBSHeader` | Badge credentials |
+| XRCP hybrid secret | `sha3_256(x25519Dh \|\| kemShared)` | `KEM.kemHybridSecret` | XRCP v1 |
+| Short-link key | `sha3_256(smpEncode FixedLinkData)` | `SL.encodeSignFixedData`, checked in `SL.decryptLinkData` | SMP v15+ |
+| Sender ID | `take 24 (sha3_384 corrId)` | `prepareConnectionLink'`, `newRcvConnSrv` in Agent.hs; `createQueue` in [Server.hs](../src/Simplex/Messaging/Server.hs) | Client-supplied sender ID with link data, SMP v15+; the server rejects a mismatch to prevent an ID oracle |
+| Ratchet associated data | `pubKeyBytes sk1 \|\| pubKeyBytes rk1` (raw X448, 112) | `CR.pqX3dh` | AD of every ratchet AEAD |
+| Security code | `sha256(rcAD)` | `ratchetVerifyCodes` in [AgentStore.hs](../src/Simplex/Messaging/Agent/Store/AgentStore.hs) | Connection verification |
+| Contact request code | `sha256(smpEncode (k1, k2, kem, sndId))` | `requestCode` in Agent.hs | Contact request binding |
+| Command authenticator | `crypto_box(sha512(authorized))` | `C.cbAuthenticate` | SMP v7+ |
+| Queue IDs hash | `xor` of `md5(queueId)` | `Protocol.queueIdsHash` | Service subscriptions |
+
+`"SimpleXSbChainInit"` and `"SimpleXSbChain"` are shared by SMP block encryption and XRCP. Their outputs are separated only by the IKM (and the salt for `sbcInit`).
+
+## Nonces and IVs
+
+| Use | Construction |
+|---|---|
+| SMP command authenticator, proxied command | `C.cbNonce corrId`, where `corrId` is a random 24-byte nonce (`C.randomCbNonce`) |
+| SMP proxied responses | `C.reverseNonce` of the request nonce |
+| Server-to-recipient message body | `C.cbNonce msgId`; `msgIdBytes = 24` in [Server/Main.hs](../src/Simplex/Messaging/Server/Main.hs) |
+| Per-queue e2e envelope, short-link data, notifications, XRCP hello, XFTP chunk download | Random 24-byte nonce, sent with the ciphertext |
+| XFTP file, local encrypted file | Random key and nonce per file (`C.randomSbKey`, `C.randomCbNonce`) |
+| SMP block, XRCP session messages | Key and nonce from `C.sbcHkdf`, one per block |
+| Ratchet header, body | 16-byte IVs from `CR.chainKdf`; header IV is sent, body IV is not |
+| WebRTC frames | 12-byte `C.GCMIV` supplied by the caller in simplex-chat |
+
+## Implementation sources
+
+| Source | Version or pin | Provides |
+|---|---|---|
+| crypton | 0.34 | Ed25519, Ed448, X25519, X448, XSalsa20, Poly1305, AES-GCM, HKDF, SHA-2, SHA-3, Keccak, MD5, ChaChaDRG, ECDSA |
+| tls, crypton-x509, crypton-x509-validation, cryptostore | 1.9.0, 1.7.6, 1.6.12, 0.3.0.1 | TLS, X.509, PKCS#8 |
+| [cbits/sntrup761.c](../cbits/sntrup761.c) | Copy of draft-josefsson-ntruprime-streamlined-00 ([cbits/README.md](../cbits/README.md)) | sntrup761 |
+| [cbits/sha512.c](../cbits/sha512.c) and system OpenSSL `libcrypto` | system | SHA-512 for sntrup761 |
+| `cbits/libbbs` submodule ([simplex-chat/libbbs](https://github.com/simplex-chat/libbbs)) | `59a0f4bf` | BBS+ (`bbs_sha256_ciphersuite`), SHA-256, SHAKE256 |
+| `cbits/blst` submodule ([supranational/blst](https://github.com/supranational/blst)) | `db3defd0` | BLS12-381 (C and `build/assembly.S`, `-D__BLST_PORTABLE__`) |
+| direct-sqlcipher | git `f814ee68` | SQLCipher |
+| `openssl` CLI | system | Server CA and certificate key generation (`createServerX509_`) |
+
+## xftp-web (TypeScript)
+
+[xftp-web](../xftp-web/) reimplements a subset for the browser XFTP client. It has no AES-GCM, HKDF, SHA-3, MD5, X448 or sntrup761 code.
+
+| Primitive | Implementation | Haskell counterpart |
+|---|---|---|
+| Ed25519 sign, verify, keys | libsodium-wrappers-sumo (`crypto_sign_*`), [src/crypto/keys.ts](../xftp-web/src/crypto/keys.ts) | `C.sign'`, `C.verify'` |
+| Ed448 verify | `@noble/curves/ed448`, keys.ts `verifyEd448` | `C.verify'` |
+| X25519 | libsodium `crypto_scalarmult`, `crypto_box_keypair` | `C.dh'` |
+| SimpleX crypto_box, secretbox, tail-tag streaming | Salsa20 block function written in TypeScript, libsodium `crypto_core_hsalsa20` and `crypto_onetimeauth_*`, [src/crypto/secretbox.ts](../xftp-web/src/crypto/secretbox.ts) | `C.cryptoBox`, `LC.sbInit_` |
+| crypto_box authenticator | [src/protocol/client.ts](../xftp-web/src/protocol/client.ts) `cbAuthenticate`, `cbVerify` | `C.cbAuthenticate`, `C.cbVerify` |
+| SHA-256, SHA-512 | libsodium `crypto_hash_sha256`, `crypto_hash_sha512*`, [src/crypto/digest.ts](../xftp-web/src/crypto/digest.ts) | `C.sha256Hash`, `C.sha512Hash` |
+| Random | WebCrypto `crypto.getRandomValues`; libsodium for key generation | `C.randomBytes` |
+
+## Tests
+
+None of the tests below compares against published reference vectors (NIST, RFC 8032, RFC 7748, RFC 5869, NaCl, the sntrup761 draft or the BBS draft). Interoperability is tested between this repository's own implementations.
+
+| Area | Test module (hspec group) | Kind |
+|---|---|---|
+| Ed25519, Ed448, X25519 crypto_box, secretbox, lazy and tail-tag secretbox, AES-GCM 12-byte IV, X.509 key encoding, X.509 chains, sntrup761, BBS+, entitlements, padding | [tests/CoreTests/CryptoTests.hs](../tests/CoreTests/CryptoTests.hs) | Round-trip; fixed lengths for BBS+; fixed bytes for padding |
+| Double ratchet (X25519 and X448), PQ KEM agreement, AES-GCM 16-byte IV | [tests/AgentTests/DoubleRatchetTests.hs](../tests/AgentTests/DoubleRatchetTests.hs) | Round-trip; decoding of a stored v2 ratchet JSON |
+| Short links (HKDF info strings, SHA3-256 link key) | [tests/AgentTests/ShortLinkTests.hs](../tests/AgentTests/ShortLinkTests.hs) | Round-trip, tamper rejection |
+| File encryption | [tests/CoreTests/CryptoFileTests.hs](../tests/CoreTests/CryptoFileTests.hs) | Round-trip |
+| XRCP session (SHA3-256 hybrid, sb chain) | [tests/RemoteControl.hs](../tests/RemoteControl.hs) | End-to-end |
+| SMP authenticators, block encryption, `IdsHash` (MD5) | [tests/ServerTests.hs](../tests/ServerTests.hs) | End-to-end; with the PostgreSQL queue store this checks Haskell MD5 against pgcrypto |
+| Haskell and xftp-web: SHA-256/512, Ed25519, Ed448 identity proof, X25519, DER keys, crypto_box, secretbox, tail-tag, authenticator, file encryption | [tests/XFTPWebTests.hs](../tests/XFTPWebTests.hs) (`XFTP Web Client`) | Byte-identical cross-language output |

protocol/diagrams/xrcp/session.mmd

@@ -17,11 +17,11 @@ note over CC, HC: TLS handshake
 par
     note over CC: validate client X509 credentials
     CC->>CI: session code from tlsUnique
-    CI-->>CC: user confirmation
+    CI-->>CC: user confirmation (optional if known)
 and
     note over HC: validate server X509 credentials
     HC->>HI: session code from tlsUnique
-    HI-->>HC: user confirmation
+    HI-->>HC: user confirmation (optional if known)
 end
 
 note over CI, HI: 3. Session verification and protocol negotiation

protocol/pqdr.md

@@ -62,14 +62,16 @@ It is possible to reduce size overhead by using only one KEM agreement and makin
 
 ## Double ratchet with encrypted headers augmented with double PQ KEM
 
-Algorithm below assumes that in addition to shared secret from the initial key agreement, there will be an encapsulation key available from the party that published its keys (Bob).
+Algorithm below assumes that in addition to shared secret from the initial key agreement, there will be an encapsulation key available from the initiating party, that sent its keys in the connection invitation to the joining party (see [agent protocol](./agent-protocol.md)). Following the double ratchet specification, the pseudo-code below names the joining party Alice and the initiating party Bob.
+
+Unlike X3DH prekey bundles, these keys are not reusable keys published for any party to use. The initiating party generates two X448 key pairs and, optionally, a sntrup761 KEM key pair for each connection, sends the public keys in the invitation, and deletes the stored private keys once the ratchet is initialized; only the second X448 private key and the agreed KEM key pair remain in the ratchet state as its initial ratchet keys. The joining party generates its keys when joining and keeps only its KEM key pair, in the ratchet state. The exception is a contact address that publishes ratchet keys in its link data: all requesters use these keys until the address owner rotates them, and the owner keeps the private keys of a few recent generations.
 
 ### Initialization
 
 The double ratchet initialization is defined in pseudo-code. This pseudo-code is identical to Signal algorithm specification except for that parts that add post-quantum key agreement.
 
 ```
-// Alice obtained Bob's keys and initializes ratchet first
+// Alice (joining party) received Bob's keys in the invitation and initializes ratchet first
 def RatchetInitAlicePQ2HE(state, SK, bob_dh_public_key, shared_hka, shared_nhkb, bob_pq_kem_encapsulation_key):
     state.DHRs = GENERATE_DH()
     state.DHRr = bob_dh_public_key
@@ -89,7 +91,7 @@ def RatchetInitAlicePQ2HE(state, SK, bob_dh_public_key, shared_hka, shared_nhkb,
     state.HKr = None
     state.NHKr = shared_nhkb
 
-// Bob initializes ratchet second, having received Alice's connection request
+// Bob (initiating party) initializes ratchet second, having received Alice's first message
 def RatchetInitBobPQ2HE(state, SK, bob_dh_key_pair, shared_hka, shared_nhkb, bob_pq_kem_key_pair):
     state.DHRs = bob_dh_key_pair
     state.DHRr = None
@@ -210,6 +212,8 @@ The outer envelope contains the encrypted header (used as associated data for bo
 
 The message body is encrypted with AES-256-GCM using the message key derived from the sending chain key (`KDF_CK`). The associated data for body encryption is the concatenation of the ratchet associated data and the encoded encrypted header.
 
+`KDF_CK(CK)` is HKDF-SHA512 with empty salt, `CK` as input key material and info `"SimpleXChainRatchet"`, producing 96 bytes split into the next chain key (32 bytes), the message key (32 bytes), the message body IV (16 bytes, not transmitted) and `headerIV` (16 bytes). Both IVs are used as 16-byte AES-256-GCM IVs, not the 12-byte IVs recommended by NIST SP 800-38D, so the initial counter block is J0 = GHASH(IV || 0^64 || [128]_64) as defined there for non-96-bit IVs. WebCrypto and other conforming implementations compute it when given the full 16-byte IV; truncating the IV to 12 bytes produces different ciphertext.
+
 ```abnf
 encRatchetMessage = versionedLength encMessageHeader msgAuthTag encMsgBody
 ; encMessageHeader is used as associated data for body decryption: AD = rcAD || encMessageHeader
@@ -274,7 +278,7 @@ As SimpleX Messaging Protocol pads messages to a fixed size, using 16kb transpor
 
 Sharing the initial keys in case of SimpleX Chat it is equivalent to sharing the invitation link. As encapsulation key is large, it may be inconvenient to share it in the link in some contexts, e.g. when QR codes are used.
 
-It is possible to postpone sharing the encapsulation key until the first message from Alice (confirmation message in SMP protocol), the party sending connection request. The upside here is that the invitation link size would not increase. The downside is that the user profile shared in this confirmation will not be encrypted with PQ-resistant algorithm.
+It is possible to postpone sharing the encapsulation key until the first message from the joining party (confirmation message in SMP protocol). The upside here is that the invitation link size would not increase. The downside is that the user profile shared in this confirmation will not be encrypted with PQ-resistant algorithm.
 
 Another consideration is pairwise ratchets in groups. Key generation in sntrup761 is quite slow - on slow devices it can be as slow as 10-20 keys per second, so using this primitive in groups larger than 10-20 members would result in slow performance.
 

protocol/security.md

@@ -15,7 +15,7 @@ This document describes the cryptographic primitives and threat model for the Si
   - [SimpleX Messaging Protocol router that proxies the messages to another SMP router](#simplex-messaging-protocol-router-that-proxies-the-messages-to-another-smp-router)
   - [An attacker who obtained Alice's (decrypted) chat database](#an-attacker-who-obtained-alices-decrypted-chat-database)
   - [A user's contact](#a-users-contact)
-  - [An attacker who observes Alice showing an introduction message to Bob](#an-attacker-who-observes-alice-showing-an-introduction-message-to-bob)
+  - [An attacker who observes the initiating party showing an introduction message to the joining party](#an-attacker-who-observes-the-initiating-party-showing-an-introduction-message-to-the-joining-party)
   - [An attacker with Internet access](#an-attacker-with-internet-access)
 
 
@@ -37,6 +37,8 @@ This document describes the cryptographic primitives and threat model for the Si
   - AES-GCM AEAD cipher,
   - SHA512-based HKDF for key derivation.
 
+All primitives in use, their lengths, domain-separation strings and the policy for new code are listed in the [cryptographic primitive registry](./crypto-registry.md).
+
 
 ## Threat Model
 
@@ -168,7 +170,7 @@ This document describes the cryptographic primitives and threat model for the Si
 
 *cannot:*
 
-- impersonate a sender and send messages to the user whose database was stolen. Doing so requires also compromising the router (to place the message in the queue, that is possible until the Double-Ratchet advances forward) or the user's device at a subsequent time (to place the message in the database).
+- impersonate a sender and send messages to the user whose database was stolen, unless the database contains a new rotation queue that is not secured yet. Doing so requires also compromising the router (to place the message in the queue, that is possible until the Double-Ratchet advances forward) or the user's device at a subsequent time (to place the message in the database).
 
 - undetectably communicate at the same time as Alice with her contacts. Doing so would result in the contact getting different messages with repeated IDs.
 
@@ -190,15 +192,15 @@ This document describes the cryptographic primitives and threat model for the Si
 
 - cannot collaborate with another of the user's contacts to confirm they are communicating with the same user.
 
-### An attacker who observes Alice showing an introduction message to Bob
+### An attacker who observes the initiating party showing an introduction message to the joining party
 
 *can:*
 
-- Impersonate Bob to Alice.
+- Impersonate the joining party to the initiating party.
 
 *cannot:*
 
-- Impersonate Alice to Bob.
+- Impersonate the initiating party to the joining party.
 
 ### An attacker with Internet access
 

protocol/simplex-messaging.md

@@ -86,7 +86,7 @@ It's designed with the focus on communication security and integrity, under the
 
 It is designed as a low level protocol for other application protocols to solve the problem of secure and private message transmission, making [MITM attack][1] very difficult at any part of the message transmission system.
 
-This document describes SMP protocol version 22. Versions 1-5 are discontinued. The version history:
+This document describes SMP protocol version 23. Versions 1-5 are discontinued. The version history:
 
 - v1: binary protocol encoding
 - v2: message flags (used to control notifications)
@@ -109,6 +109,7 @@ This document describes SMP protocol version 22. Versions 1-5 are discontinued.
 - v20: public namespaces resolver (RSLV command, RNAME response) — direct or forwarded via PFWD
 - v21: server public information in handshake
 - v22: `RNAME` says whether a name can be registered, not only what it resolves to
+- v23: version in nonces of forwarded commands, random nonces for forwarded responses
 
 ## Introduction
 
@@ -1120,13 +1121,15 @@ When the client receives PKEY response it MUST validate that:
 - the fingerprint of the received certificate matches fingerprint in the router address - it mitigates MITM attack by proxy.
 - the router session key is correctly signed with the received certificate.
 
+`smpVersionRange` in `PKEY` is not signed.
+
 The proxy router may respond with error response in case the destination router is not available or in case it has an earlier version that does not support proxied commands.
 
 #### Send command via proxy
 
 Sender can send `SKEY` and `SEND` commands via proxy after obtaining the session ID with `PRXY` command (see [Request proxied session](#request-proxied-session)).
 
-Transmission sent to proxy router should use session ID as entity ID and use a random correlation ID of 24 bytes as a nonce for crypto_box encryption of transmission to the destination router. The random ephemeral X25519 key to encrypt transmission should be unique per command, and it should be combined with the key sent by the router in the handshake header to proxy and to the client in `PKEY` command.
+Transmission sent to proxy router should use session ID as entity ID and use a random correlation ID of 24 bytes as a nonce for crypto_box encryption of transmission to the destination router. When `smpVersion` in `PFWD` is 23 or higher, the first 2 bytes of this nonce are XOR-ed with `smpVersion`. The random ephemeral X25519 key to encrypt transmission should be unique per command, and it should be combined with the key sent by the router in the handshake header to proxy and to the client in `PKEY` command.
 
 Encrypted transmission should use the received session ID from the connection between proxy router and destination router in the authorized body.
 
@@ -1138,10 +1141,11 @@ commandKey = length x509encoded
 
 The proxy router will forward the encrypted transmission in `RFWD` command (see below).
 
-Having received the `RRES` response from the destination router, proxy router will forward `PRES` response to the client. `PRES` response should use the same correlation ID as `PFWD` command. The destination router will use this correlation ID increased by 1 as a nonce for encryption of the response.
+Having received the `RRES` response from the destination router, proxy router will forward `PRES` response to the client. `PRES` response should use the same correlation ID as `PFWD` command. The destination router will use this correlation ID increased by 1 as a nonce for encryption of the response. When `smpVersion` in `PFWD` is 23 or higher, the destination router uses a random nonce instead, and sends it before the encrypted response.
 
 ```abnf
-proxyResponse = %s"PRES" SP <encrypted padded(forwardedResponse, 16226)>
+proxyResponse = %s"PRES" SP [responseNonce] <encrypted padded(forwardedResponse, 16226)>
+responseNonce = %s"0" / (%s"1" 24*24 OCTET) ; from v23
 forwardedResponse = *OCTET ; client-encrypted SMP response, decrypted by client using per-command DH secret
 ```
 
@@ -1168,7 +1172,7 @@ The shared secret for encrypting transmission bodies between proxy router and de
 
 
 ```abnf
-relayResponse = %s"RRES" SP <encrypted(responseTransmission)>
+relayResponse = %s"RRES" SP [responseNonce] <encrypted(responseTransmission)>
 responseTransmission = fwdCorrId forwardedResponse
   ; fwdCorrId and forwardedResponse defined above in RFWD section
 ```

protocol/xrcp.md

@@ -71,7 +71,7 @@ The session invitation contains this data:
 
 Host application decrypts (except the first session) and validates the invitation:
 - Session signature is valid.
-- Timestamp is within some window from the current time.
+- Timestamp of a multicast announcement is not earlier than 3660 seconds before and not later than 3600 seconds after the current time of the host. The host ignores announcements outside of this interval and continues listening.
 - Long-term key signature is valid.
 - Long-term CA and signature key are the same as in the first session.
 - Some version in the offered range is supported.
@@ -121,9 +121,15 @@ During TLS handshake, parties validate certificate chains against previously kno
 
 ### Session verification and protocol negotiation
 
-Once TLS session is established, both the host and controller devices present a "session security code" to the user who must match them (e.g., visually or via QR code scan) and confirm on the host device. The session security code must be a digest of tlsunique channel binding. As it is computed as a digest of the TLS handshake for both the controller and the host, it will validate that the same TLS certificates are used on both sides, and that the same TLS session is established, mitigating the possibility of MITM attack in the connection.
+Once TLS session is established, both the host and controller devices present a "session security code" to the user. The session security code must be a digest of tlsunique channel binding. As it is computed as a digest of the TLS handshake for both the controller and the host, it will validate that the same TLS certificates are used on both sides, and that the same TLS session is established, mitigating the possibility of MITM attack in the connection.
 
-Once the session is confirmed by the user, the host sends HELLO block to the controller.
+To verify the session, the user matches the codes (e.g., visually or via QR code scan) and confirms the session on the host device.
+
+The user MUST verify the first session with a new controller. The host stores the long-term identity of the controller (CA fingerprint and Ed25519 public key) only after this session is verified. A controller with the stored long-term identity is a known controller.
+
+The user MAY verify the subsequent sessions with a known controller. In a session that the host confirms without user action, the controller is authenticated only by its long-term identity stored in the first session (see [Threat model](#threat-model)). SimpleX Chat mobile apps require user verification of sessions with known controllers only when the user enables the "Verify connections" option.
+
+Once the session is confirmed, the host sends HELLO block to the controller.
 
 XRCP blocks inside TLS are padded to 16384 bytes.
 
@@ -305,7 +311,7 @@ to receive: (rcvKey', sk, nonce) = HKDF(rcvKey, "SimpleXSbChain", 88)
 - prevent host and controller devices from establishing the session
 
 *cannot:*
-- same as passive adversary, provided that user visually verified session code out-of-band.
+- same as passive adversary, provided that user visually verified session code out-of-band in the first session with the controller.
 
 #### An active adversary with the access to the network:
 
@@ -324,6 +330,15 @@ to receive: (rcvKey', sk, nonce) = HKDF(rcvKey, "SimpleXSbChain", 88)
 *cannot:*
 - connect to the host or make host connect to itself.
 
+#### An active adversary with the access to the network who also obtained the long-term private keys of a known controller (e.g., from the controller device):
+
+*can:*
+- make host connect to itself instead of the controller.
+- access any data of the controlled host application, within the capabilities of the provided API, if the host confirms the session without user action.
+
+*cannot:*
+- access host application data, provided that user visually verified session code out-of-band.
+
 #### Compromised controller device:
 
 *can:*

scripts/ci/prune-old-builds.sh

@@ -0,0 +1,9 @@
+#!/bin/sh
+set -eu
+
+version=$(awk '/^version:/ {print $2}' simplexmq.cabal)
+for dir in "$@"; do
+  if [ -d "$dir/build" ]; then
+    find "$dir/build" -mindepth 3 -maxdepth 3 -type d -name 'simplexmq-*' ! -name "simplexmq-$version" -exec rm -rf {} +
+  fi
+done

scripts/ci/source-timestamps.sh

@@ -0,0 +1,27 @@
+#!/bin/sh
+set -eu
+
+command=$1
+hashes=$2
+
+sources() {
+  git ls-files -z '*.hs' '*.hsc'
+}
+
+case "$command" in
+  record)
+    sources | xargs -0 sha256sum > "$hashes"
+    ;;
+  restore)
+    sources | xargs -0 touch
+    while read -r hash file; do
+      if [ -f "$file" ] && [ "$(sha256sum "$file" | cut -d' ' -f1)" = "$hash" ]; then
+        touch -d 2000-01-01 "$file"
+      fi
+    done < "$hashes"
+    ;;
+  *)
+    echo "usage: $0 record|restore <hashes file>" >&2
+    exit 1
+    ;;
+esac

scripts/resolver/.env.example

@@ -1,5 +1,8 @@
+# Copy to .env (next to docker-compose.yml) only to change a setting; without it
+# docker-compose.yml uses the values shown here.
+
 # ============================================================================
-# Required settings — the stack will not start without these.
+# Chain settings.
 # ============================================================================
 
 # Ethereum network: mainnet (the SNRC `.testing` contracts live on mainnet)
@@ -14,9 +17,16 @@ NETWORK=mainnet
 TRUSTED_NODE_URL=https://mainnet-checkpoint-sync.attestant.io
 
 # ============================================================================
-# Optional overrides — sensible defaults are baked into docker-compose.yml,
-# so leave these commented unless you need to change them.
+# Optional overrides — leave commented unless you need to change them.
 # ============================================================================
 
 # Nimbus NAT (default: any). For a stable public node set an explicit IP:
 #   NAT=extip:1.2.3.4        # your public IPv4: curl -s ifconfig.me
+
+# Resolver (defaults in the README's "Load and scaling" and "Logs" sections)
+#   SNRC_WORKERS=4                          # worker processes; default CPU count, at most 4
+#   SNRC_RPC_TIMEOUT=5                      # seconds per request to reth
+#   SNRC_LOG_FORMAT=json                    # default text (key=value)
+#   SNRC_LOG_COLOR=always                   # coloured `docker compose logs`
+#   SNRC_LOG_LEVEL=debug                    # also log health checks and idle connections
+#   SNRC_TRUSTED_PROXIES=172.16.0.0/12      # behind a reverse proxy on the host; see README

scripts/resolver/README.md

@@ -19,15 +19,15 @@ against **Ethereum mainnet** (where the `.testing` contracts live):
 
 ## 1. Configure
 
-Edit `.env` — the defaults work as-is; override only if needed:
+The defaults in `docker-compose.yml` work as-is:
 
 ```sh
 NETWORK=mainnet                                               # default
 TRUSTED_NODE_URL=https://mainnet-checkpoint-sync.attestant.io # default
 ```
 
-Everything else (NAT) has a working default baked into `docker-compose.yml`;
-uncomment the hints in `.env` only to override.
+To override these or NAT, copy `.env.example` to `.env` and edit it; `.env` is
+not committed.
 
 ## 2. Run
 
@@ -121,6 +121,15 @@ standalone for local dev (no Docker), via [`uv`](https://docs.astral.sh/uv/):
 uv run scripts/resolver/service/snrc-resolve.py  # defaults to local reth + mainnet .testing
 ```
 
+Tests and lint run from `scripts/resolver/service`:
+
+```sh
+uv run pytest
+uv run ruff check .
+```
+
+Dependencies are locked in `uv.lock`; the Docker image installs exactly those.
+
 Three routes, versioned separately from the protocol so each only changes when
 its own shape does:
 
@@ -243,7 +252,7 @@ wrong, `error` and `status` hold the same value, so one field is enough to read.
 
 `upstreamError` says only which exception type the RPC call raised. The text
 goes to the resolver's log instead, because `SNRC_RPC` can carry a provider key
-and urlopen puts the URL it failed on into the message. It is also the answer
+and the exception can carry the URL it failed on. It is also the answer
 when a registrar, controller or oracle address has no contract behind it: the
 empty reply is refused rather than read as zero, which would make every name
 look free.
@@ -374,3 +383,87 @@ here.
 To override any of them, set `SNRC_REGISTRY_<TLD>`, `SNRC_REGISTRAR_<TLD>` or
 `SNRC_CONTROLLER_<TLD>` on the `resolver` service in `docker-compose.yml`, or
 as env vars when you run the script directly.
+
+### Load and scaling
+
+A lookup reads the chain in at most three rounds: the name's status and
+registry entries, then its record from its resolver, then prices for a name
+that is free. The node runs the calls of a JSON-RPC batch one after another,
+so the contract reads of a round go to the chain as one `eth_call` to
+[Multicall3](https://github.com/mds1/multicall). A node that is slow for a
+moment therefore delays a lookup a few times, not once per read. Connections to
+the node are kept open and reused.
+
+The resolver answers HTTP/1.1, so each smp-server keeps its connections to it
+open instead of connecting for every lookup. An idle connection is closed after
+60 s; the smp-server drops its own idle ones after 30 s, so it closes them first.
+
+Each access log line ends with how long the request took, so slow requests
+show up in `docker compose logs resolver`.
+
+| Variable | Default | Meaning |
+|---|---|---|
+| `SNRC_WORKERS` | CPU count, at most 4 | processes sharing the port. If one exits, the others stop, so the container restarts |
+| `SNRC_RPC_TIMEOUT` | `5` | seconds to wait for each request to the node; the smp-server gives up after 3 |
+| `SNRC_MULTICALL` | `0xcA11bde05977b3631167028862bE2a173976CA11` | Multicall3 address. If it does not answer, each round is sent as a plain batch and the log says so once |
+
+### Logs
+
+Each event is one line: UTC time, level, event name, then `key=value` fields.
+
+```
+2026-09-26T08:06:19.992Z INFO  request  client=203.0.113.7 worker=12 method=GET path=/v2/resolve/[4fdd…].testing status=200 bytes=701 ms=35
+2026-09-26T08:06:20.051Z WARN  upstream_error  name=foobar.testing error=ConnectionRefusedError message="[Errno 111] Connection refused"
+```
+
+| Event | Level | Meaning |
+|---|---|---|
+| `listening` | info | started: address, workers, RPC endpoint, registries, trusted proxies |
+| `request` | info | one answered request; `/health` only at debug |
+| `upstream_error` | warn | a read from the node failed; the caller got 502 |
+| `multicall_unavailable` | warn | Multicall3 did not answer; rounds are sent as plain batches |
+| `client_gone` | warn | the caller hung up before the answer, usually an smp-server past its timeout |
+| `http_error` | warn | a malformed request; idle keep-alive timeouts only at debug |
+| `request_failed` | error | a request failed unexpectedly, with its traceback |
+| `worker_exited`, `worker_failed` | error | a worker stopped; the others stop too, so the container restarts |
+| `stopping` | info | stopped by a signal |
+
+| Variable | Default | Meaning |
+|---|---|---|
+| `SNRC_LOG_FORMAT` | `text` | `text` (`key=value`) or `json`, one object per line |
+| `SNRC_LOG_COLOR` | `auto` | `auto` colours only a terminal, `always`, `never`. Docker output is no terminal, so set `always` for coloured `docker compose logs` |
+| `SNRC_LOG_LEVEL` | `info` | `debug` also logs health checks and idle connections closing |
+| `SNRC_TRUSTED_PROXIES` | none | addresses or CIDRs of reverse proxies whose `X-Forwarded-For` names the client |
+
+With docker compose, set these, and those in "Load and scaling", in `.env`
+(`.env.example` lists them).
+
+Behind a reverse proxy such as Caddy on the host, the resolver sees the Docker
+network's gateway rather than the client. Trust that gateway, and the logged
+client is the last `X-Forwarded-For` address no trusted proxy added, so a caller
+through the proxy cannot claim another address. Everything else on the host that
+connects to `127.0.0.1:8000`, the smp-servers included, arrives through the same
+gateway and could set the logged address too; run the proxy in the compose network
+and trust only its address to avoid that.
+
+The gateway is
+`docker network inspect resolver_default --format '{{(index .IPAM.Config 0).Gateway}}'`,
+with the network named after the compose project. It changes when the network is
+recreated, as `docker compose down` does, so trust one of, in `.env`:
+
+- `SNRC_TRUSTED_PROXIES=172.16.0.0/12`: Docker's default pools for compose
+  networks, whatever subnet it picks. This also trusts every other container on the
+  host. Add the gateway too if Docker ever gives the network a `192.168.x.x` one.
+- a pinned subnet, so the gateway stays fixed, and only that gateway:
+
+  ```yaml
+  networks:
+    default:
+      ipam:
+        config:
+          - subnet: 172.30.0.0/24
+            gateway: 172.30.0.1
+  ```
+
+  in `docker-compose.yml`, with `SNRC_TRUSTED_PROXIES=172.30.0.1`, on a subnet no
+  other network on the host uses.

scripts/resolver/docker-compose.yml

@@ -38,14 +38,14 @@ services:
       - sh
       - -c
       - |
-        if [ -d /home/user/nimbus-eth2/build/data/${NETWORK}/db ]; then
+        if [ -d /home/user/nimbus-eth2/build/data/${NETWORK:-mainnet}/db ]; then
           echo "Nimbus data dir already initialised — skipping checkpoint sync";
           exit 0;
         fi;
         /home/user/nimbus-eth2/build/nimbus_beacon_node trustedNodeSync \
-          --network=${NETWORK} \
-          --data-dir=/home/user/nimbus-eth2/build/data/${NETWORK} \
-          --trusted-node-url=${TRUSTED_NODE_URL} \
+          --network=${NETWORK:-mainnet} \
+          --data-dir=/home/user/nimbus-eth2/build/data/${NETWORK:-mainnet} \
+          --trusted-node-url=${TRUSTED_NODE_URL:-https://mainnet-checkpoint-sync.attestant.io} \
           --backfill=false
     restart: "no"
 
@@ -69,8 +69,8 @@ services:
           echo "Reth data already initialised — skipping snapshot download";
           exit 0;
         fi;
-        echo "Downloading Reth ${NETWORK} --minimal snapshot...";
-        reth download --datadir /data --chain ${NETWORK} --minimal && \
+        echo "Downloading Reth ${NETWORK:-mainnet} --minimal snapshot...";
+        reth download --datadir /data --chain ${NETWORK:-mainnet} --minimal && \
         touch /data/.snapshot-done && \
         echo "Snapshot download complete"
     restart: "no"
@@ -92,7 +92,7 @@ services:
     command: >
       node
       --datadir /data
-      --chain ${NETWORK}
+      --chain ${NETWORK:-mainnet}
       --minimal
       --authrpc.jwtsecret /jwt/jwt.hex
       --authrpc.addr 0.0.0.0 --authrpc.port 8551
@@ -117,8 +117,8 @@ services:
       - "9000:9000/udp"
       - "127.0.0.1:5052:5052"
     command: >
-      --network=${NETWORK}
-      --data-dir=/home/user/nimbus-eth2/build/data/${NETWORK}
+      --network=${NETWORK:-mainnet}
+      --data-dir=/home/user/nimbus-eth2/build/data/${NETWORK:-mainnet}
       --el=http://reth:8551
       --jwt-secret=/jwt/jwt.hex
       --non-interactive
@@ -131,6 +131,7 @@ services:
   # smp-server points its [NAMES] resolver_endpoint at this URL.
   # To change the host port, edit the LEFT side of the port mapping below.
   resolver:
+    image: snrc-resolve:local
     build:
       context: ./service
       dockerfile: Dockerfile
@@ -156,8 +157,18 @@ services:
       # SNRC_REGISTRAR_SIMPLEX: 0x...
       # SNRC_CONTROLLER_TESTING: 0x...
       # SNRC_CONTROLLER_SIMPLEX: 0x...
+      # Set in .env (see .env.example); unset or empty means the resolver's default.
+      SNRC_WORKERS: ${SNRC_WORKERS:-}
+      SNRC_RPC_TIMEOUT: ${SNRC_RPC_TIMEOUT:-}
+      SNRC_LOG_FORMAT: ${SNRC_LOG_FORMAT:-}
+      SNRC_LOG_COLOR: ${SNRC_LOG_COLOR:-}
+      SNRC_LOG_LEVEL: ${SNRC_LOG_LEVEL:-}
+      SNRC_TRUSTED_PROXIES: ${SNRC_TRUSTED_PROXIES:-}
     ports:
       - "127.0.0.1:8000:8000"
+    # a line per lookup; the local driver keeps 5 rotated files of 20 MB
+    logging:
+      driver: local
     restart: unless-stopped
 
 volumes:

scripts/resolver/service/.dockerignore

@@ -0,0 +1,5 @@
+.venv
+**/__pycache__
+.pytest_cache
+.ruff_cache
+test_snrc_resolve.py

scripts/resolver/service/.python-version

@@ -0,0 +1 @@
+3.13

scripts/resolver/service/Dockerfile

@@ -1,9 +1,10 @@
 # syntax=docker/dockerfile:1.7
 # ---------- builder ----------
-# Use the official uv image (Astral) on top of a slim Python base.
-# uv resolves and installs the lockfile-free pyproject.toml in seconds and
-# produces a portable .venv we can copy into the runtime stage.
-FROM ghcr.io/astral-sh/uv:python3.13-bookworm-slim AS builder
+# The official uv image (Astral) on top of a slim Python base installs the
+# locked dependencies into a portable .venv for the runtime stage.
+# Pinned to the uv that wrote uv.lock (its bookworm images stopped at 0.9.30), on the
+# runtime's Debian release, so the .venv's Python path exists there.
+FROM ghcr.io/astral-sh/uv:0.12.16-python3.13-trixie-slim AS builder
 
 ENV UV_LINK_MODE=copy \
     UV_COMPILE_BYTECODE=1 \
@@ -12,17 +13,17 @@ ENV UV_LINK_MODE=copy \
 
 WORKDIR /app
 
-# Install deps first (separate layer) — script edits won't bust this cache.
-COPY pyproject.toml ./
+# Dependencies first (separate layer), exactly as locked: script edits don't bust this cache.
+COPY pyproject.toml uv.lock .python-version ./
 RUN --mount=type=cache,target=/root/.cache/uv \
-    uv sync --no-dev --no-install-project
+    uv sync --frozen --no-dev --no-install-project
 
 # Script is added after the dep layer for cache friendliness.
 COPY snrc-resolve.py ./
 
 # ---------- runtime ----------
 # Slim runtime — only the venv + script. No uv, no apt.
-FROM python:3.13-slim AS runtime
+FROM python:3.13-slim-trixie AS runtime
 
 ENV PYTHONUNBUFFERED=1 \
     PYTHONDONTWRITEBYTECODE=1 \
@@ -33,15 +34,15 @@ RUN groupadd --system --gid 10001 snrc && \
     useradd  --system --uid 10001 --gid snrc --no-create-home --shell /usr/sbin/nologin snrc
 
 WORKDIR /app
-COPY --from=builder --chown=snrc:snrc /app /app
+# owned by root, so the service cannot rewrite its own code
+COPY --from=builder /app /app
 
 USER snrc:snrc
 
 EXPOSE 8000
 
-# Liveness check hits the script's own /health route. ThreadingHTTPServer is
-# fast enough that 3s is generous for a localhost probe; restart if it stops
-# responding entirely.
+# Liveness check hits the script's own /health route. Three failures mark the
+# container unhealthy; compose does not restart it for that, but `docker ps` shows it.
 HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
     CMD ["python", "-c", "import urllib.request, sys; sys.exit(0 if urllib.request.urlopen('http://127.0.0.1:8000/health', timeout=3).status == 200 else 1)"]
 

scripts/resolver/service/pyproject.toml

@@ -9,5 +9,14 @@ dependencies = [
   "eth-hash[pycryptodome]>=0.7",
 ]
 
+[dependency-groups]
+dev = [
+  "pytest>=8",
+  "ruff>=0.6",
+]
+
 [tool.uv]
 package = false
+
+[tool.ruff.lint]
+select = ["E9", "F", "B"]

scripts/resolver/service/snrc-resolve.py

@@ -46,6 +46,15 @@ Environment:
                          (default: mainnet for .testing, empty for .simplex)
   SNRC_PORT              Listen port (default: 8000)
   SNRC_BIND              Bind address (default: 0.0.0.0)
+  SNRC_WORKERS           Worker processes sharing the port (default: CPU count, at most 4)
+  SNRC_RPC_TIMEOUT       Seconds to wait for each RPC request (default: 5)
+  SNRC_MULTICALL         Multicall3 contract that runs a round of reads as one call
+                         (default: 0xcA11bde05977b3631167028862bE2a173976CA11)
+  SNRC_LOG_FORMAT        text (key=value) or json (default: text)
+  SNRC_LOG_COLOR         auto (on a terminal), always or never (default: auto)
+  SNRC_LOG_LEVEL         debug, info, warning or error (default: info)
+  SNRC_TRUSTED_PROXIES   Comma-separated addresses or CIDRs of reverse proxies whose
+                         X-Forwarded-For names the client (default: none)
 
 Each TLD is a separate SNRC deployment with its own ENSRegistry; the
 resolver dispatches by the queried name's rightmost label.
@@ -64,19 +73,45 @@ Unrecognised payloads fall back to `0x`-prefixed raw hex.
 """
 
 import hashlib
+import ipaddress
 import json
+import logging
 import os
+import queue
+import signal
+import socket
 import sys
+import threading
 import time
+from contextlib import contextmanager
+from datetime import datetime, timezone
+from functools import lru_cache
+from http.client import BadStatusLine, HTTPConnection, HTTPSConnection
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from urllib.error import HTTPError
 from urllib.parse import unquote, urlparse
-from urllib.request import Request, urlopen
 
 from eth_hash.auto import keccak
 
 RPC = os.environ.get("SNRC_RPC", "http://127.0.0.1:8545")
 BIND = os.environ.get("SNRC_BIND", "0.0.0.0")
 PORT = int(os.environ.get("SNRC_PORT", "8000"))
+# The smp-server gives up after 3 s, so a slower call only holds a thread.
+RPC_TIMEOUT_S = float(os.environ.get("SNRC_RPC_TIMEOUT", "") or 5)
+# The node and the beacon client usually share the host, so not every core.
+MAX_DEFAULT_WORKERS = 4
+WORKERS = int(os.environ.get("SNRC_WORKERS", "") or min(MAX_DEFAULT_WORKERS, os.cpu_count() or 1))
+# Multicall3, at this address on mainnet and most chains. A node runs a JSON-RPC
+# batch one call after another, so a round of reads is sent as one eth_call.
+MULTICALL = os.environ.get("SNRC_MULTICALL", "") or "0xcA11bde05977b3631167028862bE2a173976CA11"
+LOG_FORMAT = os.environ.get("SNRC_LOG_FORMAT", "") or "text"
+LOG_COLOR = os.environ.get("SNRC_LOG_COLOR", "") or "auto"
+LOG_LEVEL = os.environ.get("SNRC_LOG_LEVEL", "") or "info"
+# Peers whose X-Forwarded-For is believed, such as the Docker gateway a reverse proxy on the host
+# connects through. Anyone else could put any address there.
+TRUSTED_PROXIES = tuple(
+    ipaddress.ip_network(p.strip(), strict=False) for p in os.environ.get("SNRC_TRUSTED_PROXIES", "").split(",") if p.strip()
+)
 
 # Each TLD is its own SNRC deployment with its own ENSRegistry. Dispatch
 # happens on the rightmost label of the queried name. Empty / unset means
@@ -121,34 +156,291 @@ COIN_ETH = 60
 COIN_BTC = 0
 COIN_XMR = 128
 COIN_DOT = 354
+RECORD_COINS = (COIN_ETH, COIN_BTC, COIN_XMR, COIN_DOT)
 
 ZERO_ADDR = "0x0000000000000000000000000000000000000000"
 
 # The registry prices in attoUSD (1e-18 USD); the protocol carries US cents.
 
 
+# ---------- Logging ----------
+
+LOGGER = logging.getLogger("snrc_resolve")
+LEVEL_NAMES = {logging.DEBUG: "DEBUG", logging.INFO: "INFO", logging.WARNING: "WARN", logging.ERROR: "ERROR"}
+# ANSI SGR codes
+DIM, BOLD, GREEN, YELLOW, RED = "2", "1", "32", "33", "31"
+LEVEL_COLORS = {"DEBUG": DIM, "INFO": GREEN, "WARN": YELLOW, "ERROR": RED}
+STATUS_COLORS = {2: GREEN, 3: GREEN, 4: YELLOW, 5: RED}
+
+
+def log_event(level: int, name: str, /, exc_info=None, **fields):
+    LOGGER.log(level, name, exc_info=exc_info, extra={"fields": fields})
+
+
+def _log_time(record: logging.LogRecord) -> str:
+    return datetime.fromtimestamp(record.created, timezone.utc).isoformat(timespec="milliseconds").replace("+00:00", "Z")
+
+
+def _log_value(value) -> str:
+    """A logfmt value: bare when it cannot be misread, JSON-quoted otherwise."""
+    if value is None:
+        return "-"
+    s = str(value)
+    if not s or any(c in s for c in ' ="\\') or not s.isprintable():
+        return json.dumps(s)
+    return s
+
+
+class TextFormatter(logging.Formatter):
+    def __init__(self, color: bool):
+        super().__init__()
+        self.color = color
+
+    def _paint(self, code: str, text: str) -> str:
+        return f"\033[{code}m{text}\033[0m" if self.color and code else text
+
+    def format(self, record: logging.LogRecord) -> str:
+        level = LEVEL_NAMES.get(record.levelno, record.levelname)
+        fields = " ".join(
+            self._paint(DIM, f"{k}=") + self._paint(STATUS_COLORS.get(v // 100, "") if k == "status" and isinstance(v, int) else "", _log_value(v))
+            for k, v in getattr(record, "fields", {}).items()
+        )
+        line = f"{self._paint(DIM, _log_time(record))} {self._paint(LEVEL_COLORS.get(level, ''), f'{level:<5}')} {self._paint(BOLD, record.getMessage())}"
+        if fields:
+            line += "  " + fields
+        if record.exc_info:
+            line += "\n" + self.formatException(record.exc_info)
+        return line
+
+
+class JsonFormatter(logging.Formatter):
+    def format(self, record: logging.LogRecord) -> str:
+        out = {
+            "time": _log_time(record),
+            "level": LEVEL_NAMES.get(record.levelno, record.levelname).lower(),
+            "event": record.getMessage(),
+            **getattr(record, "fields", {}),
+        }
+        if record.exc_info:
+            out["exception"] = self.formatException(record.exc_info)
+        return json.dumps(out, default=str)
+
+
+def setup_logging(stream=None):
+    """Sends the resolver's events to stderr in the configured format."""
+    stream = stream or sys.stderr
+    if LOG_FORMAT not in ("text", "json"):
+        raise ValueError(f"SNRC_LOG_FORMAT must be text or json, not {LOG_FORMAT!r}")
+    if LOG_COLOR not in ("auto", "always", "never"):
+        raise ValueError(f"SNRC_LOG_COLOR must be auto, always or never, not {LOG_COLOR!r}")
+    level = logging.getLevelName(LOG_LEVEL.upper())
+    if not isinstance(level, int):
+        raise ValueError(f"SNRC_LOG_LEVEL must be debug, info, warning or error, not {LOG_LEVEL!r}")
+    color = LOG_COLOR == "always" or (LOG_COLOR == "auto" and stream.isatty())
+    handler = logging.StreamHandler(stream)
+    handler.setFormatter(JsonFormatter() if LOG_FORMAT == "json" else TextFormatter(color))
+    LOGGER.handlers[:] = [handler]
+    LOGGER.setLevel(level)
+    LOGGER.propagate = False
+
+
 # ---------- RPC + ABI helpers (mirrors ens-lookup.py shape) ----------
 
+RPC_URL = urlparse(RPC)
+# Set a non-default User-Agent; Cloudflare-fronted public RPCs (drpc,
+# publicnode, etc.) reject `Python-urllib/3.x` with 403.
+RPC_HEADERS = {"Content-Type": "application/json", "User-Agent": "snrc-resolve/1.0"}
+
+# Idle keep-alive connections to SNRC_RPC. A connection per call costs most of
+# a lookup's CPU and leaves a TIME_WAIT socket per call, which exhausts local
+# ports at a few dozen lookups per second.
+# Idle connections kept per worker; more are closed after use. 4 workers stay well
+# under reth's default limit of 500 connections.
+RPC_POOL_SIZE = 32
+_rpc_pool = queue.LifoQueue(maxsize=RPC_POOL_SIZE)
+
+
+def _new_rpc_connection():
+    conn_class = HTTPSConnection if RPC_URL.scheme == "https" else HTTPConnection
+    return conn_class(RPC_URL.hostname, RPC_URL.port, timeout=RPC_TIMEOUT_S)
+
+
+def _post_rpc(conn, body: bytes) -> bytes:
+    path = (RPC_URL.path or "/") + (f"?{RPC_URL.query}" if RPC_URL.query else "")
+    conn.request("POST", path, body, RPC_HEADERS)
+    res = conn.getresponse()
+    data = res.read()
+    # HTTPError, as urlopen raised: RuntimeError means the call itself failed
+    if not 200 <= res.status < 300:
+        raise HTTPError(RPC, res.status, res.reason, res.headers, None)
+    return data
+
+
+def _post_pooled(conn, body: bytes) -> bytes:
+    try:
+        data = _post_rpc(conn, body)
+    except BaseException:
+        conn.close()
+        raise
+    try:
+        _rpc_pool.put_nowait(conn)
+    except queue.Full:
+        conn.close()
+    return data
+
+
+def _send_rpc(payload) -> object:
+    body = json.dumps(payload).encode()
+    try:
+        idle = _rpc_pool.get_nowait()
+    except queue.Empty:
+        return json.loads(_post_pooled(_new_rpc_connection(), body))
+    try:
+        return json.loads(_post_pooled(idle, body))
+    except (ConnectionError, BadStatusLine):
+        # the node closed the idle connection; every call is a read, so resending is safe
+        return json.loads(_post_pooled(_new_rpc_connection(), body))
+
+
+# Reads prefetched for the request being answered, keyed by _read_key. Only set
+# inside request_reads(), so code called outside a request reads one call at a time.
+_request = threading.local()
+_UNREAD = object()
+
+
+def _read_key(method, params) -> str:
+    return method + json.dumps(params, sort_keys=True)
+
+
+@contextmanager
+def request_reads():
+    _request.reads = {}
+    try:
+        yield
+    finally:
+        del _request.reads
+
+
 def rpc(method, params):
-    body = json.dumps(
-        {"jsonrpc": "2.0", "method": method, "params": params, "id": 1}
-    ).encode()
-    # Set a non-default User-Agent; Cloudflare-fronted public RPCs (drpc,
-    # publicnode, etc.) reject `Python-urllib/3.x` with 403.
-    req = Request(
-        RPC,
-        data=body,
-        headers={
-            "Content-Type": "application/json",
-            "User-Agent": "snrc-resolve/1.0",
-        },
-    )
-    res = json.loads(urlopen(req, timeout=15).read())
+    reads = getattr(_request, "reads", None)
+    if reads is not None:
+        read = reads.get(_read_key(method, params), _UNREAD)
+        if isinstance(read, RuntimeError):
+            raise read
+        if read is not _UNREAD:
+            return read
+    res = _send_rpc({"jsonrpc": "2.0", "method": method, "params": params, "id": 1})
     if "error" in res:
         raise RuntimeError(res["error"])
     return res["result"]
 
 
+def _send_batch(requests):
+    """Answers of a JSON-RPC batch in request order, None for a request left
+    unanswered; all None when the node does not batch."""
+    res = _send_rpc([{"jsonrpc": "2.0", "id": i, "method": m, "params": p} for i, (m, p) in enumerate(requests)])
+    # a node that does not batch answers with a single error object
+    by_id = {r.get("id"): r for r in res if isinstance(r, dict)} if isinstance(res, list) else {}
+    return [by_id.get(i) for i in range(len(requests))]
+
+
+def _is_latest_call(method, params) -> bool:
+    return method == "eth_call" and params[1] == "latest"
+
+
+_multicall_failed_logged = False
+
+
+def _remember(reads, requests, answers):
+    for (m, p), r in zip(requests, answers, strict=True):
+        if r is not None and ("result" in r or "error" in r):
+            reads[_read_key(m, p)] = RuntimeError(r["error"]) if "error" in r else r["result"]
+
+
+def prefetch(requests):
+    """Sends reads the request will make in one round trip, with its contract
+    reads as one multicall. A read left unanswered is made on its own when the
+    code gets to it."""
+    global _multicall_failed_logged
+    reads = getattr(_request, "reads", None)
+    if reads is None:
+        return
+    todo = [(m, p) for m, p in requests if _read_key(m, p) not in reads]
+    if len(todo) < 2:
+        return
+    calls = [(m, p) for m, p in todo if _is_latest_call(m, p)]
+    if len(calls) < 2:
+        _remember(reads, todo, _send_batch(todo))
+        return
+    others = [(m, p) for m, p in todo if not _is_latest_call(m, p)]
+    multicall = eth_call_read(MULTICALL, encode_aggregate3([(p[0]["to"], p[0]["data"]) for _, p in calls]))
+    answers = _send_batch(others + [multicall])
+    if all(r is None for r in answers):
+        return
+    _remember(reads, others, answers[:-1])
+    try:
+        result = (answers[-1] or {}).get("result")
+        if not isinstance(result, str):
+            raise ValueError(f"multicall answered {answers[-1]!r}")
+        results = decode_aggregate3(result)
+        if len(results) != len(calls):
+            raise ValueError("multicall answered a different number of calls")
+    except (KeyError, TypeError, ValueError) as e:
+        if not _multicall_failed_logged:
+            _multicall_failed_logged = True
+            log_event(logging.WARNING, "multicall_unavailable", multicall=MULTICALL, error=repr(e), fallback="batch")
+        _remember(reads, calls, _send_batch(calls))
+        return
+    for (m, p), (success, data) in zip(calls, results, strict=True):
+        reads[_read_key(m, p)] = "0x" + data.hex() if success else RuntimeError("execution reverted")
+
+
+AGGREGATE3 = "0x82ad56cb"  # aggregate3((address,bool,bytes)[])
+
+
+def encode_aggregate3(calls) -> str:
+    """Calldata for Multicall3.aggregate3 with every call allowed to fail."""
+    tuples = []
+    for to, data in calls:
+        b = bytes.fromhex(data[2:])
+        tuples.append(
+            int(to, 16).to_bytes(32, "big")
+            + (1).to_bytes(32, "big")
+            + (0x60).to_bytes(32, "big")
+            + len(b).to_bytes(32, "big")
+            + b
+            + b"\x00" * ((-len(b)) % 32)
+        )
+    offsets, at = [], 32 * len(tuples)
+    for t in tuples:
+        offsets.append(at.to_bytes(32, "big"))
+        at += len(t)
+    body = (0x20).to_bytes(32, "big") + len(tuples).to_bytes(32, "big") + b"".join(offsets) + b"".join(tuples)
+    return AGGREGATE3 + body.hex()
+
+
+def decode_aggregate3(hex_data: str):
+    """Multicall3.aggregate3's (bool success, bytes returnData)[]."""
+    raw = bytes.fromhex(hex_data[2:] if hex_data.startswith("0x") else hex_data)
+
+    def word(at: int) -> int:
+        if at + 32 > len(raw):
+            raise ValueError("multicall answer is truncated")
+        return int.from_bytes(raw[at:at + 32], "big")
+
+    array = word(0)
+    base = array + 32
+    out = []
+    for i in range(word(array)):
+        item = base + word(base + 32 * i)
+        data = item + word(item + 32)
+        length = word(data)
+        if data + 32 + length > len(raw):
+            raise ValueError("multicall answer is truncated")
+        out.append((word(item) != 0, raw[data + 32:data + 32 + length]))
+    return out
+
+
 def namehash(name: str) -> bytes:
     node = b"\x00" * 32
     if name:
@@ -184,11 +476,14 @@ def node_of(name: str) -> bytes:
 # ---------- Registration status ----------
 
 
+BLOCK_READ = ("eth_getBlockByNumber", ["latest", False])
+
+
 def head_block():
     """How far behind the node is. Unlike expiry, this is the one thing that has
     to be measured against the host clock: a node that stops still has a block."""
     try:
-        block = rpc("eth_getBlockByNumber", ["latest", False])
+        block = rpc(*BLOCK_READ)
         return {
             "blockNumber": decode_uint(block["number"]),
             "chainLagSeconds": int(time.time()) - decode_uint(block["timestamp"]),
@@ -199,13 +494,13 @@ def head_block():
 
 def chain_now() -> int:
     """Expiry is compared against the block timestamp, never the host clock."""
-    block = rpc("eth_getBlockByNumber", ["latest", False])
+    block = rpc(*BLOCK_READ)
     return decode_uint(block["timestamp"])
 
 
 def grace_period(registrar: str) -> int:
     """A deployment can configure a different window, so it is read on chain."""
-    return decode_uint(eth_call(registrar, selector("GRACE_PERIOD()")))
+    return decode_uint(eth_call(*grace_call(registrar)))
 
 
 def expiry_status(expires: int, grace: int, now: int) -> str:
@@ -225,8 +520,7 @@ def reservation_reason(tld: str, token: int) -> int:
     controller = CONTROLLERS.get(tld)
     if not controller:
         return 0
-    raw = eth_call(controller, selector("reservedNames(bytes32)") + encode_uint(token))
-    return decode_uint(raw)
+    return decode_uint(eth_call(*reserved_call(controller, token)))
 
 
 def pricing_params(tld: str):
@@ -235,7 +529,8 @@ def pricing_params(tld: str):
     controller = CONTROLLERS.get(tld)
     if not controller:
         return None
-    oracle = decode_address(eth_call(controller, selector("prices()")))
+    prefetch([eth_call_read(*prices_call(controller)), eth_call_read(*min_length_call(controller))])
+    oracle = decode_address(eth_call(*prices_call(controller)))
     if oracle == ZERO_ADDR:
         return None
     try:
@@ -247,6 +542,8 @@ def pricing_params(tld: str):
 
 
 SECONDS_PER_YEAR = 31536000
+# an ENS-shaped oracle prices names by length up to six letters
+LETTER_TIERS = range(1, 7)
 ATTO_PER_CENT = 10**16
 
 
@@ -255,13 +552,15 @@ def read_oracle_prices(controller: str, oracle: str):
     protocol carries. An ENS-shaped oracle prices in attoUSD per second and
     charges a premium on lapsed names that it does not expose, so a quote from
     it is only safe for a name that was never registered."""
+    # either oracle shape is answered in the same round trip
+    prefetch([eth_call_read(*prices_call(oracle))] + [eth_call_read(*letter_price_call(oracle, n)) for n in LETTER_TIERS])
     try:
-        base, tiers = decode_prices(eth_call(oracle, selector("prices()")))
+        base, tiers = decode_prices(eth_call(*prices_call(oracle)))
         premium_unknown = False
     except RuntimeError:
         base, tiers = decode_letter_prices(oracle)
         premium_unknown = True
-    min_len = decode_uint(eth_call(controller, selector("minCharLength()")))
+    min_len = decode_uint(eth_call(*min_length_call(controller)))
     return {
         # lengths the registry refuses are left out rather than priced at zero
         "registrationPrices": {n: c for n, c in tiers.items() if n >= min_len},
@@ -277,9 +576,9 @@ def decode_letter_prices(oracle: str):
     the six-letter tier stops at five, and charges its highest tier for anything
     longer, which is what basePrice means here."""
     tiers = {}
-    for n in range(1, 7):
+    for n in LETTER_TIERS:
         try:
-            rate = decode_uint(eth_call(oracle, selector(f"price{n}Letter()")))
+            rate = decode_uint(eth_call(*letter_price_call(oracle, n)))
         except RuntimeError:
             if n <= 5:
                 raise
@@ -301,6 +600,9 @@ def decode_prices(hex_data: str):
     base = int.from_bytes(raw[:32], "big")
     at = int.from_bytes(raw[32:64], "big")
     count = int.from_bytes(raw[at:at + 32], "big")
+    # the count comes from the answer, so it is checked against the answer's length
+    if at + 32 + count * 64 > len(raw):
+        raise RuntimeError("prices(): short response")
     tiers = {}
     for i in range(count):
         item = at + 32 + i * 64
@@ -328,9 +630,7 @@ def name_status(name: str):
     # The 2LD's label is that key at any depth. node_of decodes a bracket only
     # in a two-label name, so a bracket subname gets a status but no record.
     token = label_token(labels[-2])
-    expires = decode_uint(
-        eth_call(registrar, selector("nameExpires(uint256)") + encode_uint(token))
-    )
+    expires = decode_uint(eth_call(*expires_call(registrar, token)))
     grace = grace_period(registrar) if expires else 0
     now = chain_now()
     status = expiry_status(expires, grace, now)
@@ -358,12 +658,17 @@ def name_status(name: str):
     return out
 
 
+@lru_cache(maxsize=None)
 def selector(signature: str) -> str:
     return "0x" + keccak(signature.encode())[:4].hex()
 
 
+def eth_call_read(to: str, data: str):
+    return "eth_call", [{"to": to, "data": data}, "latest"]
+
+
 def eth_call(to: str, data: str) -> str:
-    result = rpc("eth_call", [{"to": to, "data": data}, "latest"])
+    result = rpc(*eth_call_read(to, data))
     if result == "0x":
         raise RuntimeError(f"empty return from {to}: no contract at that address?")
     return result
@@ -385,7 +690,7 @@ def registered_label(registrar: str, token: int):
     """The plaintext label the registrar recorded at registration, keyed by the
     hash of that label. None when the name was registered without
     registerWithLabel, so the registrar cannot name it."""
-    raw = decode_bytes(eth_call(registrar, selector("labelOf(uint256)") + encode_uint(token)))
+    raw = decode_bytes(eth_call(*label_call(registrar, token)))
     return raw.decode("utf-8", errors="replace") if raw else None
 
 
@@ -427,7 +732,7 @@ def encode_text_call(node: bytes, key: str) -> str:
 
 
 def text(resolver: str, node: bytes, key: str) -> str:
-    raw = decode_bytes(eth_call(resolver, encode_text_call(node, key)))
+    raw = decode_bytes(eth_call(*text_call(resolver, node, key)))
     return raw.decode("utf-8", errors="replace") if raw else ""
 
 
@@ -446,7 +751,7 @@ def addr_multicoin(resolver: str, node: bytes, coin_type: int):
     payload doesn't match any recognised on-chain shape. Returns None when
     the record is unset."""
     try:
-        raw = decode_bytes(eth_call(resolver, encode_addr_multicoin_call(node, coin_type)))
+        raw = decode_bytes(eth_call(*addr_call(resolver, node, coin_type)))
     except RuntimeError:
         return None
     if not raw:
@@ -465,6 +770,81 @@ def addr_multicoin(resolver: str, node: bytes, coin_type: int):
         return "0x" + raw.hex()
 
 
+# ---------- Contract calls, as (to, data) ----------
+# Shared by the reads and prefetch, so a prefetched read is found by the code that makes it.
+
+
+def expires_call(registrar: str, token: int):
+    return registrar, selector("nameExpires(uint256)") + encode_uint(token)
+
+
+def grace_call(registrar: str):
+    return registrar, selector("GRACE_PERIOD()")
+
+
+def label_call(registrar: str, token: int):
+    return registrar, selector("labelOf(uint256)") + encode_uint(token)
+
+
+def reserved_call(controller: str, token: int):
+    return controller, selector("reservedNames(bytes32)") + encode_uint(token)
+
+
+def prices_call(contract: str):
+    return contract, selector("prices()")
+
+
+def letter_price_call(oracle: str, letters: int):
+    return oracle, selector(f"price{letters}Letter()")
+
+
+def min_length_call(controller: str):
+    return controller, selector("minCharLength()")
+
+
+def resolver_call(registry: str, node: bytes):
+    return registry, selector("resolver(bytes32)") + node.hex()
+
+
+def owner_call(registry: str, node: bytes):
+    return registry, selector("owner(bytes32)") + node.hex()
+
+
+def text_call(resolver: str, node: bytes, key: str):
+    return resolver, encode_text_call(node, key)
+
+
+def addr_call(resolver: str, node: bytes, coin_type: int):
+    return resolver, encode_addr_multicoin_call(node, coin_type)
+
+
+def lookup_reads(name: str):
+    """The reads a lookup makes before it knows the name's resolver."""
+    labels = name.split(".")
+    tld = labels[-1]
+    reads = []
+    registrar = REGISTRARS.get(tld)
+    if registrar and len(labels) >= 2:
+        token = label_token(labels[-2])
+        reads += [BLOCK_READ, eth_call_read(*expires_call(registrar, token)), eth_call_read(*grace_call(registrar))]
+        if len(labels) == 2 and is_encoded_labelhash(labels[0]):
+            reads.append(eth_call_read(*label_call(registrar, token)))
+        if CONTROLLERS.get(tld):
+            reads.append(eth_call_read(*reserved_call(CONTROLLERS[tld], token)))
+    registry = REGISTRIES.get(tld)
+    if registry:
+        node = node_of(name)
+        reads += [eth_call_read(*resolver_call(registry, node)), eth_call_read(*owner_call(registry, node))]
+    return reads
+
+
+def record_reads(resolver: str, node: bytes):
+    """The reads of a name's record from its resolver."""
+    return [eth_call_read(*text_call(resolver, node, k)) for k in TEXT_KEYS] + [
+        eth_call_read(*addr_call(resolver, node, coin)) for coin in RECORD_COINS
+    ]
+
+
 # ---------- Coin-specific address encoders ----------
 # Each takes raw bytes as stored under ENSIP-9 and returns the canonical
 # user-facing string for that chain (EIP-55 for ETH, bech32/base58check
@@ -665,9 +1045,9 @@ def split_links(value: str) -> list:
 
 
 def upstream_error(subject: dict, e: Exception) -> dict:
-    """urlopen puts the failing URL into its message and SNRC_RPC can carry a
-    provider key, so the text goes to the log and only the type to the caller."""
-    print(f"upstream error: {type(e).__name__}: {e}", file=sys.stderr)
+    """The exception can carry the failing URL and SNRC_RPC can carry a provider
+    key, so the text goes to the log and only the type to the caller."""
+    log_event(logging.WARNING, "upstream_error", **subject, error=type(e).__name__, message=str(e))
     return {
         **subject,
         "error": "upstreamError",
@@ -680,9 +1060,8 @@ def name_record(name: str):
     has one, with every field unset."""
     registry = REGISTRIES[name.rsplit(".", 1)[-1]]
     node = node_of(name)
-    node_hex = node.hex()
-    resolver_addr = decode_address(eth_call(registry, selector("resolver(bytes32)") + node_hex))
-    owner = decode_address(eth_call(registry, selector("owner(bytes32)") + node_hex))
+    resolver_addr = decode_address(eth_call(*resolver_call(registry, node)))
+    owner = decode_address(eth_call(*owner_call(registry, node)))
     rec = {
         "name": canonical_name(name),
         "nickname": "",
@@ -699,6 +1078,7 @@ def name_record(name: str):
     }
     if resolver_addr == ZERO_ADDR:
         return rec
+    prefetch(record_reads(resolver_addr, node))
     texts = {}
     for k in TEXT_KEYS:
         try:
@@ -734,6 +1114,7 @@ def registration(name: str):
     tld = name.rsplit(".", 1)[-1]
     if not REGISTRIES.get(tld):
         return 400, {"name": name, "error": "tldNotConfigured"}
+    prefetch(lookup_reads(name))
     reg = name_status(name)
     status = reg["status"]
     if status in ("registered", "grace"):
@@ -788,7 +1169,7 @@ def resolve(name: str):
         }
 
     node = node_of(name)
-    node_hex = node.hex()
+    prefetch(lookup_reads(name))
 
     # Before the resolver lookup, so a lapsed name is not reported as noResolver.
     reg = name_status(name)
@@ -806,13 +1187,12 @@ def resolve(name: str):
         }
         return (404 if reg["status"] == "unregistered" else 410), body
 
-    resolver_raw = eth_call(registry, selector("resolver(bytes32)") + node_hex)
-    resolver_addr = decode_address(resolver_raw)
+    resolver_addr = decode_address(eth_call(*resolver_call(registry, node)))
     if resolver_addr == ZERO_ADDR:
         # A registered name always resolves: with no resolver set the record is
         # still returned with every field unset, so "taken until <date>" stays
         # answerable. For a subname, no owner means nobody created it.
-        owner = decode_address(eth_call(registry, selector("owner(bytes32)") + node_hex))
+        owner = decode_address(eth_call(*owner_call(registry, node)))
         if len(name.split(".")) > 2 and owner == ZERO_ADDR:
             return 404, {
                 "name": name,
@@ -837,9 +1217,9 @@ def resolve(name: str):
             **reg,
         }
 
-    owner_raw = eth_call(registry, selector("owner(bytes32)") + node_hex)
-    owner = decode_address(owner_raw)
+    owner = decode_address(eth_call(*owner_call(registry, node)))
 
+    prefetch(record_reads(resolver_addr, node))
     texts = {}
     for k in TEXT_KEYS:
         try:
@@ -878,12 +1258,34 @@ def resolve(name: str):
 # ---------- HTTP layer ----------
 
 class Handler(BaseHTTPRequestHandler):
+    # the smp-server reuses a connection only after an HTTP/1.1 response
+    protocol_version = "HTTP/1.1"
+    # idle seconds before a kept-alive connection is closed; the smp-server closes
+    # its idle ones after 30-35 s, so it is the one to close them
+    timeout = 60
+
+    def setup(self):
+        super().setup()
+        # headers and body go out in separate writes, which Nagle holds for the client's delayed ACK
+        self.request.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
+
+    def handle_one_request(self):
+        # on a kept-alive connection these still hold the previous request's values
+        self._started = self._sent = self.path = self.headers = None
+        super().handle_one_request()
+
     def do_GET(self):  # noqa: N802 - http.server contract
+        self._started = time.monotonic()
+        if self._has_body():
+            # an unread body would be taken for the next request on this connection, which a
+            # reverse proxy reusing it would pass to another client (request smuggling)
+            self._respond(400, {"error": "unexpectedBody"}, close=True)
+            return
         path = urlparse(self.path).path
         parts = [unquote(p) for p in path.split("/") if p]
 
         if parts == ["health"]:
-            self._respond(200, {"ok": True, "rpc": RPC, "registries": REGISTRIES, **head_block()})
+            self._respond(200, {"ok": True, "rpc": public_endpoint(RPC), "registries": REGISTRIES, **head_block()})
             return
 
         if len(parts) == 3 and parts[0] == "v2" and parts[1] == "resolve":
@@ -892,7 +1294,8 @@ class Handler(BaseHTTPRequestHandler):
                 self._respond(400, {"name": name, "error": "notFullyQualified"})
                 return
             try:
-                status, body = registration(name)
+                with request_reads():
+                    status, body = registration(name)
             except Exception as e:  # surface upstream errors as 502
                 status, body = 502, upstream_error({"name": name}, e)
             self._respond(status, body)
@@ -915,7 +1318,8 @@ class Handler(BaseHTTPRequestHandler):
                 )
                 return
             try:
-                status, body = resolve(name)
+                with request_reads():
+                    status, body = resolve(name)
             except Exception as e:  # surface upstream errors as 502
                 status, body = 502, upstream_error({"name": name}, e)
             self._respond(status, body)
@@ -930,34 +1334,176 @@ class Handler(BaseHTTPRequestHandler):
             },
         )
 
-    def _respond(self, status: int, body: dict):
+    def _has_body(self) -> bool:
+        # a malformed header line ends parsing, leaving the rest, Content-Length included, as payload
+        if self.headers.defects or self.headers.get_payload():
+            return True
+        if self.headers.get_all("Transfer-Encoding"):
+            return True
+        return any(length.strip() != "0" for length in self.headers.get_all("Content-Length") or [])
+
+    def _respond(self, status: int, body: dict, close: bool = False):
         data = json.dumps(body, indent=2).encode()
+        # send_response logs the request, so the size is known before it
+        self._sent = len(data)
         self.send_response(status)
         self.send_header("Content-Type", "application/json")
         self.send_header("Content-Length", str(len(data)))
+        if close:
+            # also sets close_connection, so the server closes it after this response
+            self.send_header("Connection", "close")
         self.end_headers()
         self.wfile.write(data)
 
+    def address_string(self) -> str:
+        forwarded = self.headers.get_all("X-Forwarded-For") if self.headers else None
+        return client_address(self.client_address[0], ",".join(forwarded) if forwarded else None)
+
+    def log_request(self, code="-", size="-"):
+        started = self._started
+        path = urlparse(self.path).path if self.path else None
+        log_event(
+            # the container's health check runs every 30 s
+            logging.DEBUG if path == "/health" else logging.INFO,
+            "request",
+            client=self.address_string(),
+            worker=os.getpid(),
+            method=getattr(self, "command", None),
+            path=unquote(self.path) if self.path else None,
+            status=getattr(code, "value", code),
+            bytes=self._sent,
+            ms=round((time.monotonic() - started) * 1000) if started else None,
+        )
+
+    def log_error(self, fmt, *args):
+        message = fmt % args
+        # a kept-alive connection left idle is closed on purpose
+        level = logging.DEBUG if message.startswith("Request timed out") else logging.WARNING
+        log_event(level, "http_error", client=self.address_string(), worker=os.getpid(), message=message)
+
     def log_message(self, fmt, *args):
-        # Quiet the default per-request access log; route to stderr in one line.
-        sys.stderr.write(f"{self.address_string()} - {fmt % args}\n")
+        log_event(logging.INFO, "http", client=self.address_string(), message=fmt % args)
+
+
+class ResolverServer(ThreadingHTTPServer):
+    # socketserver's default of 5 drops simultaneous connections, each retried by
+    # TCP after 1 s, and the smp-server gives up after 3 s. The kernel caps it at somaxconn.
+    request_queue_size = 128
+
+    def handle_error(self, request, client_address):
+        error = sys.exc_info()[1]
+        if isinstance(error, ConnectionError):
+            # the smp-server hung up, usually after its own timeout
+            log_event(logging.WARNING, "client_gone", client=client_address[0], worker=os.getpid(), error=type(error).__name__)
+        else:
+            log_event(logging.ERROR, "request_failed", client=client_address[0], worker=os.getpid(), exc_info=True)
+
+
+def public_endpoint(url: str) -> str:
+    """The RPC endpoint without credentials, path or query, where a provider key would be."""
+    parsed = urlparse(url)
+    return f"{parsed.scheme}://{parsed.netloc.rpartition('@')[2]}"
+
+
+def client_address(peer: str, forwarded_for: str | None) -> str:
+    """The peer, or behind trusted proxies the last X-Forwarded-For address none of them added."""
+    if not forwarded_for or not _trusted(peer):
+        return peer
+    hops = [hop.strip() for hop in forwarded_for.split(",")]
+    for hop in reversed(hops):
+        # an address it cannot read, and anything claimed before it, is not believed
+        if _address(hop) is None:
+            return peer
+        if not _trusted(hop):
+            return hop
+    return hops[0]
+
+
+def _address(text: str):
+    try:
+        ip = ipaddress.ip_address(text)
+    except ValueError:
+        return None
+    return ip.ipv4_mapped or ip if ip.version == 6 else ip
+
+
+def _trusted(text: str) -> bool:
+    ip = _address(text)
+    return ip is not None and any(ip in net for net in TRUSTED_PROXIES)
+
+
+def serve(reuse_port: bool):
+    server = ResolverServer((BIND, PORT), Handler, bind_and_activate=False)
+    server.allow_reuse_port = reuse_port
+    try:
+        server.server_bind()
+        server.server_activate()
+        server.serve_forever()
+    finally:
+        server.server_close()
+
+
+def stop_on(signum, _frame):
+    log_event(logging.INFO, "stopping", signal=signal.Signals(signum).name)
+    sys.exit(0)
+
+
+def supervise(workers: int):
+    """Runs the workers, each with its own socket on the shared port. One worker
+    exiting stops the others, so the container restarts instead of running short."""
+    children = []
+
+    def stop(*_):
+        for pid in children:
+            try:
+                os.kill(pid, signal.SIGTERM)
+            except ProcessLookupError:
+                pass
+
+    def stop_and_exit(signum, frame):
+        stop()
+        stop_on(signum, frame)
+
+    # before forking, so a signal during startup cannot orphan the workers
+    signal.signal(signal.SIGTERM, stop_and_exit)
+    signal.signal(signal.SIGINT, stop_and_exit)
+    for _ in range(workers):
+        pid = os.fork()
+        if pid == 0:
+            signal.signal(signal.SIGTERM, signal.SIG_DFL)
+            signal.signal(signal.SIGINT, signal.SIG_DFL)
+            code = 0
+            try:
+                serve(reuse_port=True)
+            except BaseException:
+                log_event(logging.ERROR, "worker_failed", worker=os.getpid(), exc_info=True)
+                code = 1
+            os._exit(code)
+        children.append(pid)
+    pid, status = os.wait()
+    log_event(logging.ERROR, "worker_exited", worker=pid, status=status, action="stopping")
+    stop()
+    sys.exit(1)
 
 
 def main():
-    server = ThreadingHTTPServer((BIND, PORT), Handler)
-    sys.stderr.write(
-        f"snrc-resolve listening on {BIND}:{PORT}\n"
-        f"  RPC = {RPC}\n"
-        f"  Registries:\n"
+    setup_logging()
+    log_event(
+        logging.INFO,
+        "listening",
+        bind=BIND,
+        port=PORT,
+        workers=WORKERS,
+        rpc=public_endpoint(RPC),
+        registries=",".join(f"{tld}={addr or '-'}" for tld, addr in REGISTRIES.items()),
+        trusted_proxies=",".join(map(str, TRUSTED_PROXIES)) or None,
     )
-    for tld, addr in REGISTRIES.items():
-        sys.stderr.write(f"    .{tld:<8s} = {addr or '(not configured)'}\n")
-    sys.stderr.write("  GET /v2/resolve/<name>   GET /v1/resolve/<name>   GET /health\n")
-    try:
-        server.serve_forever()
-    except KeyboardInterrupt:
-        sys.stderr.write("\nshutting down\n")
-        server.server_close()
+    if WORKERS > 1:
+        supervise(WORKERS)
+    else:
+        signal.signal(signal.SIGTERM, stop_on)
+        signal.signal(signal.SIGINT, stop_on)
+        serve(reuse_port=False)
 
 
 if __name__ == "__main__":

scripts/resolver/service/test_snrc_resolve.py

@@ -1,15 +1,30 @@
 #!/usr/bin/env python3
 """Unit tests for snrc-resolve helpers.
 
-Run with `python3 -m unittest scripts/resolver/service/test_snrc_resolve.py`.
+Run with `uv run pytest` from scripts/resolver/service.
 """
 
 import contextlib
+import http.client
 import importlib.util
 import io
+import ipaddress
+import json
+import logging
 import os
+import queue
+import re
+import signal
+import socket
+import subprocess
+import sys
+import threading
 import time
 import unittest
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from urllib.error import HTTPError
+from urllib.parse import urlparse
+from urllib.request import Request, urlopen
 
 # snrc-resolve.py has a hyphen, so import it via importlib instead of `import`.
 _HERE = os.path.dirname(os.path.abspath(__file__))
@@ -636,6 +651,25 @@ class PricingTests(unittest.TestCase):
         self.assertEqual(body["basePrice"], self.BASE)
 
 
+class DecodePricesTests(unittest.TestCase):
+    def test_a_count_longer_than_the_answer_is_refused(self):
+        """The count comes from the oracle; decoding it unchecked could loop for ever."""
+        huge = "0x" + snrc.encode_uint(200) + snrc.encode_uint(0x40) + snrc.encode_uint(2**64)
+        outcome = []
+
+        def decode():
+            try:
+                snrc.decode_prices(huge)
+            except RuntimeError as e:
+                outcome.append(e)
+
+        worker = threading.Thread(target=decode, daemon=True)
+        worker.start()
+        worker.join(5)
+        self.assertFalse(worker.is_alive(), "decoding did not stop")
+        self.assertRegex(str(outcome[0]), "short response")
+
+
 class EnsOracleTests(unittest.TestCase):
     """.testing runs an ENS-shaped oracle: it prices in attoUSD per second and
     charges a premium on lapsed names that it does not expose."""
@@ -777,13 +811,17 @@ class ErrorCodeTests(unittest.TestCase):
                 self.assertNotEqual(body["error"], body["message"])
 
     def test_an_upstream_failure_does_not_echo_the_exception(self):
-        with contextlib.redirect_stderr(io.StringIO()) as log:
+        with self.assertLogs("snrc_resolve", "WARNING") as logs:
             body = snrc.upstream_error(
                 {"name": "alice.testing"},
                 RuntimeError("http://user:[email protected]/kEy8 refused"),
             )
         # the operator still sees the detail in the log
-        self.assertIn("secret", log.getvalue())
+        [record] = logs.records
+        self.assertEqual(record.getMessage(), "upstream_error")
+        self.assertEqual(record.fields["name"], "alice.testing")
+        self.assertEqual(record.fields["error"], "RuntimeError")
+        self.assertIn("secret", record.fields["message"])
         self.assertEqual(body["error"], "upstreamError")
         self.assertIn("RuntimeError", body["message"])
         self.assertNotIn("secret", body["message"])
@@ -1035,5 +1073,797 @@ class RegistrationV2Tests(unittest.TestCase):
         self.assertEqual(res["lastBlockTs"], self.now)
         self.assertEqual(res["registration"]["type"], "available")
 
-if __name__ == "__main__":
-    unittest.main()
+
+@unittest.skipUnless(sys.platform.startswith("linux"), "Linux drops SYNs on a full accept queue")
+class ListenBacklogTests(unittest.TestCase):
+    """The smp-server opens a connection per lookup and gives up after 3 s, so a
+    burst the accept queue cannot hold fails: TCP retries a dropped SYN after 1 s."""
+
+    BURST = 50
+
+    def test_a_burst_of_connections_is_queued_while_the_server_is_busy(self):
+        # never accepts, so every connection must wait in the queue
+        server = snrc.ResolverServer(("127.0.0.1", 0), snrc.Handler)
+        clients = []
+        try:
+            for i in range(self.BURST):
+                c = socket.socket()
+                clients.append(c)
+                c.settimeout(0.5)
+                try:
+                    c.connect(server.server_address)
+                except TimeoutError:
+                    self.fail(f"connection {i + 1} of {self.BURST} was not queued")
+        finally:
+            for c in clients:
+                c.close()
+            server.server_close()
+
+
+def _word(value: int) -> bytes:
+    return value.to_bytes(32, "big")
+
+
+def _decode_aggregate3_calls(data: str):
+    """Multicall3.aggregate3 calldata back to (to, data) pairs, written apart
+    from the resolver's encoder so the two check each other."""
+    raw = bytes.fromhex(data[len(snrc.AGGREGATE3):])
+
+    def word(at):
+        return int.from_bytes(raw[at:at + 32], "big")
+
+    array = word(0)
+    base = array + 32
+    calls = []
+    for i in range(word(array)):
+        item = base + word(base + 32 * i)
+        call = item + word(item + 64)
+        calls.append(("0x" + raw[item + 12:item + 32].hex(), "0x" + raw[call + 32:call + 32 + word(call)].hex()))
+    return calls
+
+
+def _encode_aggregate3_results(results) -> str:
+    tuples = [
+        _word(int(ok)) + _word(0x40) + _word(len(data)) + data + b"\x00" * ((-len(data)) % 32)
+        for ok, data in results
+    ]
+    offsets, at = b"", 32 * len(tuples)
+    for t in tuples:
+        offsets += _word(at)
+        at += len(t)
+    return "0x" + (_word(0x20) + _word(len(tuples)) + offsets + b"".join(tuples)).hex()
+
+
+class FakeChain:
+    """Contract state for one registered name and one free name. Any other
+    call reverts, as a view function asked for something unset does."""
+
+    REGISTRY = "0x58fc46996d975c57883564648bda5206d1a0102b"
+    REGISTRAR = "0xef47eb4384b46c89e4482a677c2cbcbd2a6fd85a"
+    CONTROLLER = "0x281ca41311c2aa808c917c4674639d7567b75714"
+    ORACLE = "0x1e0c9a2b9d1a4c8f7b3e5d6a9c2f4b8e1d7a3c50"
+    OWNER = "0xd83bd7e0e6b8a4c1f2593a7b0c4e8d1a6f9b2c37"
+    RESOLVER = "0x80fa2b1c3d4e5f60718293a4b5c6d7e8f9012345"
+    GRACE = 90 * 86400
+    TEXTS = {"nickname": "Acme", "url": "https://acme.example", "simplex.channel": "https://a.example/c#1;https://b.example/c#2"}
+
+    def __init__(self):
+        self.now = int(time.time())
+        acme, free = snrc.label_token("acme"), snrc.label_token("free")
+        node = snrc.node_of("acme.testing")
+        abi_bytes = RegistrationV2Tests._abi_bytes
+        prices = RegistrationV2Tests._prices_return(RegistrationV2Tests())
+        self.answers = {
+            snrc.expires_call(self.REGISTRAR, acme): "0x" + snrc.encode_uint(self.now + 3600),
+            snrc.expires_call(self.REGISTRAR, free): "0x" + snrc.encode_uint(0),
+            snrc.grace_call(self.REGISTRAR): "0x" + snrc.encode_uint(self.GRACE),
+            snrc.label_call(self.REGISTRAR, acme): abi_bytes(b"acme"),
+            snrc.reserved_call(self.CONTROLLER, acme): "0x" + snrc.encode_uint(0),
+            snrc.reserved_call(self.CONTROLLER, free): "0x" + snrc.encode_uint(0),
+            snrc.resolver_call(self.REGISTRY, node): "0x" + snrc.encode_uint(int(self.RESOLVER, 16)),
+            snrc.owner_call(self.REGISTRY, node): "0x" + snrc.encode_uint(int(self.OWNER, 16)),
+            snrc.addr_call(self.RESOLVER, node, snrc.COIN_ETH): abi_bytes(bytes.fromhex(self.OWNER[2:])),
+            snrc.prices_call(self.CONTROLLER): "0x" + snrc.encode_uint(int(self.ORACLE, 16)),
+            snrc.prices_call(self.ORACLE): prices,
+            snrc.min_length_call(self.CONTROLLER): "0x" + snrc.encode_uint(3),
+        }
+        for key, value in self.TEXTS.items():
+            self.answers[snrc.text_call(self.RESOLVER, node, key)] = abi_bytes(value.encode())
+
+    def call(self, to, data):
+        answer = self.answers.get((to.lower(), data))
+        if answer is None:
+            raise RuntimeError("execution reverted")
+        return answer
+
+
+class FakeNode(ThreadingHTTPServer):
+    """A JSON-RPC node over HTTP/1.1 keep-alive, serving FakeChain, with
+    batches and Multicall3, each of which a test can take away."""
+
+    daemon_threads = True
+
+    def __init__(self):
+        super().__init__(("127.0.0.1", 0), _FakeNodeHandler)
+        self.chain = FakeChain()
+        self.block = 100
+        self.requests = 0
+        self.connections = 0
+        self.batch = True
+        self.multicall = True
+        self.multicall_null = False
+        self.status = 200
+        self.hang_up = False
+        self.drop_after_reply = False
+        threading.Thread(target=self.serve_forever, args=(0.05,), daemon=True).start()
+
+    @property
+    def url(self):
+        return f"http://127.0.0.1:{self.server_address[1]}/"
+
+    def stop(self):
+        self.shutdown()
+        self.server_close()
+
+    def answer(self, req):
+        out = {"jsonrpc": "2.0", "id": req.get("id")}
+        method, params = req["method"], req["params"]
+        if method == "eth_blockNumber":
+            out["result"] = hex(self.block)
+        elif method == "eth_getBlockByNumber":
+            out["result"] = {"number": hex(self.block), "timestamp": hex(self.chain.now)}
+        elif method == "eth_call" and params[0]["to"].lower() == snrc.MULTICALL.lower():
+            if not self.multicall:
+                out["error"] = {"code": -32000, "message": "no contract code"}
+            elif self.multicall_null:
+                out["result"] = None
+            else:
+                results = []
+                for to, data in _decode_aggregate3_calls(params[0]["data"]):
+                    try:
+                        results.append((True, bytes.fromhex(self.chain.call(to, data)[2:])))
+                    except RuntimeError:
+                        results.append((False, b""))
+                out["result"] = _encode_aggregate3_results(results)
+        elif method == "eth_call":
+            try:
+                out["result"] = self.chain.call(params[0]["to"], params[0]["data"])
+            except RuntimeError:
+                out["error"] = {"code": 3, "message": "execution reverted"}
+        else:
+            out["error"] = {"code": -32601, "message": "method not found"}
+        return out
+
+
+class _FakeNodeHandler(BaseHTTPRequestHandler):
+    protocol_version = "HTTP/1.1"
+
+    def setup(self):
+        super().setup()
+        # headers and body go out in separate writes, which Nagle holds for the client's delayed ACK
+        self.request.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
+        self.server.connections += 1
+
+    def do_POST(self):  # noqa: N802 - http.server contract
+        request = json.loads(self.rfile.read(int(self.headers["Content-Length"])))
+        node = self.server
+        node.requests += 1
+        if node.hang_up:
+            self.close_connection = True
+            return
+        if node.status != 200:
+            reply = {"error": "unavailable"}
+        elif isinstance(request, list):
+            reply = [node.answer(r) for r in request] if node.batch else {"jsonrpc": "2.0", "id": None, "error": {"code": -32600, "message": "batch not supported"}}
+        else:
+            reply = node.answer(request)
+        data = json.dumps(reply).encode()
+        self.send_response(node.status)
+        self.send_header("Content-Type", "application/json")
+        self.send_header("Content-Length", str(len(data)))
+        self.end_headers()
+        self.wfile.write(data)
+        # closes without `Connection: close`, as a node dropping an idle connection does
+        self.close_connection = node.drop_after_reply
+
+    def log_message(self, fmt, *args):
+        pass
+
+
+class FakeNodeTestCase(unittest.TestCase):
+    """Points the resolver at a FakeNode and at FakeChain's contracts."""
+
+    def setUp(self):
+        self.node = FakeNode()
+        self._saved = (snrc.RPC, snrc.RPC_URL, snrc.REGISTRIES, snrc.REGISTRARS, snrc.CONTROLLERS)
+        snrc.RPC = self.node.url
+        snrc.RPC_URL = urlparse(snrc.RPC)
+        snrc.REGISTRIES = {"testing": FakeChain.REGISTRY, "simplex": ""}
+        snrc.REGISTRARS = {"testing": FakeChain.REGISTRAR}
+        snrc.CONTROLLERS = {"testing": FakeChain.CONTROLLER}
+        snrc._multicall_failed_logged = False
+        self._drain_pool()
+
+    def tearDown(self):
+        self._drain_pool()
+        snrc.RPC, snrc.RPC_URL, snrc.REGISTRIES, snrc.REGISTRARS, snrc.CONTROLLERS = self._saved
+        self.node.stop()
+
+    def _drain_pool(self):
+        while not snrc._rpc_pool.empty():
+            snrc._rpc_pool.get_nowait().close()
+
+    def requests_made(self, action):
+        before = self.node.requests
+        result = action()
+        return result, self.node.requests - before
+
+
+class RpcTransportTests(FakeNodeTestCase):
+    """A lookup makes several reads, and a new connection per read costs CPU
+    and leaves a TIME_WAIT socket each, which exhausts local ports under load."""
+
+    def test_reads_share_one_connection(self):
+        for _ in range(18):
+            self.assertEqual(snrc.rpc("eth_blockNumber", []), hex(self.node.block))
+        self.assertEqual(self.node.connections, 1)
+
+    def test_a_connection_the_node_closed_is_replaced(self):
+        self.node.drop_after_reply = True
+        for _ in range(3):
+            self.assertEqual(snrc.rpc("eth_blockNumber", []), hex(self.node.block))
+        self.assertEqual(self.node.connections, 3)
+
+    def test_a_fresh_connection_that_fails_is_not_retried(self):
+        """Only a pooled connection can be stale; a new one failing means the
+        node is down, and resending would only double the wait."""
+        self.node.hang_up = True
+        with self.assertRaises(ConnectionError):
+            snrc.rpc("eth_blockNumber", [])
+        self.assertEqual(self.node.requests, 1)
+
+    def test_a_node_failure_is_not_a_reverted_call(self):
+        """Callers read RuntimeError as the call reverting and fall back to an
+        empty value, so a node failure must not look like one."""
+        self.node.status = 502
+        with self.assertRaises(HTTPError) as cm:
+            snrc.rpc("eth_blockNumber", [])
+        self.assertNotIsInstance(cm.exception, RuntimeError)
+        self.assertEqual(cm.exception.code, 502)
+
+    def test_a_reverted_call_is_a_runtime_error_and_keeps_the_connection(self):
+        with self.assertRaises(RuntimeError):
+            snrc.eth_call("0x" + "11" * 20, "0xdeadbeef")
+        snrc.rpc("eth_blockNumber", [])
+        self.assertEqual(self.node.connections, 1)
+
+
+class RpcPoolTests(FakeNodeTestCase):
+    def test_connections_beyond_the_pool_are_closed_after_use(self):
+        """Every worker would otherwise keep its peak concurrency open to the node."""
+        saved, snrc._rpc_pool = snrc._rpc_pool, queue.LifoQueue(maxsize=1)
+        try:
+            body = json.dumps({"jsonrpc": "2.0", "id": 1, "method": "eth_blockNumber", "params": []}).encode()
+            kept, extra = snrc._new_rpc_connection(), snrc._new_rpc_connection()
+            snrc._post_pooled(kept, body)
+            # returning a connection to a full pool must not wait for a free slot
+            done = threading.Thread(target=snrc._post_pooled, args=(extra, body), daemon=True)
+            done.start()
+            done.join(5)
+            self.assertFalse(done.is_alive(), "returning a connection to a full pool blocked")
+            self.assertEqual(snrc._rpc_pool.qsize(), 1)
+            self.assertIsNotNone(kept.sock)
+            self.assertIsNone(extra.sock)
+        finally:
+            while not snrc._rpc_pool.empty():
+                snrc._rpc_pool.get_nowait().close()
+            snrc._rpc_pool = saved
+
+
+class BatchedReadsTests(FakeNodeTestCase):
+    """Inside a request a round of reads is one round trip, and its contract
+    reads one multicall, because a node runs the calls of a JSON-RPC batch one
+    after another. Answers must be exactly those of reading one call at a time."""
+
+    def batched(self, answer, name):
+        def action():
+            with snrc.request_reads():
+                return answer(name)
+        return self.requests_made(action)
+
+    def assert_same_answer(self, answer, name, round_trips):
+        one_by_one, one_by_one_trips = self.requests_made(lambda: answer(name))
+        batched, batched_trips = self.batched(answer, name)
+        self.assertEqual(batched, one_by_one)
+        self.assertEqual(batched_trips, round_trips)
+        self.assertGreater(one_by_one_trips, round_trips)
+        return batched
+
+    def test_a_registered_name_takes_two_round_trips(self):
+        status, body = self.assert_same_answer(snrc.registration, "acme.testing", 2)
+        record = body["registration"]["nameRecord"]
+        self.assertEqual(record["nickname"], "Acme")
+        self.assertEqual(record["simplexChannel"], ["https://a.example/c#1", "https://b.example/c#2"])
+        self.assertIsNone(record["btc"])
+
+    def test_a_hashed_query_is_named_from_the_same_round_trip(self):
+        hashed = "[" + snrc.keccak(b"acme").hex() + "].testing"
+        status, body = self.assert_same_answer(snrc.registration, hashed, 2)
+        self.assertEqual(body["registration"]["nameRecord"]["name"], "acme.testing")
+
+    def test_an_available_name_is_priced_in_three_round_trips(self):
+        status, body = self.assert_same_answer(snrc.registration, "free.testing", 3)
+        self.assertEqual(body["registration"]["type"], "available")
+
+    def test_v1_answers_the_same(self):
+        self.assert_same_answer(snrc.resolve, "acme.testing", 2)
+
+    def test_without_multicall_a_round_is_still_one_batch(self):
+        self.node.multicall = False
+        with self.assertLogs("snrc_resolve", "WARNING") as logs:
+            self.assert_same_answer(snrc.registration, "acme.testing", 4)
+        [record] = logs.records
+        self.assertEqual((record.getMessage(), record.fields["fallback"]), ("multicall_unavailable", "batch"))
+
+    def test_a_multicall_without_a_result_falls_back_to_a_batch(self):
+        self.node.multicall_null = True
+        with self.assertLogs("snrc_resolve", "WARNING") as logs:
+            self.assert_same_answer(snrc.registration, "acme.testing", 4)
+        self.assertEqual([r.getMessage() for r in logs.records], ["multicall_unavailable"])
+
+    def test_an_answer_with_neither_result_nor_error_is_not_taken_as_one(self):
+        reads = {}
+        snrc._remember(reads, [("eth_blockNumber", [])], [{"jsonrpc": "2.0", "id": 0}])
+        self.assertEqual(reads, {})
+
+    def test_a_node_that_does_not_batch_is_read_one_call_at_a_time(self):
+        self.node.batch = False
+        one_by_one, one_by_one_trips = self.requests_made(lambda: snrc.registration("acme.testing"))
+        with self.assertNoLogs("snrc_resolve"):
+            batched, batched_trips = self.batched(snrc.registration, "acme.testing")
+        self.assertEqual(batched, one_by_one)
+        # one refused batch per round, then the reads the batch would have made
+        self.assertEqual(batched_trips, one_by_one_trips + 2)
+
+    def test_a_read_reverted_in_the_multicall_is_a_reverted_call(self):
+        with snrc.request_reads():
+            snrc.prefetch([snrc.eth_call_read(*snrc.grace_call(FakeChain.REGISTRAR)), snrc.eth_call_read(FakeChain.REGISTRY, "0xdeadbeef")])
+            _, trips = self.requests_made(lambda: self.assertRaises(RuntimeError, snrc.eth_call, FakeChain.REGISTRY, "0xdeadbeef"))
+        self.assertEqual(trips, 0)
+
+    def test_outside_a_request_nothing_is_prefetched(self):
+        _, trips = self.requests_made(lambda: snrc.prefetch(snrc.lookup_reads("acme.testing")))
+        self.assertEqual(trips, 0)
+
+
+class Aggregate3Tests(unittest.TestCase):
+    def test_calls_are_encoded_as_multicall3_reads_them(self):
+        calls = [(FakeChain.REGISTRY, "0x0178b8bf" + "11" * 32), (FakeChain.RESOLVER, "0x59d1d43c" + "22" * 100)]
+        data = snrc.encode_aggregate3(calls)
+        self.assertTrue(data.startswith(snrc.AGGREGATE3))
+        self.assertEqual(_decode_aggregate3_calls(data), calls)
+
+    def test_results_are_decoded_with_their_success_flags(self):
+        results = [(True, b"\x01" * 40), (False, b""), (True, b"")]
+        self.assertEqual(snrc.decode_aggregate3(_encode_aggregate3_results(results)), results)
+
+    def test_a_truncated_answer_is_refused(self):
+        whole = _encode_aggregate3_results([(True, b"\x01" * 40)])
+        with self.assertRaises(ValueError):
+            snrc.decode_aggregate3(whole[:-64])
+        with self.assertRaises(ValueError):
+            snrc.decode_aggregate3("0x")
+
+
+class RequestLogTests(unittest.TestCase):
+    """Each request is one event with the client it came from, behind a reverse
+    proxy the client the proxy names rather than the proxy itself."""
+
+    def setUp(self):
+        self.server = snrc.ResolverServer(("127.0.0.1", 0), snrc.Handler)
+        threading.Thread(target=self.server.serve_forever, args=(0.05,), daemon=True).start()
+        self._saved = snrc.TRUSTED_PROXIES
+
+    def tearDown(self):
+        snrc.TRUSTED_PROXIES = self._saved
+        self.server.shutdown()
+        self.server.server_close()
+
+    def request(self, path, headers=None, level="INFO"):
+        url = f"http://127.0.0.1:{self.server.server_address[1]}{path}"
+        with self.assertLogs("snrc_resolve", level) as logs:
+            try:
+                urlopen(Request(url, headers=headers or {}), timeout=5).read()
+            except HTTPError:
+                pass
+            deadline = time.monotonic() + 2
+            while not any(r.getMessage() == "request" for r in logs.records) and time.monotonic() < deadline:
+                time.sleep(0.01)
+        [record] = [r for r in logs.records if r.getMessage() == "request"]
+        return record
+
+    def test_a_request_is_logged_with_its_outcome(self):
+        record = self.request("/v2/resolve/x.simplex")
+        fields = record.fields
+        self.assertEqual(record.levelname, "INFO")
+        self.assertEqual(
+            (fields["client"], fields["method"], fields["path"], fields["status"], fields["worker"]),
+            ("127.0.0.1", "GET", "/v2/resolve/x.simplex", 400, os.getpid()),
+        )
+        self.assertGreater(fields["bytes"], 0)
+        self.assertGreaterEqual(fields["ms"], 0)
+
+    def test_a_trusted_proxy_names_the_client(self):
+        snrc.TRUSTED_PROXIES = (ipaddress.ip_network("127.0.0.1/32"),)
+        record = self.request("/v2/resolve/x.simplex", {"X-Forwarded-For": "203.0.113.7"})
+        self.assertEqual(record.fields["client"], "203.0.113.7")
+
+    def test_any_other_peer_cannot_name_itself(self):
+        record = self.request("/v2/resolve/x.simplex", {"X-Forwarded-For": "203.0.113.7"})
+        self.assertEqual(record.fields["client"], "127.0.0.1")
+
+    def test_a_bad_request_is_not_logged_with_the_previous_ones_fields(self):
+        """On a kept-alive connection the handler still held the last request's path and size."""
+        with self.assertLogs("snrc_resolve", "INFO") as logs:
+            with socket.create_connection(self.server.server_address, timeout=5) as sock:
+                sock.sendall(b"GET /v2/resolve/x.simplex HTTP/1.1\r\nHost: x\r\n\r\n")
+                sock.recv(65536)
+                sock.sendall(b"BOGUS\r\n\r\n")
+                while sock.recv(65536):
+                    pass
+        first, second = [r for r in logs.records if r.getMessage() == "request"]
+        self.assertEqual(first.fields["path"], "/v2/resolve/x.simplex")
+        self.assertEqual((second.fields["status"], second.fields["path"], second.fields["bytes"], second.fields["ms"]), (400, None, None, None))
+
+    def test_health_checks_are_logged_only_at_debug(self):
+        """The container checks /health every 30 s."""
+        record = self.request("/health", level="DEBUG")
+        self.assertEqual(record.levelname, "DEBUG")
+
+
+class RequestBodyTests(unittest.TestCase):
+    """Behind a reverse proxy that reuses connections, a body left unread would be
+    answered as another request, and that answer given to the proxy's next client."""
+
+    def setUp(self):
+        self.server = snrc.ResolverServer(("127.0.0.1", 0), snrc.Handler)
+        threading.Thread(target=self.server.serve_forever, args=(0.05,), daemon=True).start()
+        self._log = contextlib.redirect_stderr(io.StringIO())
+        self._log.__enter__()
+
+    def tearDown(self):
+        self._log.__exit__(None, None, None)
+        self.server.shutdown()
+        self.server.server_close()
+
+    def exchange(self, head: bytes, body: bytes = b"") -> bytes:
+        with socket.create_connection(self.server.server_address, timeout=5) as sock:
+            sock.sendall(head + body)
+            data = b""
+            while chunk := sock.recv(65536):
+                data += chunk
+                if data.count(b"HTTP/1.1 ") > 1:
+                    break
+            return data
+
+    SMUGGLED = b"GET /v2/resolve/y.simplex HTTP/1.1\r\nHost: x\r\n\r\n"
+
+    def test_a_body_is_refused_and_the_connection_closed(self):
+        data = self.exchange(b"GET /v2/resolve/x.simplex HTTP/1.1\r\nHost: x\r\nContent-Length: %d\r\n\r\n" % len(self.SMUGGLED), self.SMUGGLED)
+        self.assertEqual(data.count(b"HTTP/1.1 "), 1)
+        self.assertIn(b"HTTP/1.1 400 ", data)
+        self.assertIn(b"Connection: close", data)
+        self.assertNotIn(b"y.simplex", data)
+
+    def test_a_length_after_a_malformed_header_line_is_still_seen(self):
+        """Python stops parsing headers at a malformed line and keeps the rest as payload."""
+        data = self.exchange(b"GET /v2/resolve/x.simplex HTTP/1.1\r\nX-Junk : 1\r\nContent-Length: %d\r\n\r\n" % len(self.SMUGGLED), self.SMUGGLED)
+        self.assertEqual(data.count(b"HTTP/1.1 "), 1)
+        self.assertIn(b"unexpectedBody", data)
+        self.assertNotIn(b"y.simplex", data)
+
+    def test_a_chunked_body_is_refused_and_the_connection_closed(self):
+        chunked = b"%x\r\n" % len(self.SMUGGLED) + self.SMUGGLED + b"\r\n0\r\n\r\n"
+        data = self.exchange(b"GET /v2/resolve/x.simplex HTTP/1.1\r\nHost: x\r\nTransfer-Encoding: chunked\r\n\r\n", chunked)
+        self.assertEqual(data.count(b"HTTP/1.1 "), 1)
+        self.assertNotIn(b"y.simplex", data)
+
+    def test_an_empty_body_keeps_the_connection(self):
+        conn = http.client.HTTPConnection(*self.server.server_address, timeout=5)
+        answers = []
+        for _ in range(2):
+            conn.request("GET", "/v2/resolve/x.simplex", headers={"Content-Length": "0"})
+            res = conn.getresponse()
+            answers.append((res.status, json.loads(res.read())["error"], conn.sock))
+        conn.close()
+        self.assertEqual([a[:2] for a in answers], [(400, "tldNotConfigured")] * 2)
+        self.assertIs(answers[0][2], answers[1][2])
+
+
+class PublicEndpointTests(unittest.TestCase):
+    def test_credentials_path_and_query_are_not_shown(self):
+        self.assertEqual(snrc.public_endpoint("https://user:[email protected]:8443/v2/KEY?apikey=K"), "https://rpc.example:8443")
+
+    def test_a_plain_endpoint_is_unchanged(self):
+        self.assertEqual(snrc.public_endpoint("http://reth:8545"), "http://reth:8545")
+
+    def test_an_ipv6_host_keeps_its_brackets(self):
+        self.assertEqual(snrc.public_endpoint("http://u:p@[::1]:8545/key"), "http://[::1]:8545")
+
+    def test_a_malformed_port_does_not_break_health(self):
+        self.assertEqual(snrc.public_endpoint("http://reth:port"), "http://reth:port")
+
+
+class ClientAddressTests(unittest.TestCase):
+    PROXY = "172.18.0.1"
+
+    def setUp(self):
+        self._saved = snrc.TRUSTED_PROXIES
+        snrc.TRUSTED_PROXIES = (ipaddress.ip_network("172.16.0.0/12"),)
+
+    def tearDown(self):
+        snrc.TRUSTED_PROXIES = self._saved
+
+    def test_a_peer_that_is_no_proxy_is_the_client(self):
+        self.assertEqual(snrc.client_address("198.51.100.9", "203.0.113.7"), "198.51.100.9")
+
+    def test_without_the_header_the_proxy_is_the_client(self):
+        self.assertEqual(snrc.client_address(self.PROXY, None), self.PROXY)
+
+    def test_the_address_the_proxy_added_is_the_client(self):
+        self.assertEqual(snrc.client_address(self.PROXY, "203.0.113.7"), "203.0.113.7")
+
+    def test_addresses_a_client_sent_itself_are_not_believed(self):
+        """A proxy appends the address it saw, so only the last untrusted one is known."""
+        self.assertEqual(snrc.client_address(self.PROXY, "6.6.6.6, 203.0.113.7"), "203.0.113.7")
+
+    def test_trusted_proxies_in_a_chain_are_skipped(self):
+        self.assertEqual(snrc.client_address(self.PROXY, "203.0.113.7, 172.18.0.5"), "203.0.113.7")
+
+    def test_an_address_that_does_not_parse_is_not_believed(self):
+        self.assertEqual(snrc.client_address(self.PROXY, "203.0.113.7, not-an-ip"), self.PROXY)
+
+    def test_an_ipv4_mapped_peer_is_matched_as_ipv4(self):
+        self.assertEqual(snrc.client_address("::ffff:172.18.0.1", "203.0.113.7"), "203.0.113.7")
+
+    def test_an_ipv6_client_is_kept(self):
+        self.assertEqual(snrc.client_address(self.PROXY, "2001:db8::7"), "2001:db8::7")
+
+
+class RequestErrorTests(unittest.TestCase):
+    def setUp(self):
+        self.server = snrc.ResolverServer(("127.0.0.1", 0), snrc.Handler)
+
+    def tearDown(self):
+        self.server.server_close()
+
+    def test_a_client_that_hung_up_is_a_warning_without_a_traceback(self):
+        """What an smp-server that gave up after its timeout leaves behind."""
+        with self.assertLogs("snrc_resolve", "WARNING") as logs:
+            try:
+                raise BrokenPipeError
+            except BrokenPipeError:
+                self.server.handle_error(None, ("198.51.100.9", 4000))
+        [record] = logs.records
+        self.assertEqual((record.levelname, record.getMessage(), record.fields["error"]), ("WARNING", "client_gone", "BrokenPipeError"))
+        self.assertIsNone(record.exc_info)
+
+    def test_a_failure_is_an_error_with_its_traceback(self):
+        with self.assertLogs("snrc_resolve", "ERROR") as logs:
+            try:
+                raise KeyError("boom")
+            except KeyError:
+                self.server.handle_error(None, ("198.51.100.9", 4000))
+        [record] = logs.records
+        self.assertEqual((record.getMessage(), record.fields["client"]), ("request_failed", "198.51.100.9"))
+        self.assertIsNotNone(record.exc_info)
+
+
+class KeepAliveTests(unittest.TestCase):
+    """The smp-server keeps a resolver connection only after an HTTP/1.1
+    response, and otherwise connects for every lookup."""
+
+    def setUp(self):
+        self.server = snrc.ResolverServer(("127.0.0.1", 0), snrc.Handler)
+        threading.Thread(target=self.server.serve_forever, args=(0.05,), daemon=True).start()
+        self._saved_timeout = snrc.Handler.timeout
+        self._log = contextlib.redirect_stderr(io.StringIO())
+        self._log.__enter__()
+
+    def tearDown(self):
+        self._log.__exit__(None, None, None)
+        snrc.Handler.timeout = self._saved_timeout
+        self.server.shutdown()
+        self.server.server_close()
+
+    def test_requests_share_one_connection_without_delay(self):
+        conn = http.client.HTTPConnection(*self.server.server_address, timeout=5)
+        start = time.monotonic()
+        for i in range(20):
+            conn.request("GET", "/v2/resolve/x.simplex")
+            res = conn.getresponse()
+            res.read()
+            self.assertEqual((res.version, res.status), (11, 400))
+            if i == 0:
+                sock = conn.sock
+            self.assertIs(conn.sock, sock)
+        conn.close()
+        # a response held by Nagle for the delayed ACK takes ~40 ms, 20 of them over 0.8 s
+        self.assertLess(time.monotonic() - start, 0.5)
+
+    def test_idle_connections_outlast_the_smp_servers(self):
+        """http-client drops a connection idle for 30 s, checked every 5 s. A
+        resolver that closed sooner would race the client reusing it, and one
+        that never closed would keep a thread per dead connection."""
+        self.assertIsNotNone(self._saved_timeout)
+        self.assertGreater(self._saved_timeout, 35)
+
+    def test_an_idle_connection_is_closed(self):
+        snrc.Handler.timeout = 0.2
+        with socket.create_connection(self.server.server_address, timeout=5) as sock:
+            time.sleep(0.5)
+            self.assertEqual(sock.recv(1), b"")
+
+
+@unittest.skipUnless(sys.platform.startswith("linux"), "reads worker processes from /proc")
+class WorkerProcessesTests(unittest.TestCase):
+    """Workers share the port, and the service stops as a whole, so the
+    container restarts rather than serving on fewer workers."""
+
+    def setUp(self):
+        self.node = FakeNode()
+        with socket.socket() as s:
+            s.bind(("127.0.0.1", 0))
+            self.port = s.getsockname()[1]
+        env = dict(os.environ, SNRC_RPC=self.node.url, SNRC_BIND="127.0.0.1", SNRC_PORT=str(self.port), SNRC_WORKERS="2",
+                   SNRC_REGISTRY_TESTING=FakeChain.REGISTRY, SNRC_REGISTRAR_TESTING=FakeChain.REGISTRAR, SNRC_CONTROLLER_TESTING=FakeChain.CONTROLLER)
+        self.service = subprocess.Popen([sys.executable, os.path.join(_HERE, "snrc-resolve.py")], env=env, stderr=subprocess.PIPE, text=True)
+        self.workers = self._wait_for_workers(2)
+
+    def tearDown(self):
+        if self.service.poll() is None:
+            self.service.kill()
+            self.service.wait()
+        self.service.stderr.close()
+        for pid in self.workers:
+            with contextlib.suppress(ProcessLookupError):
+                os.kill(pid, signal.SIGKILL)
+        self.node.stop()
+
+    def _wait_for_workers(self, count):
+        deadline = time.monotonic() + 10
+        while time.monotonic() < deadline:
+            workers = [int(p) for p in os.listdir("/proc") if p.isdigit() and self._parent(p) == self.service.pid]
+            if len(workers) == count and self._serving():
+                return workers
+            time.sleep(0.05)
+        self.fail("workers did not start")
+
+    @staticmethod
+    def _parent(pid):
+        try:
+            with open(f"/proc/{pid}/stat") as f:
+                return int(f.read().rsplit(")", 1)[1].split()[1])
+        except (FileNotFoundError, ProcessLookupError):
+            return None
+
+    def _serving(self):
+        try:
+            with socket.create_connection(("127.0.0.1", self.port), timeout=1):
+                return True
+        except OSError:
+            return False
+
+    def _gone(self, pid):
+        deadline = time.monotonic() + 5
+        while time.monotonic() < deadline:
+            if self._parent(pid) != self.service.pid:
+                return True
+            time.sleep(0.05)
+        return False
+
+    def test_workers_answer_on_the_shared_port(self):
+        for _ in range(20):
+            with urlopen(f"http://127.0.0.1:{self.port}/v2/resolve/acme.testing", timeout=5) as res:
+                self.assertEqual(json.loads(res.read())["registration"]["type"], "registered")
+
+    def test_stopping_the_service_stops_every_worker(self):
+        self.service.send_signal(signal.SIGTERM)
+        self.assertEqual(self.service.wait(timeout=5), 0)
+        self.assertTrue(all(self._gone(pid) for pid in self.workers))
+        self.assertRegex(self.service.stderr.read(), r"INFO  stopping  signal=SIGTERM")
+
+    def test_a_worker_exiting_stops_the_service(self):
+        os.kill(self.workers[0], signal.SIGKILL)
+        self.assertEqual(self.service.wait(timeout=5), 1)
+        self.assertTrue(self._gone(self.workers[1]))
+
+
+TIME = r"\d{4}-\d\d-\d\dT\d\d:\d\d:\d\d\.\d{3}Z"
+
+
+class LogTestCase(unittest.TestCase):
+    def setUp(self):
+        self._saved = (snrc.LOG_FORMAT, snrc.LOG_COLOR, snrc.LOG_LEVEL)
+        snrc.LOG_FORMAT, snrc.LOG_COLOR, snrc.LOG_LEVEL = "text", "never", "info"
+        self.out = io.StringIO()
+
+    def tearDown(self):
+        snrc.LOG_FORMAT, snrc.LOG_COLOR, snrc.LOG_LEVEL = self._saved
+        snrc.LOGGER.handlers[:] = []
+        snrc.LOGGER.setLevel(logging.NOTSET)
+        snrc.LOGGER.propagate = True
+
+    def emit(self, level=logging.INFO, name="request", **fields):
+        snrc.setup_logging(self.out)
+        snrc.log_event(level, name, **fields)
+        return self.out.getvalue()
+
+
+class TextFormatTests(LogTestCase):
+    def test_a_line_is_utc_time_level_event_and_fields(self):
+        line = self.emit(client="203.0.113.7", path="/v2/resolve/[4fdd].testing", status=200, ms=7)
+        self.assertRegex(line, rf"^{TIME} INFO  request  client=203\.0\.113\.7 path=/v2/resolve/\[4fdd\]\.testing status=200 ms=7\n$")
+
+    def test_values_that_would_be_misread_are_quoted(self):
+        line = self.emit(message='says "hi" here', empty="", missing=None, eq="a=b")
+        self.assertIn('message="says \\"hi\\" here" empty="" missing=- eq="a=b"', line)
+
+    def test_levels_are_named_in_five_columns(self):
+        self.assertRegex(self.emit(logging.WARNING, "upstream_error"), rf"^{TIME} WARN  upstream_error\n$")
+
+    def test_below_the_configured_level_nothing_is_written(self):
+        self.assertEqual(self.emit(logging.DEBUG), "")
+
+    def test_an_exception_follows_its_line(self):
+        snrc.setup_logging(self.out)
+        try:
+            raise KeyError("boom")
+        except KeyError:
+            snrc.log_event(logging.ERROR, "request_failed", exc_info=True)
+        self.assertRegex(self.out.getvalue(), rf"(?s)^{TIME} ERROR request_failed\nTraceback .*KeyError: 'boom'\n$")
+
+
+class ColorTests(LogTestCase):
+    def test_colours_mark_the_level_and_the_status_class(self):
+        snrc.LOG_COLOR = "always"
+        line = self.emit(status=503)
+        self.assertIn("\033[32mINFO ", line)
+        self.assertIn("\033[31m503\033[0m", line)
+
+    def test_auto_leaves_output_that_is_no_terminal_plain(self):
+        snrc.LOG_COLOR = "auto"
+        self.assertNotIn("\033[", self.emit(status=200))
+
+    def test_never_is_plain(self):
+        self.assertNotIn("\033[", self.emit(status=200))
+
+
+class JsonFormatTests(LogTestCase):
+    def test_a_line_is_one_json_object(self):
+        snrc.LOG_FORMAT = "json"
+        record = json.loads(self.emit(client="203.0.113.7", status=200, missing=None))
+        self.assertRegex(record.pop("time"), rf"^{TIME}$")
+        self.assertEqual(record, {"level": "info", "event": "request", "client": "203.0.113.7", "status": 200, "missing": None})
+
+    def test_an_exception_is_a_field(self):
+        snrc.LOG_FORMAT = "json"
+        snrc.setup_logging(self.out)
+        try:
+            raise KeyError("boom")
+        except KeyError:
+            snrc.log_event(logging.ERROR, "request_failed", exc_info=True)
+        self.assertIn("KeyError: 'boom'", json.loads(self.out.getvalue())["exception"])
+
+
+class SetupTests(LogTestCase):
+    def test_unknown_settings_are_refused_at_start(self):
+        for setting, value in (("LOG_FORMAT", "yaml"), ("LOG_COLOR", "sometimes"), ("LOG_LEVEL", "loud")):
+            with self.subTest(setting=setting):
+                saved = getattr(snrc, setting)
+                setattr(snrc, setting, value)
+                try:
+                    with self.assertRaisesRegex(ValueError, re.escape(f"SNRC_{setting}")):
+                        snrc.setup_logging(self.out)
+                finally:
+                    setattr(snrc, setting, saved)

scripts/resolver/service/uv.lock

@@ -0,0 +1,156 @@
+version = 1
+revision = 3
+requires-python = ">=3.11"
+
+[[package]]
+name = "colorama"
+version = "0.4.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44", size = 27697, upload-time = "2022-10-25T02:36:22.414Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335, upload-time = "2022-10-25T02:36:20.889Z" },
+]
+
+[[package]]
+name = "eth-hash"
+version = "0.8.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3c/f5/c67fc24f2f676aa9b7ab29679d44f113f314c817207cd4319353356f62da/eth_hash-0.8.0.tar.gz", hash = "sha256:b009752b620da2e9c7668014849d1f5fadbe4f138603f1871cc5d4ca706896b1", size = 12225, upload-time = "2026-03-25T16:36:55.099Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/87/87/b36792150ca0b28e4df683a34be15a61461ca0e349e5b5cf3ec8f694edb9/eth_hash-0.8.0-py3-none-any.whl", hash = "sha256:523718a51b369ab89866b929a5c93c52978cd866ea309192ad980dd8271f9fac", size = 7965, upload-time = "2026-03-25T16:36:54.205Z" },
+]
+
+[package.optional-dependencies]
+pycryptodome = [
+    { name = "pycryptodome" },
+]
+
+[[package]]
+name = "iniconfig"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730", size = 20503, upload-time = "2025-10-18T21:55:43.219Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12", size = 7484, upload-time = "2025-10-18T21:55:41.639Z" },
+]
+
+[[package]]
+name = "packaging"
+version = "26.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7d/fa/3944b40b07da9ce895c0e6303a5ab7d53da063554f534556b134a54d6093/packaging-26.3.tar.gz", hash = "sha256:94edc256424af38762eb31306eed28beb9f0efc50a8837492c9d6fd6004aed79", size = 313412, upload-time = "2026-08-04T18:15:28.737Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/63/34/ba1c580383c9eada3711951fef0795c80b829a078d72188184bcab9dd527/packaging-26.3-py3-none-any.whl", hash = "sha256:d7193f7c8e4e93f444fde0262bf90af30e16fa0ad0ad44cb553c87339b23cd1c", size = 129956, upload-time = "2026-08-04T18:15:27.159Z" },
+]
+
+[[package]]
+name = "pluggy"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3", size = 69412, upload-time = "2025-05-15T12:30:07.975Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746", size = 20538, upload-time = "2025-05-15T12:30:06.134Z" },
+]
+
+[[package]]
+name = "pycryptodome"
+version = "3.23.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8e/a6/8452177684d5e906854776276ddd34eca30d1b1e15aa1ee9cefc289a33f5/pycryptodome-3.23.0.tar.gz", hash = "sha256:447700a657182d60338bab09fdb27518f8856aecd80ae4c6bdddb67ff5da44ef", size = 4921276, upload-time = "2025-05-17T17:21:45.242Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/04/5d/bdb09489b63cd34a976cc9e2a8d938114f7a53a74d3dd4f125ffa49dce82/pycryptodome-3.23.0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0011f7f00cdb74879142011f95133274741778abba114ceca229adbf8e62c3e4", size = 2495152, upload-time = "2025-05-17T17:20:20.833Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/ce/7840250ed4cc0039c433cd41715536f926d6e86ce84e904068eb3244b6a6/pycryptodome-3.23.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:90460fc9e088ce095f9ee8356722d4f10f86e5be06e2354230a9880b9c549aae", size = 1639348, upload-time = "2025-05-17T17:20:23.171Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/f0/991da24c55c1f688d6a3b5a11940567353f74590734ee4a64294834ae472/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4764e64b269fc83b00f682c47443c2e6e85b18273712b98aa43bcb77f8570477", size = 2184033, upload-time = "2025-05-17T17:20:25.424Z" },
+    { url = "https://files.pythonhosted.org/packages/54/16/0e11882deddf00f68b68dd4e8e442ddc30641f31afeb2bc25588124ac8de/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eb8f24adb74984aa0e5d07a2368ad95276cf38051fe2dc6605cbcf482e04f2a7", size = 2270142, upload-time = "2025-05-17T17:20:27.808Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/fc/4347fea23a3f95ffb931f383ff28b3f7b1fe868739182cb76718c0da86a1/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d97618c9c6684a97ef7637ba43bdf6663a2e2e77efe0f863cce97a76af396446", size = 2309384, upload-time = "2025-05-17T17:20:30.765Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/d9/c5261780b69ce66d8cfab25d2797bd6e82ba0241804694cd48be41add5eb/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9a53a4fe5cb075075d515797d6ce2f56772ea7e6a1e5e4b96cf78a14bac3d265", size = 2183237, upload-time = "2025-05-17T17:20:33.736Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/6f/3af2ffedd5cfa08c631f89452c6648c4d779e7772dfc388c77c920ca6bbf/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:763d1d74f56f031788e5d307029caef067febf890cd1f8bf61183ae142f1a77b", size = 2343898, upload-time = "2025-05-17T17:20:36.086Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/dc/9060d807039ee5de6e2f260f72f3d70ac213993a804f5e67e0a73a56dd2f/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:954af0e2bd7cea83ce72243b14e4fb518b18f0c1649b576d114973e2073b273d", size = 2269197, upload-time = "2025-05-17T17:20:38.414Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/34/e6c8ca177cb29dcc4967fef73f5de445912f93bd0343c9c33c8e5bf8cde8/pycryptodome-3.23.0-cp313-cp313t-win32.whl", hash = "sha256:257bb3572c63ad8ba40b89f6fc9d63a2a628e9f9708d31ee26560925ebe0210a", size = 1768600, upload-time = "2025-05-17T17:20:40.688Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/1d/89756b8d7ff623ad0160f4539da571d1f594d21ee6d68be130a6eccb39a4/pycryptodome-3.23.0-cp313-cp313t-win_amd64.whl", hash = "sha256:6501790c5b62a29fcb227bd6b62012181d886a767ce9ed03b303d1f22eb5c625", size = 1799740, upload-time = "2025-05-17T17:20:42.413Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/61/35a64f0feaea9fd07f0d91209e7be91726eb48c0f1bfc6720647194071e4/pycryptodome-3.23.0-cp313-cp313t-win_arm64.whl", hash = "sha256:9a77627a330ab23ca43b48b130e202582e91cc69619947840ea4d2d1be21eb39", size = 1703685, upload-time = "2025-05-17T17:20:44.388Z" },
+    { url = "https://files.pythonhosted.org/packages/db/6c/a1f71542c969912bb0e106f64f60a56cc1f0fabecf9396f45accbe63fa68/pycryptodome-3.23.0-cp37-abi3-macosx_10_9_universal2.whl", hash = "sha256:187058ab80b3281b1de11c2e6842a357a1f71b42cb1e15bce373f3d238135c27", size = 2495627, upload-time = "2025-05-17T17:20:47.139Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/4e/a066527e079fc5002390c8acdd3aca431e6ea0a50ffd7201551175b47323/pycryptodome-3.23.0-cp37-abi3-macosx_10_9_x86_64.whl", hash = "sha256:cfb5cd445280c5b0a4e6187a7ce8de5a07b5f3f897f235caa11f1f435f182843", size = 1640362, upload-time = "2025-05-17T17:20:50.392Z" },
+    { url = "https://files.pythonhosted.org/packages/50/52/adaf4c8c100a8c49d2bd058e5b551f73dfd8cb89eb4911e25a0c469b6b4e/pycryptodome-3.23.0-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:67bd81fcbe34f43ad9422ee8fd4843c8e7198dd88dd3d40e6de42ee65fbe1490", size = 2182625, upload-time = "2025-05-17T17:20:52.866Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/e9/a09476d436d0ff1402ac3867d933c61805ec2326c6ea557aeeac3825604e/pycryptodome-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c8987bd3307a39bc03df5c8e0e3d8be0c4c3518b7f044b0f4c15d1aa78f52575", size = 2268954, upload-time = "2025-05-17T17:20:55.027Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/c5/ffe6474e0c551d54cab931918127c46d70cab8f114e0c2b5a3c071c2f484/pycryptodome-3.23.0-cp37-abi3-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:aa0698f65e5b570426fc31b8162ed4603b0c2841cbb9088e2b01641e3065915b", size = 2308534, upload-time = "2025-05-17T17:20:57.279Z" },
+    { url = "https://files.pythonhosted.org/packages/18/28/e199677fc15ecf43010f2463fde4c1a53015d1fe95fb03bca2890836603a/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:53ecbafc2b55353edcebd64bf5da94a2a2cdf5090a6915bcca6eca6cc452585a", size = 2181853, upload-time = "2025-05-17T17:20:59.322Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/ea/4fdb09f2165ce1365c9eaefef36625583371ee514db58dc9b65d3a255c4c/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_i686.whl", hash = "sha256:156df9667ad9f2ad26255926524e1c136d6664b741547deb0a86a9acf5ea631f", size = 2342465, upload-time = "2025-05-17T17:21:03.83Z" },
+    { url = "https://files.pythonhosted.org/packages/22/82/6edc3fc42fe9284aead511394bac167693fb2b0e0395b28b8bedaa07ef04/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:dea827b4d55ee390dc89b2afe5927d4308a8b538ae91d9c6f7a5090f397af1aa", size = 2267414, upload-time = "2025-05-17T17:21:06.72Z" },
+    { url = "https://files.pythonhosted.org/packages/59/fe/aae679b64363eb78326c7fdc9d06ec3de18bac68be4b612fc1fe8902693c/pycryptodome-3.23.0-cp37-abi3-win32.whl", hash = "sha256:507dbead45474b62b2bbe318eb1c4c8ee641077532067fec9c1aa82c31f84886", size = 1768484, upload-time = "2025-05-17T17:21:08.535Z" },
+    { url = "https://files.pythonhosted.org/packages/54/2f/e97a1b8294db0daaa87012c24a7bb714147c7ade7656973fd6c736b484ff/pycryptodome-3.23.0-cp37-abi3-win_amd64.whl", hash = "sha256:c75b52aacc6c0c260f204cbdd834f76edc9fb0d8e0da9fbf8352ef58202564e2", size = 1799636, upload-time = "2025-05-17T17:21:10.393Z" },
+    { url = "https://files.pythonhosted.org/packages/18/3d/f9441a0d798bf2b1e645adc3265e55706aead1255ccdad3856dbdcffec14/pycryptodome-3.23.0-cp37-abi3-win_arm64.whl", hash = "sha256:11eeeb6917903876f134b56ba11abe95c0b0fd5e3330def218083c7d98bbcb3c", size = 1703675, upload-time = "2025-05-17T17:21:13.146Z" },
+]
+
+[[package]]
+name = "pygments"
+version = "2.21.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/49/2e/ced460408999b33da6b31b0021b0f37d329e202d4169aeb164493778f25b/pygments-2.21.0.tar.gz", hash = "sha256:610ca751c9bc2492b38eb9a38a7fbc93edbbb2d7182edaf34e66ae493dee5c8c", size = 5005329, upload-time = "2026-08-17T08:02:48.824Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/71/46/17f022dd3e953bf20a04a028a21ec746d942f8d2af30fa0f124fa0e6a684/pygments-2.21.0-py3-none-any.whl", hash = "sha256:2363c69b61c4a97c838da3b130dcd6468f4848992b21a82f2a63ec34377137d9", size = 1250147, upload-time = "2026-08-17T08:02:44.912Z" },
+]
+
+[[package]]
+name = "pytest"
+version = "9.1.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "iniconfig" },
+    { name = "packaging" },
+    { name = "pluggy" },
+    { name = "pygments" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e4/47/b9efed96c114afcfa3c9d3fe98a76a1d14c74a9e266d397cf6eb64be5e01/pytest-9.1.1.tar.gz", hash = "sha256:1088fbde8f2b49d95a549a195707afa7a76a3ce9bcadc26b6d71f0ffda5fe313", size = 1636369, upload-time = "2026-06-19T10:58:32.857Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/24/25/1de2678b631f5a49215c6c96fff41ba892b0a34df68d6d80292b1b48aa7f/pytest-9.1.1-py3-none-any.whl", hash = "sha256:37a86b45efb9a47a61a36449063e8e18d0cab3161329fc099eb21783169c4f0c", size = 386536, upload-time = "2026-06-19T10:58:31.347Z" },
+]
+
+[[package]]
+name = "ruff"
+version = "0.16.9"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/96/bf/c935ca98e73fe8ce65b87ef08a280c0c1e85295d569228c15e87d8fdfaf1/ruff-0.16.9.tar.gz", hash = "sha256:12b625c6cfba78d285d9f48eda5f053374f1e53cb10ef17342a383750db99161", size = 4948764, upload-time = "2026-09-24T20:37:49.416Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0d/26/df51322b52ee1ada7eff2d071ea09d11d5c2d1dcc9f02594f5785c4d1635/ruff-0.16.9-py3-none-linux_armv6l.whl", hash = "sha256:95e6f022090368ab3b824c36276839c53b2adf1a3f4c09fefc33dfc400f6da96", size = 10082922, upload-time = "2026-09-24T20:37:13.045Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/27/7bf51f5a7aa375e9f339280a303aab44ca75dc1525f1cdc5991761685b0f/ruff-0.16.9-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:a5f27be168556594a86d2f415db0cf43f5291917849318f873c7e2791f7a8c67", size = 10236360, upload-time = "2026-09-24T20:37:16.049Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/63/09659283f92f02dff45809da194a70da2f688d87c55d8875c4fae3536072/ruff-0.16.9-py3-none-macosx_11_0_arm64.whl", hash = "sha256:1632eb1d6197f33bd00b1acbc5b71009e89a8895c158e2d2b03a834fac964ab6", size = 9892940, upload-time = "2026-09-24T20:37:17.957Z" },
+    { url = "https://files.pythonhosted.org/packages/24/58/98de1b72ec172f5f8f1731236fe21585b3998bf7dfe9fcc44ae9ba626012/ruff-0.16.9-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b3f951b14d865d5952c89d40a5ca07e87abe24fa5453299878411e127748fb1c", size = 10032114, upload-time = "2026-09-24T20:37:19.942Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/7d/f1e17c54ab59d4bad1dce8ee3e22a7a1d0ef4745240decacdcf3832b5bb2/ruff-0.16.9-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:447fc07e1573afff7cb02803462b12b6c8ece7cf10e2cd78565fa6d7a1c0bf8d", size = 9910227, upload-time = "2026-09-24T20:37:21.872Z" },
+    { url = "https://files.pythonhosted.org/packages/34/19/436f647a65075bbd3bab2668b3bdaa5120559b294694018cdcefabbbf30b/ruff-0.16.9-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8a3e039a6a40ed976c491722b60e0ae4a4aa1a86057f540ee7a37a5d19ae9120", size = 10547484, upload-time = "2026-09-24T20:37:24.229Z" },
+    { url = "https://files.pythonhosted.org/packages/03/59/38430a6bc2f6d8095447ac39625cf8b6e9344a47e6c26225c8ba1bff3ffb/ruff-0.16.9-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4684dded7db60aa57cb118fa158630f5feade4af5782903b6053484bdf9bd129", size = 11412367, upload-time = "2026-09-24T20:37:26.307Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/bd/bbb6d7fc7f208c8b8c50dd5dc8206e4cfdb1e7a8fb852606a3adf370c880/ruff-0.16.9-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d29c934357e45642fda2f34c0b1f4025b4a6c01e15e4bf0016879d60078a142c", size = 10869787, upload-time = "2026-09-24T20:37:28.35Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/b8/9c543074918061abbefc3bd139bee22de35abedb00dde0f2d27288838962/ruff-0.16.9-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a21713e629d3e5bdb2f5c2def1cc7f04f47fa8e1a7eb0571b4a28e1da64bc728", size = 10406494, upload-time = "2026-09-24T20:37:30.624Z" },
+    { url = "https://files.pythonhosted.org/packages/35/7a/5a8851bd146e7ccf8fd4b003f6c75c11f8fbdb0e60673b45097986b6bf41/ruff-0.16.9-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:7baa24ef5fc8e77aa93879e1d3f43754a01ae488e869f1ae30cf431afd4d2452", size = 10590083, upload-time = "2026-09-24T20:37:32.439Z" },
+    { url = "https://files.pythonhosted.org/packages/87/f0/4c3467188f23f806960b46fa76575a7cd0514c9ba90562650b71efc96980/ruff-0.16.9-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:a41aac6230aadfaa133bdfa1614488531ffa3e0837567ae04c0da2058a9c0f9e", size = 10119151, upload-time = "2026-09-24T20:37:34.581Z" },
+    { url = "https://files.pythonhosted.org/packages/15/34/5a4def5adea572ce6aea0bb64f21f928ee317b80e0db747d7979b01d7261/ruff-0.16.9-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:c2529fb5896d49115b0e9aa8f887490b34bbe76baf879ec2264ac59406869ce7", size = 9911544, upload-time = "2026-09-24T20:37:36.796Z" },
+    { url = "https://files.pythonhosted.org/packages/62/5d/d15ebea7499eef9373318c0ee6ca127832927c6529731f6d48e18dce7ca9/ruff-0.16.9-py3-none-musllinux_1_2_i686.whl", hash = "sha256:41e3870277694177429b56406d65dfbdb2c2802c52b715edaf6a0b829c69d4ee", size = 10269884, upload-time = "2026-09-24T20:37:38.857Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/56/c5d3cd119ded7a3c7aba0e961b69cb3df701c91662c98ad694467d060ce1/ruff-0.16.9-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:8adbe4e58af167f767d7b2ba5e83c42e878350796cf78c2f5e14ab9903a92588", size = 10749366, upload-time = "2026-09-24T20:37:41.042Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/fe/734ec7527029ac757ecf821f143c9f3fcf69149c21f53a044a899b430f5a/ruff-0.16.9-py3-none-win32.whl", hash = "sha256:0e1dbc2073624dee6618d41d0098690a7244654af746704b64759e12b6b6b385", size = 10152355, upload-time = "2026-09-24T20:37:43.025Z" },
+    { url = "https://files.pythonhosted.org/packages/14/21/26e4643629b3ebb44f0a06f9c9a53058d63d989415f63a9a3c28e2ee7f22/ruff-0.16.9-py3-none-win_amd64.whl", hash = "sha256:6bd40fec8cd4c8a3d4dd589bd8ad4e6320c13c29234159bfd959a40d529d597b", size = 10592965, upload-time = "2026-09-24T20:37:44.944Z" },
+    { url = "https://files.pythonhosted.org/packages/51/60/5fb1a39dbb5ae314d5f59bc7348a63c1d5c20f3cd83914c4b5cb0be31d2d/ruff-0.16.9-py3-none-win_arm64.whl", hash = "sha256:ed1a252039200f57a59eebc063b54beabea67bfbaaca0eeaa7f54b5fbcda2284", size = 10458649, upload-time = "2026-09-24T20:37:46.882Z" },
+]
+
+[[package]]
+name = "snrc-resolve"
+version = "0.1.0"
+source = { virtual = "." }
+dependencies = [
+    { name = "eth-hash", extra = ["pycryptodome"] },
+]
+
+[package.dev-dependencies]
+dev = [
+    { name = "pytest" },
+    { name = "ruff" },
+]
+
+[package.metadata]
+requires-dist = [{ name = "eth-hash", extras = ["pycryptodome"], specifier = ">=0.7" }]
+
+[package.metadata.requires-dev]
+dev = [
+    { name = "pytest", specifier = ">=8" },
+    { name = "ruff", specifier = ">=0.6" },
+]

simplexmq.cabal

@@ -1,7 +1,7 @@
 cabal-version: 3.0
 
 name:           simplexmq
-version:        7.1.0.8
+version:        7.1.0.11
 synopsis:       SimpleXMQ message broker
 description:    This package includes <./docs/Simplex-Messaging-Server.html server>,
                 <./docs/Simplex-Messaging-Client.html client> and
@@ -194,6 +194,9 @@ library
           Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260411_service_certs
           Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260712_address_dr_rpc
           Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260823_snd_files_entitlement
+          Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260919_ratchet_verify_codes
+          Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260929_ratchet_indexes
+          Simplex.Messaging.Agent.Store.Postgres.Migrations.M20261008_rcv_switch_status
   else
       exposed-modules:
           Simplex.Messaging.Agent.Store.SQLite
@@ -248,6 +251,10 @@ library
           Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260411_service_certs
           Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260712_address_dr_rpc
           Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260823_snd_files_entitlement
+          Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260919_ratchet_verify_codes
+          Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260929_ratchet_indexes
+          Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261005_commands_autoincrement
+          Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261008_rcv_switch_status
           Simplex.Messaging.Agent.Store.SQLite.Util
   if flag(client_postgres) || flag(server_postgres)
       exposed-modules:
@@ -595,7 +602,7 @@ test-suite simplexmq-test
   default-extensions:
       StrictData
   -- add -fhpc to ghc-options to run tests with coverage
-  ghc-options: -Weverything -Wno-missing-exported-signatures -Wno-missing-import-lists -Wno-missed-specialisations -Wno-all-missed-specialisations -Wno-unsafe -Wno-safe -Wno-missing-local-signatures -Wno-missing-kind-signatures -Wno-missing-deriving-strategies -Wno-monomorphism-restriction -Wno-prepositive-qualified-module -Wno-implicit-prelude -Wno-missing-safe-haskell-mode -Wno-missing-export-lists -Wno-partial-fields -Wcompat -Werror=incomplete-record-updates -Werror=incomplete-patterns -Werror=incomplete-uni-patterns -Werror=missing-methods -Werror=tabs -Wredundant-constraints -Wincomplete-record-updates -Wunused-type-patterns -O2 -threaded -rtsopts -with-rtsopts=-A64M -with-rtsopts=-N1
+  ghc-options: -Weverything -Wno-missing-exported-signatures -Wno-missing-import-lists -Wno-missed-specialisations -Wno-all-missed-specialisations -Wno-unsafe -Wno-safe -Wno-missing-local-signatures -Wno-missing-kind-signatures -Wno-missing-deriving-strategies -Wno-monomorphism-restriction -Wno-prepositive-qualified-module -Wno-implicit-prelude -Wno-missing-safe-haskell-mode -Wno-missing-export-lists -Wno-partial-fields -Wcompat -Werror=incomplete-record-updates -Werror=incomplete-patterns -Werror=incomplete-uni-patterns -Werror=missing-methods -Werror=tabs -Wredundant-constraints -Wincomplete-record-updates -Wunused-type-patterns -O2 -threaded -rtsopts "-with-rtsopts=-A64M -N1"
   build-depends:
       base
     , aeson
@@ -633,7 +640,6 @@ test-suite simplexmq-test
     , stm
     , text
     , time
-    , timeit ==2.0.*
     , tls >=1.9.0 && <1.10
     , transformers
     , unliftio

src/Simplex/FileTransfer/Agent.hs

@@ -17,11 +17,16 @@ module Simplex.FileTransfer.Agent
     toFSFilePath,
     -- Receiving files
     xftpReceiveFile',
+    xftpPrepareReceiveFile',
+    xftpStartReceiveFile',
     xftpDeleteRcvFile',
     xftpDeleteRcvFiles',
     -- Sending files
     xftpSendFile',
     xftpSendDescription',
+    xftpPrepareSendFile',
+    xftpPrepareSendDescription',
+    xftpStartSendFile',
     deleteSndFileInternal,
     deleteSndFilesInternal,
     deleteSndFileRemote,
@@ -127,7 +132,12 @@ closeXFTPAgent a = do
     stopWorkers workers = atomically (swapTVar workers M.empty) >>= mapM_ (liftIO . cancelWorker)
 
 xftpReceiveFile' :: AgentClient -> UserId -> ValidFileDescription 'FRecipient -> Maybe CryptoFileArgs -> Bool -> AM RcvFileId
-xftpReceiveFile' c userId (ValidFileDescription fd@FileDescription {chunks, redirect}) cfArgs approvedRelays = do
+xftpReceiveFile' c userId vfd cfArgs approvedRelays = do
+  fId <- xftpPrepareReceiveFile' c userId vfd cfArgs approvedRelays
+  fId <$ xftpStartReceiveFile' c fId
+
+xftpPrepareReceiveFile' :: AgentClient -> UserId -> ValidFileDescription 'FRecipient -> Maybe CryptoFileArgs -> Bool -> AM RcvFileId
+xftpPrepareReceiveFile' c userId (ValidFileDescription fd@FileDescription {redirect}) cfArgs approvedRelays = do
   g <- asks random
   prefixPath <- lift $ getPrefixPath "rcv.xftp"
   createDirectory prefixPath
@@ -137,7 +147,7 @@ xftpReceiveFile' c userId (ValidFileDescription fd@FileDescription {chunks, redi
   lift $ createDirectory =<< toFSFilePath relTmpPath
   lift $ createEmptyFile =<< toFSFilePath relSavePath
   let saveFile = CryptoFile relSavePath cfArgs
-  fId <- case redirect of
+  case redirect of
     Nothing -> withStore c $ \db -> createRcvFile db g userId fd relPrefixPath relTmpPath saveFile approvedRelays
     Just _ -> do
       -- prepare description paths
@@ -149,8 +159,11 @@ xftpReceiveFile' c userId (ValidFileDescription fd@FileDescription {chunks, redi
       let saveFileRedirect = CryptoFile relSavePathRedirect $ Just cfArgsRedirect
       -- create download tasks
       withStore c $ \db -> createRcvFileRedirect db g userId fd relPrefixPath relTmpPathRedirect saveFileRedirect relTmpPath saveFile approvedRelays
-  forM_ chunks (downloadChunk c)
-  pure fId
+
+xftpStartReceiveFile' :: AgentClient -> RcvFileId -> AM ()
+xftpStartReceiveFile' c rcvFileEntityId = do
+  srvs <- withStore c (`startPreparedRcvFile` rcvFileEntityId)
+  lift $ forM_ srvs $ void . getXFTPRcvWorker True c . Just
 
 downloadChunk :: AgentClient -> FileChunk -> AM ()
 downloadChunk c FileChunk {replicas = (FileChunkReplica {server} : _)} = do
@@ -353,6 +366,11 @@ notify c entId cmd = atomically $ writeTBQueue (subQ c) ("", entId, AEvt (sAEnti
 
 xftpSendFile' :: AgentClient -> UserId -> CryptoFile -> Int -> Maybe Word32 -> AM SndFileId
 xftpSendFile' c userId file numRecipients storageHours = do
+  fId <- xftpPrepareSendFile' c userId file numRecipients storageHours
+  fId <$ xftpStartSendFile' c fId
+
+xftpPrepareSendFile' :: AgentClient -> UserId -> CryptoFile -> Int -> Maybe Word32 -> AM SndFileId
+xftpPrepareSendFile' c userId file numRecipients storageHours = do
   g <- asks random
   prefixPath <- lift $ getPrefixPath "snd.xftp"
   createDirectory prefixPath
@@ -360,12 +378,15 @@ xftpSendFile' c userId file numRecipients storageHours = do
   key <- atomically $ C.randomSbKey g
   nonce <- atomically $ C.randomCbNonce g
   -- saving absolute filePath will not allow to restore file encryption after app update, but it's a short window
-  fId <- withStore c $ \db -> createSndFile db g userId file numRecipients relPrefixPath key nonce Nothing storageHours
-  lift . void $ getXFTPSndWorker True c Nothing
-  pure fId
+  withStore c $ \db -> createSndFile db g userId file numRecipients relPrefixPath key nonce Nothing storageHours
 
 xftpSendDescription' :: AgentClient -> UserId -> ValidFileDescription 'FRecipient -> Int -> AM SndFileId
-xftpSendDescription' c userId (ValidFileDescription fdDirect@FileDescription {size, digest}) numRecipients = do
+xftpSendDescription' c userId vfd numRecipients = do
+  fId <- xftpPrepareSendDescription' c userId vfd numRecipients
+  fId <$ xftpStartSendFile' c fId
+
+xftpPrepareSendDescription' :: AgentClient -> UserId -> ValidFileDescription 'FRecipient -> Int -> AM SndFileId
+xftpPrepareSendDescription' c userId (ValidFileDescription fdDirect@FileDescription {size, digest}) numRecipients = do
   g <- asks random
   prefixPath <- lift $ getPrefixPath "snd.xftp"
   createDirectory prefixPath
@@ -376,9 +397,12 @@ xftpSendDescription' c userId (ValidFileDescription fdDirect@FileDescription {si
   liftError (FILE . FILE_IO . show) $ CF.writeFile file (LB.fromStrict $ strEncode fdDirect)
   key <- atomically $ C.randomSbKey g
   nonce <- atomically $ C.randomCbNonce g
-  fId <- withStore c $ \db -> createSndFile db g userId file numRecipients relPrefixPath key nonce (Just RedirectFileInfo {size, digest}) Nothing
+  withStore c $ \db -> createSndFile db g userId file numRecipients relPrefixPath key nonce (Just RedirectFileInfo {size, digest}) Nothing
+
+xftpStartSendFile' :: AgentClient -> SndFileId -> AM ()
+xftpStartSendFile' c sndFileEntityId = do
+  withStore c (`startPreparedSndFile` sndFileEntityId)
   lift . void $ getXFTPSndWorker True c Nothing
-  pure fId
 
 resumeXFTPSndWork :: AgentClient -> Maybe XFTPServer -> AM' ()
 resumeXFTPSndWork = void .: getXFTPSndWorker False
@@ -544,9 +568,10 @@ runXFTPSndWorker c srv Worker {doWork} = do
       notify c sndFileEntityId $ SFPROG uploaded total
       when complete $ do
         (sndDescr, rcvDescrs) <- sndFileToDescrs sf
+        flip catchAllErrors (\e -> logError $ "XFTP snd worker error: " <> tshow e) $ do
+          lift . forM_ prefixPath $ removePath <=< toFSFilePath
+          withStore' c $ \db -> updateSndFileComplete db sndFileId
         notify c sndFileEntityId $ SFDONE sndDescr rcvDescrs (sndFileExpiresAt chunks)
-        lift . forM_ prefixPath $ removePath <=< toFSFilePath
-        withStore' c $ \db -> updateSndFileComplete db sndFileId
       where
         addRecipients :: SndFileChunk -> SndFileChunkReplica -> AM SndFileChunkReplica
         addRecipients ch@SndFileChunk {numRecipients} cr@SndFileChunkReplica {sndChunkReplicaId, rcvIdsKeys}

src/Simplex/FileTransfer/Protocol.hs

@@ -80,7 +80,7 @@ import Simplex.Messaging.Protocol
     tParse,
   )
 import Simplex.Messaging.Transport (THandleParams (..), TransportError (..), TransportPeer (..))
-import Simplex.Messaging.Util ((<$?>))
+import Simplex.Messaging.Util (eitherToMaybe, (<$?>))
 
 xftpBlockSize :: Int
 xftpBlockSize = 16384
@@ -383,7 +383,7 @@ checkParty' c = case testEquality (sFileParty @p) (sFileParty @p') of
 xftpEncodeAuthTransmission :: ProtocolEncoding XFTPVersion XFTPErrorType c => THandleParams XFTPVersion 'TClient -> C.APrivateAuthKey -> Transmission c -> Either TransportError ByteString
 xftpEncodeAuthTransmission thParams@THandleParams {thAuth} pKey t@(corrId, _, _) = do
   let TransmissionForAuth {tForAuth, tToSend} = encodeTransmissionForAuth thParams t
-  xftpEncodeBatch1 . (,tToSend) =<< authTransmission thAuth False (Just pKey) (C.cbNonce $ bs corrId) tForAuth
+  xftpEncodeBatch1 . (,tToSend) =<< authTransmission thAuth False (Just pKey) (eitherToMaybe $ C.cbNonce $ bs corrId) tForAuth
 
 xftpEncodeTransmission :: ProtocolEncoding XFTPVersion XFTPErrorType c => THandleParams XFTPVersion p -> Transmission c -> Either TransportError ByteString
 xftpEncodeTransmission thParams t = xftpEncodeBatch1 (Nothing, encodeTransmission thParams t)

src/Simplex/FileTransfer/Server.hs

@@ -25,7 +25,6 @@ import Control.Monad.Reader
 import Control.Monad.Trans.Except
 import qualified Data.Aeson as J
 import Data.Bifunctor (first)
-import qualified Data.ByteString.Base64.URL as B64
 import Data.ByteString.Builder (Builder, byteString)
 import Data.ByteString.Char8 (ByteString)
 import qualified Data.ByteString.Char8 as B
@@ -594,7 +593,7 @@ processXFTPRequest ent HTTP2Body {bodyPart} = \case
               \used -> let used' = used + fromIntegral size in if used' <= quota then (True, used') else (False, used)
           receive = do
             path <- asks $ filesPath . config
-            let fPath = path </> B.unpack (B64.encode $ unEntityId senderId)
+            let fPath = path </> serverFileName senderId
             receiveChunk (XFTPRcvChunkSpec fPath size digest) >>= \case
               Right () -> do
                 stats <- asks serverStats

src/Simplex/FileTransfer/Server/StoreLog.hs

@@ -17,6 +17,7 @@ module Simplex.FileTransfer.Server.StoreLog
     logDeleteFile,
     logBlockFile,
     logAckFile,
+    serverFileName,
   )
 where
 
@@ -24,6 +25,7 @@ import Control.Applicative (optional, (<|>))
 import Control.Concurrent.STM
 import Control.Monad.Except
 import qualified Data.Attoparsec.ByteString.Char8 as A
+import qualified Data.ByteString.Base64.URL as B64
 import qualified Data.ByteString.Char8 as B
 import qualified Data.ByteString.Lazy.Char8 as LB
 import Data.Composition ((.:), (.::.))
@@ -35,10 +37,11 @@ import Simplex.FileTransfer.Protocol (FileInfo (..))
 import Simplex.FileTransfer.Server.Store
 import Simplex.FileTransfer.Transport (XFTPErrorType (..))
 import Simplex.Messaging.Encoding.String
-import Simplex.Messaging.Protocol (BlockingInfo, RcvPublicAuthKey, RecipientId, SenderId)
+import Simplex.Messaging.Protocol (BlockingInfo, EntityId (..), RcvPublicAuthKey, RecipientId, SenderId)
 import Simplex.Messaging.Server.QueueStore (ServerEntityStatus (..))
 import Simplex.Messaging.Server.StoreLog
 import Simplex.Messaging.Util (bshow)
+import System.FilePath (takeFileName)
 import System.IO
 
 data FileStoreLogRecord
@@ -99,6 +102,9 @@ logBlockFile s fId = logFileStoreRecord s . BlockFile fId
 logAckFile :: StoreLog 'WriteMode -> RecipientId -> IO ()
 logAckFile s = logFileStoreRecord s . AckFile
 
+serverFileName :: SenderId -> FilePath
+serverFileName = B.unpack . B64.encode . unEntityId
+
 readWriteFileStore :: FilePath -> STMFileStore -> IO (StoreLog 'WriteMode)
 readWriteFileStore = readWriteStoreLog readFileStore writeFileStore
 
@@ -115,7 +121,9 @@ readFileStore f st = mapM_ (addFileLogRecord . LB.toStrict) . LB.lines =<< LB.re
       AddFile sId file createdAt expiresAt status
         | size file > 0 -> addFile st sId file createdAt expiresAt status
         | otherwise -> pure $ Left SIZE
-      PutFile qId path -> setFilePath st qId path
+      PutFile qId path
+        | takeFileName path == serverFileName qId -> setFilePath st qId path
+        | otherwise -> pure $ Left AUTH
       AddRecipients sId rcps -> runExceptT $ addRecipients sId rcps
       DeleteFile sId -> deleteFile st sId
       BlockFile sId info -> blockFile st sId info True

src/Simplex/FileTransfer/Types.hs

@@ -93,7 +93,8 @@ data RcvFile = RcvFile
   deriving (Show)
 
 data RcvFileStatus
-  = RFSReceiving
+  = RFSPrepared
+  | RFSReceiving
   | RFSReceived
   | RFSDecrypting
   | RFSComplete
@@ -106,6 +107,7 @@ instance ToField RcvFileStatus where toField = toField . textEncode
 
 instance TextEncoding RcvFileStatus where
   textDecode = \case
+    "prepared" -> Just RFSPrepared
     "receiving" -> Just RFSReceiving
     "received" -> Just RFSReceived
     "decrypting" -> Just RFSDecrypting
@@ -113,6 +115,7 @@ instance TextEncoding RcvFileStatus where
     "error" -> Just RFSError
     _ -> Nothing
   textEncode = \case
+    RFSPrepared -> "prepared"
     RFSReceiving -> "receiving"
     RFSReceived -> "received"
     RFSDecrypting -> "decrypting"
@@ -177,7 +180,8 @@ sndFileEncPath :: FilePath -> FilePath
 sndFileEncPath prefixPath = prefixPath </> "xftp.encrypted"
 
 data SndFileStatus
-  = SFSNew -- db record created
+  = SFSPrepared
+  | SFSNew -- db record created
   | SFSEncrypting -- encryption started
   | SFSEncrypted -- encryption complete
   | SFSUploading -- all chunk replicas are created on servers
@@ -191,6 +195,7 @@ instance ToField SndFileStatus where toField = toField . textEncode
 
 instance TextEncoding SndFileStatus where
   textDecode = \case
+    "prepared" -> Just SFSPrepared
     "new" -> Just SFSNew
     "encrypting" -> Just SFSEncrypting
     "encrypted" -> Just SFSEncrypted
@@ -199,6 +204,7 @@ instance TextEncoding SndFileStatus where
     "error" -> Just SFSError
     _ -> Nothing
   textEncode = \case
+    SFSPrepared -> "prepared"
     SFSNew -> "new"
     SFSEncrypting -> "encrypting"
     SFSEncrypted -> "encrypted"

src/Simplex/Messaging/Agent/Client.hs

@@ -312,7 +312,7 @@ import Simplex.Messaging.Session
 import Simplex.Messaging.SystemTime
 import Simplex.Messaging.TMap (TMap)
 import qualified Simplex.Messaging.TMap as TM
-import Simplex.Messaging.Transport (HandshakeError (..), SMPServiceRole (..), SMPVersion, ServiceCredentials (..), SessionId, THClientService' (..), THandleAuth (..), THandleParams (sessionId, thAuth, thVersion, serverInfo), TransportError (..), TransportPeer (..), shortLinksSMPVersion, newNtfCredsSMPVersion)
+import Simplex.Messaging.Transport (HandshakeError (..), SMPServiceRole (..), SMPVersion, ServiceCredentials (..), SessionId, THClientService' (..), THandleAuth (..), THandleParams (sessionId, thAuth, thVersion, serverInfo), TransportError (..), newNtfCredsSMPVersion)
 import Simplex.Messaging.Transport.Client (TransportHost (..))
 import Simplex.Messaging.Transport.Credentials
 import Simplex.Messaging.Util
@@ -1261,9 +1261,12 @@ sendOrProxySMPCommand c nm userId destSrv@ProtocolServer {host = destHosts} conn
           Left e -> throwE e
 
 ipAddressProtected :: NetworkConfig -> ProtocolServer p -> Bool
-ipAddressProtected NetworkConfig {socksProxy, hostMode} (ProtocolServer _ hosts _ _) = do
-  isJust socksProxy || (hostMode == HMOnion && any isOnionHost hosts)
+ipAddressProtected NetworkConfig {socksProxy, socksMode, hostMode} (ProtocolServer _ hosts _ _)
+  | isJust socksProxy = socksMode == SMAlways || if hostMode == HMPublic then allOnion else anyOnion
+  | otherwise = hostMode == HMOnion && anyOnion
   where
+    anyOnion = any isOnionHost hosts
+    allOnion = all isOnionHost hosts
     isOnionHost = \case THOnionHost _ -> True; _ -> False
 
 withNtfClient :: AgentClient -> NetworkRequestMode -> NtfServer -> EntityId -> ByteString -> (NtfClient -> ExceptT NtfClientError IO a) -> AM a
@@ -1496,7 +1499,7 @@ newRcvQueue_ c nm userId connId (ProtoServerWithAuth srv auth) vRange cqrd enabl
   let sessServiceId = (\THClientService {serviceId = sId} -> sId) <$> (clientService =<< thAuth thParams')
   when (isJust serviceId && serviceId /= sessServiceId) $ logError "incorrect service ID in NEW response"
   liftIO . logServer "<--" c srv NoEntity $ B.unwords ["IDS", logSecret rcvId, logSecret sndId]
-  shortLink <- mkShortLinkCreds thParams' qik
+  shortLink <- mkShortLinkCreds qik
   let rq =
         RcvQueue
           { userId,
@@ -1537,8 +1540,8 @@ newRcvQueue_ c nm userId connId (ProtoServerWithAuth srv auth) vRange cqrd enabl
       (Just ((ntfPublicKey, ntfPrivateKey), dhpk), Just (ServerNtfCreds notifierId dhk')) ->
         Just ClientNtfCreds {ntfPublicKey, ntfPrivateKey, notifierId, rcvNtfDhSecret = C.dh' dhk' dhpk}
       _ -> Nothing
-    mkShortLinkCreds :: THandleParams SMPVersion 'TClient -> QueueIdsKeys -> AM (Maybe ShortLinkCreds)
-    mkShortLinkCreds thParams' QIK {sndId, queueMode, linkId} = case (cqrd, queueMode) of
+    mkShortLinkCreds :: QueueIdsKeys -> AM (Maybe ShortLinkCreds)
+    mkShortLinkCreds QIK {sndId, queueMode, linkId} = case (cqrd, queueMode) of
       (CQRMessaging ld, Just QMMessaging) ->
         withLinkData ld $ \lnkId CQRData {linkKey, privSigKey, srvReq = (sndId', d)} ->
           if sndId == sndId'
@@ -1552,11 +1555,9 @@ newRcvQueue_ c nm userId connId (ProtoServerWithAuth srv auth) vRange cqrd enabl
       (_, Nothing) -> newErr "unexpected link ID"
       _ -> newErr "unexpected queue mode"
       where
-        v = thVersion thParams'
         withLinkData :: Maybe d -> (SMP.LinkId -> d -> AM (Maybe ShortLinkCreds)) -> AM (Maybe ShortLinkCreds)
         withLinkData ld_ mkLink = case (ld_, linkId) of
           (Just ld, Just lnkId) -> mkLink lnkId ld
-          (Just _, Nothing) | v < shortLinksSMPVersion -> pure Nothing
           (Nothing, Nothing) -> pure Nothing
           _ -> newErr "unexpected or absent link ID"
         newErr :: String -> AM (Maybe ShortLinkCreds)
@@ -1981,7 +1982,7 @@ decryptSMPMessage :: RcvQueue -> SMP.RcvMessage -> AM SMP.ClientRcvMsgBody
 decryptSMPMessage rq SMP.RcvMessage {msgId, msgBody = SMP.EncRcvMsgBody body} =
   liftEither $ parse SMP.clientRcvMsgBodyP (AGENT $ A_MESSAGE "decrypt message") =<< decrypt body
   where
-    decrypt = agentCbDecrypt (rcvDhSecret rq) (C.cbNonce msgId)
+    decrypt = agentCbDecrypt (rcvDhSecret rq) msgId
 
 secureQueue :: AgentClient -> NetworkRequestMode -> RcvQueue -> SndPublicAuthKey -> AM ()
 secureQueue c nm rq@RcvQueue {rcvId, rcvPrivateKey} senderKey =
@@ -2079,9 +2080,9 @@ disableQueuesNtfs c = sendTSessionBatches "NDEL" (mkSMPTSession . snd) disableQu
     queueCreds :: DisableQueueNtfReq -> (SMP.RecipientId, SMP.RcvPrivateAuthKey)
     queueCreds (_, RcvQueue {rcvPrivateKey, rcvId}) = (rcvId, rcvPrivateKey)
 
-sendAck :: AgentClient -> RcvQueue -> MsgId -> AM ()
+sendAck :: AgentClient -> RcvQueue -> MsgId -> AM QueueDrained
 sendAck c rq@RcvQueue {rcvId, rcvPrivateKey} msgId =
-  withSMPClient c NRMBackground rq ("ACK:" <> logSecret' msgId) $ \smp ->
+  withSMPClient c NRMBackground rq ("ACK:" <> logSecret' (SMP.unMsgId msgId)) $ \smp ->
     ackSMPMessage smp rcvPrivateKey rcvId msgId
 
 hasGetLock :: SomeRcvQueue q => AgentClient -> q -> IO Bool
@@ -2276,7 +2277,7 @@ agentCbEncryptOnce clientVersion dhRcvPubKey msg = do
 
 -- | NaCl crypto-box decrypt - both for messages received from the server
 -- and per-queue E2E encrypted messages from the sender that were inside.
-agentCbDecrypt :: C.DhSecretX25519 -> C.CbNonce -> ByteString -> Either AgentErrorType ByteString
+agentCbDecrypt :: C.CbNonceI n => C.DhSecretX25519 -> n -> ByteString -> Either AgentErrorType ByteString
 agentCbDecrypt dhSecret nonce msg =
   first cryptoError $
     C.cbDecrypt dhSecret nonce msg

src/Simplex/Messaging/Agent/Env/SQLite.hs

@@ -162,6 +162,7 @@ data AgentConfig = AgentConfig
     connDeleteDeliveryTimeout :: NominalDiffTime,
     helloTimeout :: NominalDiffTime,
     quotaExceededTimeout :: NominalDiffTime,
+    commandQuotaRetryInterval :: RetryInterval,
     persistErrorInterval :: NominalDiffTime,
     initialCleanupDelay :: Int64,
     cleanupInterval :: Int64,
@@ -241,6 +242,7 @@ defaultAgentConfig =
       connDeleteDeliveryTimeout = 2 * nominalDay,
       helloTimeout = 2 * nominalDay,
       quotaExceededTimeout = 7 * nominalDay,
+      commandQuotaRetryInterval = RetryInterval {initialInterval = 30_000000, increaseAfter = 0, maxInterval = 3600_000000},
       persistErrorInterval = 3, -- seconds
       initialCleanupDelay = 30 * 1000000, -- 30 seconds
       cleanupInterval = 5 * 60 * 1000000, -- 5 minutes

src/Simplex/Messaging/Agent/Protocol.hs

@@ -79,6 +79,7 @@ module Simplex.Messaging.Agent.Protocol
     SMPConfirmation (..),
     AgentMsgEnvelope (..),
     AgentMessage (..),
+    RatchetInfo (..),
     RequestSignature (..),
     AgentMessageType (..),
     APrivHeader (..),
@@ -166,6 +167,8 @@ module Simplex.Messaging.Agent.Protocol
     ConnId,
     ConfirmationId,
     InvitationId,
+    ConnVerifyCodes (..),
+    ContactRequestBinding (..),
     MsgIntegrity (..),
     MsgErrorType (..),
     QueueStatus (..),
@@ -243,7 +246,6 @@ import Simplex.Messaging.Crypto.Ratchet
     RcvE2ERatchetParams,
     RcvE2ERatchetParamsUri,
     SndE2ERatchetParams,
-    RcvE2EPrivRatchetParams,
     pattern PQSupportOff,
     pattern PQSupportOn,
   )
@@ -404,7 +406,7 @@ data AEvent (e :: AEntity) where
   LINK :: ConnShortLink 'CMContact -> UserConnLinkData 'CMContact -> AEvent AEConn
   LDATA :: FixedLinkData 'CMContact -> ConnLinkData 'CMContact -> ConnectionRequestUri 'CMContact -> AEvent AEConn
   CONF :: ConfirmationId -> PQSupport -> [SMPServer] -> ConnInfo -> AEvent AEConn -- ConnInfo is from sender, [SMPServer] will be empty only in v1 handshake
-  REQ :: InvitationId -> PQSupport -> NonEmpty SMPServer -> ConnInfo -> Bool -> AEvent AEConn -- ConnInfo is from sender; Bool - rejection reason can be sent
+  REQ :: InvitationId -> PQSupport -> NonEmpty SMPServer -> ConnInfo -> ContactRequestBinding -> Bool -> AEvent AEConn -- ConnInfo is from sender; Bool - rejection reason can be sent
   SREQ :: InvitationId -> Maybe C.PublicKeyEd25519 -> MsgBody -> AEvent AEConn
   SSENT :: AgentMsgId -> Maybe SMPServer -> AEvent AEConn
   RJCT :: ConnInfo -> AEvent AEConn
@@ -609,6 +611,7 @@ data RcvSwitchStatus
   | RSSendingQADD
   | RSSendingQUSE
   | RSReceivedMessage
+  | RSReceivedQEND
   deriving (Eq, Show)
 
 instance StrEncoding RcvSwitchStatus where
@@ -617,12 +620,14 @@ instance StrEncoding RcvSwitchStatus where
     RSSendingQADD -> "sending_qadd"
     RSSendingQUSE -> "sending_quse"
     RSReceivedMessage -> "received_message"
+    RSReceivedQEND -> "received_qend"
   strP =
     A.takeTill (== ' ') >>= \case
       "switch_started" -> pure RSSwitchStarted
       "sending_qadd" -> pure RSSendingQADD
       "sending_quse" -> pure RSSendingQUSE
       "received_message" -> pure RSReceivedMessage
+      "received_qend" -> pure RSReceivedQEND
       _ -> fail "bad RcvSwitchStatus"
 
 instance ToField RcvSwitchStatus where toField = toField . decodeLatin1 . strEncode
@@ -926,7 +931,7 @@ data AgentMessage
   | -- AgentConnInfoReply is used by accepting party in duplexHandshake mode (v2), allowing to include reply queue(s) in the initial confirmation.
     -- It made removed REPLY message unnecessary.
     AgentConnInfoReply (NonEmpty SMPQueueInfo) ConnInfo
-  | AgentRatchetInfo ByteString
+  | AgentRatchetInfo RatchetInfo
   | AgentMessage APrivHeader AMessage
   | AgentServiceRequest (NonEmpty SMPQueueInfo) (Maybe RequestSignature) MsgBody
   | AgentServiceResponse MsgBody
@@ -937,7 +942,7 @@ instance Encoding AgentMessage where
   smpEncode = \case
     AgentConnInfo cInfo -> smpEncode ('I', Tail cInfo)
     AgentConnInfoReply smpQueues cInfo -> smpEncode ('D', smpQueues, Tail cInfo) -- 'D' stands for "duplex"
-    AgentRatchetInfo info -> smpEncode ('R', Tail info)
+    AgentRatchetInfo info -> smpEncode ('R', info)
     AgentMessage hdr aMsg -> smpEncode ('M', hdr, aMsg)
     AgentServiceRequest qs sig_ body -> smpEncode ('A', qs, sig_, Tail body)
     AgentServiceResponse body -> smpEncode ('P', Tail body)
@@ -946,13 +951,23 @@ instance Encoding AgentMessage where
     smpP >>= \case
       'I' -> AgentConnInfo . unTail <$> smpP
       'D' -> AgentConnInfoReply <$> smpP <*> (unTail <$> smpP)
-      'R' -> AgentRatchetInfo . unTail <$> smpP
+      'R' -> AgentRatchetInfo <$> smpP
       'M' -> AgentMessage <$> smpP <*> smpP
       'A' -> AgentServiceRequest <$> smpP <*> smpP <*> (unTail <$> smpP)
       'P' -> AgentServiceResponse . unTail <$> smpP
       'J' -> AgentRejection . unTail <$> smpP
       _ -> fail "bad AgentMessage"
 
+data RatchetInfo = RatchetInfo {answeredKeyHash :: Maybe ByteString}
+  deriving (Eq, Show)
+
+instance Encoding RatchetInfo where
+  smpEncode RatchetInfo {answeredKeyHash} = smpEncode answeredKeyHash
+  smpP = do
+    answeredKeyHash <- smpP <|> pure Nothing
+    _ <- A.takeByteString
+    pure RatchetInfo {answeredKeyHash}
+
 -- internal type for storing message type in the database
 data AgentMessageType
   = AM_CONN_INFO
@@ -1372,6 +1387,15 @@ type ConfirmationId = ByteString
 
 type InvitationId = ByteString
 
+data ConnVerifyCodes = ConnVerifyCodes
+  { codeAD :: ByteString,
+    codePQ :: Maybe ByteString
+  }
+  deriving (Eq, Show)
+
+data ContactRequestBinding = CRBRatchet ConnVerifyCodes | CRBRequest ByteString
+  deriving (Eq, Show)
+
 extraSMPServerHosts :: Map TransportHost TransportHost
 extraSMPServerHosts =
   M.fromList
@@ -1602,7 +1626,7 @@ newtype LinkKey = LinkKey ByteString -- sha3-256(fixed_data)
 instance ToField LinkKey where toField (LinkKey s) = toField $ Binary s
 
 -- | Parameters for creating a connection with a prepared link.
-data PreparedLinkParams = PreparedLinkParams
+data PreparedLinkParams k = PreparedLinkParams
   { -- | Correlation ID / determines sender ID
     plpNonce :: C.CbNonce,
     -- | Queue E2EE DH key pair
@@ -1617,8 +1641,8 @@ data PreparedLinkParams = PreparedLinkParams
     plpSrvWithAuth :: SMPServerWithAuth,
     -- | Initial PQ keys
     plpInitKeys :: InitialKeys,
-    -- | Contact address double ratchet keys
-    plpAddressKeys :: Maybe (RatchetKeyId, RcvE2EPrivRatchetParams 'C.X448)
+    -- | Contact address double ratchet keys or invitation X3DH keys
+    plpRatchetKeys :: k
   }
 
 instance ConnectionModeI c => ToField (ConnectionLink c) where toField = toField . Binary . strEncode

src/Simplex/Messaging/Agent/RetryInterval.hs

@@ -10,6 +10,7 @@ module Simplex.Messaging.Agent.RetryInterval
     withRetryInterval,
     withRetryIntervalCount,
     withRetryForeground,
+    withRetryInterval2,
     withRetryLock2,
     updateRetryInterval2,
     nextRetryDelay,
@@ -86,9 +87,24 @@ withRetryForeground ri isForeground isOnline action = callAction 0 $ initialInte
                 | otherwise = (elapsed + delay, nextRetryDelay elapsed' delay ri)
           callAction elapsed' delay'
 
+withRetryInterval2 :: forall m. MonadIO m => RetryInterval2 -> (RI2State -> (RetryIntervalMode -> m ()) -> m ()) -> m ()
+withRetryInterval2 = withRetryWait2 $ liftIO . threadDelay'
+
 -- This function allows action to toggle between slow and fast retry intervals.
 withRetryLock2 :: forall m. MonadIO m => RetryInterval2 -> TMVar () -> (RI2State -> (RetryIntervalMode -> m ()) -> m ()) -> m ()
-withRetryLock2 RetryInterval2 {riSlow, riFast} lock action =
+withRetryLock2 ri lock = withRetryWait2 wait ri
+  where
+    wait delay = do
+      waiting <- newTVarIO True
+      _ <- liftIO . forkIO $ do
+        threadDelay' delay
+        atomically $ whenM (readTVar waiting) $ void $ tryPutTMVar lock ()
+      atomically $ do
+        takeTMVar lock
+        writeTVar waiting False
+
+withRetryWait2 :: forall m. Monad m => (Int64 -> m ()) -> RetryInterval2 -> (RI2State -> (RetryIntervalMode -> m ()) -> m ()) -> m ()
+withRetryWait2 wait RetryInterval2 {riSlow, riFast} action =
   callAction (0, initialInterval riSlow) (0, initialInterval riFast)
   where
     callAction :: (Int64, Int64) -> (Int64, Int64) -> m ()
@@ -102,14 +118,6 @@ withRetryLock2 RetryInterval2 {riSlow, riFast} lock action =
           let elapsed' = elapsed + delay
               delay' = nextRetryDelay elapsed' delay ri
           call (elapsed', delay')
-        wait delay = do
-          waiting <- newTVarIO True
-          _ <- liftIO . forkIO $ do
-            threadDelay' delay
-            atomically $ whenM (readTVar waiting) $ void $ tryPutTMVar lock ()
-          atomically $ do
-            takeTMVar lock
-            writeTVar waiting False
 
 nextRetryDelay :: Int64 -> Int64 -> RetryInterval -> Int64
 nextRetryDelay elapsed delay RetryInterval {increaseAfter, maxInterval} =

src/Simplex/Messaging/Agent/Store.hs

@@ -219,6 +219,7 @@ canAbortRcvSwitch ConnData {connAgentVersion} = maybe False canAbort . rcvSwchSt
       -- if switch is in RSReceivedMessage status, aborting switch (deleting new queue)
       -- will break the connection because the sender would have original queue deleted
       RSReceivedMessage -> False
+      RSReceivedQEND -> False
 
 data ClientNtfCreds = ClientNtfCreds
   { -- | key pair to be used by the notification server to authorize transmissions
@@ -489,7 +490,8 @@ data PendingCommand = PendingCommand
     corrId :: ACorrId,
     userId :: UserId,
     connId :: ConnId,
-    command :: AgentCommand
+    command :: AgentCommand,
+    createdAt :: UTCTime
   }
 
 data AgentCmdType = ACClient | ACInternal

src/Simplex/Messaging/Agent/Store/AgentStore.hs

@@ -76,6 +76,7 @@ module Simplex.Messaging.Agent.Store.AgentStore
     getExpiredServiceConns,
     deleteExpiredServiceRequests,
     getDeletedWaitingDeliveryConnIds,
+    getConnRatchetSync,
     setConnRatchetSync,
     addProcessedRatchetKeyHash,
     checkRatchetKeyHashExists,
@@ -99,6 +100,7 @@ module Simplex.Messaging.Agent.Store.AgentStore
     setSndQueuePrimary,
     deleteConnRcvQueue,
     incRcvDeleteErrors,
+    getEndedRcvQueueConnIds,
     deleteConnSndQueue,
     getPrimaryRcvQueue,
     getRcvQueue,
@@ -165,6 +167,8 @@ module Simplex.Messaging.Agent.Store.AgentStore
     createRatchet,
     deleteRatchet,
     getRatchet,
+    getRatchetVerifyCodes,
+    ratchetVerifyCodes,
     getRatchetForUpdate,
     getSkippedMsgKeys,
     updateRatchet,
@@ -213,6 +217,7 @@ module Simplex.Messaging.Agent.Store.AgentStore
     -- Rcv files
     createRcvFile,
     createRcvFileRedirect,
+    startPreparedRcvFile,
     lockRcvFileForUpdate,
     getRcvFile,
     getRcvFileByEntityId,
@@ -234,6 +239,7 @@ module Simplex.Messaging.Agent.Store.AgentStore
     getRcvFilesExpired,
     -- Snd files
     createSndFile,
+    startPreparedSndFile,
     lockSndFileForUpdate,
     getSndFile,
     getSndFileByEntityId,
@@ -291,6 +297,7 @@ import Data.ByteString (ByteString)
 import qualified Data.ByteString.Base64.URL as U
 import qualified Data.ByteString.Char8 as B
 import qualified Data.ByteString.Lazy as LB
+import Data.Either (partitionEithers)
 import Data.Functor (($>))
 import Data.Int (Int64)
 import Data.List (foldl', sortBy)
@@ -757,6 +764,10 @@ incRcvDeleteErrors :: DB.Connection -> RcvQueue -> IO ()
 incRcvDeleteErrors db RcvQueue {connId, dbQueueId} =
   DB.execute db "UPDATE rcv_queues SET delete_errors = delete_errors + 1 WHERE conn_id = ? AND rcv_queue_id = ?" (connId, dbQueueId)
 
+getEndedRcvQueueConnIds :: DB.Connection -> IO [ConnId]
+getEndedRcvQueueConnIds db =
+  map fromOnly <$> DB.query db "SELECT conn_id FROM rcv_queues WHERE switch_status = ? AND deleted = 0" (Only RSReceivedQEND)
+
 deleteConnRcvQueue :: DB.Connection -> RcvQueue -> IO ()
 deleteConnRcvQueue db RcvQueue {connId, dbQueueId} =
   DB.execute db "DELETE FROM rcv_queues WHERE conn_id = ? AND rcv_queue_id = ?" (connId, dbQueueId)
@@ -1249,7 +1260,7 @@ getLastMsg db connId msgId =
         LEFT JOIN snd_messages s ON s.conn_id = r.conn_id AND s.rcpt_internal_id = r.internal_id
         WHERE r.conn_id = ? AND r.broker_id = ?
       |]
-      (connId, Binary msgId)
+      (connId, msgId)
 
 toRcvMsg :: (Int64, InternalTs, BrokerId, BrokerTs) :. (AgentMsgId, MsgIntegrity, MsgHash, AgentMessageType, MsgBody, PQEncryption, Maybe AgentMsgId, Maybe MsgReceiptStatus, BoolInt) -> RcvMsg
 toRcvMsg ((agentMsgId, internalTs, brokerId, brokerTs) :. (sndMsgId, integrity, internalHash, msgType, msgBody, pqEncryption, rcptInternalId_, rcptStatus_, BI userAck)) =
@@ -1281,7 +1292,7 @@ checkRcvMsgHashExists db connId hash =
 getRcvMsgBrokerTs :: DB.Connection -> ConnId -> SMP.MsgId -> IO (Either StoreError BrokerTs)
 getRcvMsgBrokerTs db connId msgId =
   firstRow fromOnly (SEMsgNotFound "getRcvMsgBrokerTs") $
-    DB.query db "SELECT broker_ts FROM rcv_messages WHERE conn_id = ? AND broker_id = ?" (connId, Binary msgId)
+    DB.query db "SELECT broker_ts FROM rcv_messages WHERE conn_id = ? AND broker_id = ?" (connId, msgId)
 
 deleteMsg :: DB.Connection -> ConnId -> InternalId -> IO ()
 deleteMsg db connId msgId =
@@ -1474,9 +1485,11 @@ createSndRatchet db connId ratchetState (CR.AE2ERatchetParams s (CR.E2ERatchetPa
     db
     [sql|
       INSERT INTO ratchets
-        (conn_id, ratchet_state, x3dh_pub_key_1, x3dh_pub_key_2, pq_pub_kem) VALUES (?, ?, ?, ?, ?)
+        (conn_id, ratchet_state, rc_verify_code_ad, rc_verify_code_pq, x3dh_pub_key_1, x3dh_pub_key_2, pq_pub_kem) VALUES (?, ?, ?, ?, ?, ?, ?)
       ON CONFLICT (conn_id) DO UPDATE SET
         ratchet_state = EXCLUDED.ratchet_state,
+        rc_verify_code_ad = EXCLUDED.rc_verify_code_ad,
+        rc_verify_code_pq = EXCLUDED.rc_verify_code_pq,
         x3dh_priv_key_1 = NULL,
         x3dh_priv_key_2 = NULL,
         x3dh_pub_key_1 = EXCLUDED.x3dh_pub_key_1,
@@ -1484,7 +1497,13 @@ createSndRatchet db connId ratchetState (CR.AE2ERatchetParams s (CR.E2ERatchetPa
         pq_priv_kem = NULL,
         pq_pub_kem = EXCLUDED.pq_pub_kem
     |]
-    (connId, ratchetState, x3dhPubKey1, x3dhPubKey2, CR.ARKP s <$> pqPubKem)
+    ((connId, ratchetState) :. verifyCodesRow (ratchetVerifyCodes ratchetState) :. (x3dhPubKey1, x3dhPubKey2, CR.ARKP s <$> pqPubKem))
+
+ratchetVerifyCodes :: RatchetX448 -> ConnVerifyCodes
+ratchetVerifyCodes CR.Ratchet {rcAD, rcVCPQ} = ConnVerifyCodes {codeAD = C.sha256Hash $ unStr rcAD, codePQ = unStr <$> rcVCPQ}
+
+verifyCodesRow :: ConnVerifyCodes -> (Binary ByteString, Maybe (Binary ByteString))
+verifyCodesRow ConnVerifyCodes {codeAD, codePQ} = (Binary codeAD, Binary <$> codePQ)
 
 getSndRatchet :: DB.Connection -> ConnId -> CR.VersionE2E -> IO (Either StoreError (RatchetX448, CR.AE2ERatchetParams 'C.X448))
 getSndRatchet db connId v =
@@ -1505,10 +1524,12 @@ createRatchet db connId rc =
   DB.execute
     db
     [sql|
-      INSERT INTO ratchets (conn_id, ratchet_state)
-      VALUES (?, ?)
+      INSERT INTO ratchets (conn_id, ratchet_state, rc_verify_code_ad, rc_verify_code_pq)
+      VALUES (?, ?, ?, ?)
       ON CONFLICT (conn_id) DO UPDATE SET
-        ratchet_state = ?,
+        ratchet_state = EXCLUDED.ratchet_state,
+        rc_verify_code_ad = EXCLUDED.rc_verify_code_ad,
+        rc_verify_code_pq = EXCLUDED.rc_verify_code_pq,
         x3dh_priv_key_1 = NULL,
         x3dh_priv_key_2 = NULL,
         x3dh_pub_key_1 = NULL,
@@ -1516,7 +1537,42 @@ createRatchet db connId rc =
         pq_priv_kem = NULL,
         pq_pub_kem = NULL
     |]
-    (connId, rc, rc)
+    ((connId, rc) :. verifyCodesRow (ratchetVerifyCodes rc))
+
+getRatchetVerifyCodes :: DB.Connection -> [ConnId] -> IO (Map ConnId ConnVerifyCodes)
+getRatchetVerifyCodes db connIds = do
+  (saved, computed) <- partitionEithers . mapMaybe rowCodes <$> selectCodes
+  stored <- saveCodes computed
+  pure $ M.fromList $ saved <> stored
+  where
+    rowCodes (connId, codeAD_, codePQ_, rc_) = case codeAD_ of
+      Just codeAD -> Just $ Left $ toCodes (connId, codeAD, codePQ_)
+      Nothing -> Right . (connId,) . ratchetVerifyCodes <$> rc_
+    toCodes (connId, Binary codeAD, codePQ_) = (connId, ConnVerifyCodes {codeAD, codePQ = fromBinary <$> codePQ_})
+    codesRow (connId, cs) = verifyCodesRow cs :. Only connId
+    codesQuery :: Query
+    codesQuery = "SELECT conn_id, rc_verify_code_ad, rc_verify_code_pq, CASE WHEN rc_verify_code_ad IS NULL THEN ratchet_state END FROM ratchets"
+    selectCodes :: IO [(ConnId, Maybe (Binary ByteString), Maybe (Binary ByteString), Maybe RatchetX448)]
+    saveCodes :: [(ConnId, ConnVerifyCodes)] -> IO [(ConnId, ConnVerifyCodes)]
+#if defined(dbPostgres)
+    selectCodes = DB.query db (codesQuery <> " WHERE conn_id IN ?") (Only (In connIds))
+    saveCodes cs =
+      map toCodes
+        <$> DB.returning
+          db
+          [sql|
+            UPDATE ratchets r
+            SET rc_verify_code_ad = COALESCE(r.rc_verify_code_ad, (upd.rc_verify_code_ad :: BYTEA)),
+              rc_verify_code_pq = CASE WHEN r.rc_verify_code_ad IS NULL THEN (upd.rc_verify_code_pq :: BYTEA) ELSE r.rc_verify_code_pq END
+            FROM (VALUES(?, ?, ?)) AS upd(rc_verify_code_ad, rc_verify_code_pq, conn_id)
+            WHERE r.conn_id = (upd.conn_id :: BYTEA)
+            RETURNING r.conn_id, r.rc_verify_code_ad, r.rc_verify_code_pq
+          |]
+          (map codesRow cs)
+#else
+    selectCodes = concat <$> mapM (DB.query db (codesQuery <> " WHERE conn_id = ?") . Only) connIds
+    saveCodes cs = cs <$ unless (null cs) (DB.executeMany db "UPDATE ratchets SET rc_verify_code_ad = ?, rc_verify_code_pq = ? WHERE conn_id = ?" $ map codesRow cs)
+#endif
 
 deleteRatchet :: DB.Connection -> ConnId -> IO ()
 deleteRatchet db connId =
@@ -1544,12 +1600,16 @@ getRatchet_ q db connId =
   where
     ratchet = maybe (Left SERatchetNotFound) Right . fromOnly
 
-getSkippedMsgKeys :: DB.Connection -> ConnId -> IO SkippedMsgKeys
-getSkippedMsgKeys db connId =
-  skipped <$> DB.query db "SELECT header_key, msg_n, msg_key FROM skipped_messages WHERE conn_id = ?" (Only connId)
+getSkippedMsgKeys :: DB.Connection -> ConnId -> Int -> IO SkippedMsgKeys
+getSkippedMsgKeys db connId maxKeys = do
+  (keys, oldKeys) <- splitAt maxKeys <$> DB.query db "SELECT skipped_message_id, header_key, msg_n, msg_key FROM skipped_messages WHERE conn_id = ? ORDER BY skipped_message_id DESC LIMIT ?" (connId, maxKeys + 1)
+  case oldKeys of
+    (skippedMsgId :: Int64, _, _, _) : _ -> DB.execute db "DELETE FROM skipped_messages WHERE conn_id = ? AND skipped_message_id <= ?" (connId, skippedMsgId)
+    [] -> pure ()
+  pure $ skipped keys
   where
     skipped = foldl' addSkippedKey M.empty
-    addSkippedKey smks (hk, msgN, mk) = M.alter (Just . addMsgKey) hk smks
+    addSkippedKey smks (_, hk, msgN, mk) = M.alter (Just . addMsgKey) hk smks
       where
         addMsgKey = maybe (M.singleton msgN mk) (M.insert msgN mk)
 
@@ -1668,7 +1728,7 @@ getPendingServerCommand db connId srv_ = getWorkItem "command" getCmdId getComma
         DB.query
           db
           [sql|
-            SELECT c.corr_id, cs.user_id, c.command
+            SELECT c.corr_id, cs.user_id, c.command, c.created_at
             FROM commands c
             JOIN connections cs USING (conn_id)
             WHERE c.command_id = ?
@@ -1676,7 +1736,7 @@ getPendingServerCommand db connId srv_ = getWorkItem "command" getCmdId getComma
           (Only cmdId)
       where
         err = SEInternal $ "command  " <> bshow cmdId <> " returned []"
-        pendingCommand (corrId, userId, command) = PendingCommand {cmdId, corrId, userId, connId, command}
+        pendingCommand (corrId, userId, command, createdAt) = PendingCommand {cmdId, corrId, userId, connId, command, createdAt}
     markCommandFailed cmdId = DB.execute db "UPDATE commands SET failed = 1 WHERE command_id = ?" (Only cmdId)
 
 updateCommandServer :: DB.Connection -> AsyncCmdId -> SMPServer -> IO (Either StoreError ())
@@ -2450,12 +2510,12 @@ getSubscriptionServers db onlyNeeded =
 
 -- TODO [certs rcv] check index for getting queues with service present
 getUserServerRcvQueueSubs :: DB.Connection -> UserId -> SMPServer -> Bool -> ServiceAssoc -> Int -> Maybe SMP.RecipientId -> IO [RcvQueueSub]
-getUserServerRcvQueueSubs db userId (SMPServer h p kh) onlyNeeded hasService limit cursor_ =
-  map toRcvQueueSub <$> case cursor_ of
+getUserServerRcvQueueSubs db userId srv@(SMPServer h p kh) onlyNeeded hasService limit cursor_ =
+  map (toServerRcvQueueSub srv) <$> case cursor_ of
     Nothing -> DB.query db (q <> orderLimit) (userId, h, p, kh, limit)
     Just cursor -> DB.query db (q <> " AND q.rcv_id > ? " <> orderLimit) (userId, h, p, kh, cursor, limit)
   where
-    q = rcvQueueSubQuery <> toSubscribe <> " c.deleted = 0 AND q.deleted = 0 AND c.user_id = ? AND q.host = ? AND q.port = ? AND COALESCE(q.server_key_hash, s.key_hash) = ?" <> serviceCond
+    q = serverRcvQueueSubQuery <> toSubscribe <> " c.deleted = 0 AND q.deleted = 0 AND c.user_id = ? AND q.host = ? AND q.port = ? AND COALESCE(q.server_key_hash, s.key_hash) = ?" <> serviceCond
     orderLimit = " ORDER BY q.rcv_id LIMIT ?"
     toSubscribe
       | onlyNeeded = " WHERE q.to_subscribe = 1 AND "
@@ -2468,7 +2528,7 @@ unassocUserServerRcvQueueSubs :: DB.Connection -> UserId -> SMPServer -> IO [Rcv
 unassocUserServerRcvQueueSubs db userId srv@(SMPServer h p kh) = do
   deleteClientService db userId srv
 #if defined(dbPostgres)
-  map toRcvQueueSub
+  map (toServerRcvQueueSub srv)
     <$> DB.query
       db
       (removeRcvAssocsQuery <> " " <> returningColumns)
@@ -2476,15 +2536,15 @@ unassocUserServerRcvQueueSubs db userId srv@(SMPServer h p kh) = do
   where
     returningColumns =
       [sql|
-        RETURNING c.user_id, rcv_queues.conn_id, rcv_queues.host, rcv_queues.port, COALESCE(rcv_queues.server_key_hash, s.key_hash),
+        RETURNING c.user_id, rcv_queues.conn_id,
           rcv_queues.rcv_id, rcv_queues.rcv_private_key, rcv_queues.status, c.enable_ntfs, rcv_queues.client_notice_id,
           rcv_queues.rcv_queue_id, rcv_queues.rcv_primary, rcv_queues.replace_rcv_queue_id
       |]
 #else
-  qs <- map toRcvQueueSub
+  qs <- map (toServerRcvQueueSub srv)
     <$> DB.query
       db
-      (rcvQueueSubQuery <> " WHERE c.user_id = ? AND q.host = ? AND q.port = ? AND COALESCE(q.server_key_hash, s.key_hash) = ? AND q.rcv_service_assoc = 1")
+      (serverRcvQueueSubQuery <> " WHERE c.user_id = ? AND q.host = ? AND q.port = ? AND COALESCE(q.server_key_hash, s.key_hash) = ? AND q.rcv_service_assoc = 1")
       (userId, h, p, kh)
   DB.execute db removeRcvAssocsQuery (h, p, userId, kh)
   pure qs
@@ -2743,6 +2803,11 @@ getDeletedWaitingDeliveryConnIds :: DB.Connection -> IO [ConnId]
 getDeletedWaitingDeliveryConnIds db =
   map fromOnly <$> DB.query_ db "SELECT conn_id FROM connections WHERE deleted_at_wait_delivery IS NOT NULL"
 
+getConnRatchetSync :: DB.Connection -> ConnId -> IO (Either StoreError RatchetSyncState)
+getConnRatchetSync db connId =
+  firstRow fromOnly SEConnNotFound $
+    DB.query db "SELECT ratchet_sync_state FROM connections WHERE conn_id = ?" (Only connId)
+
 setConnRatchetSync :: DB.Connection -> ConnId -> RatchetSyncState -> IO ()
 setConnRatchetSync db connId ratchetSyncState =
   DB.execute db "UPDATE connections SET ratchet_sync_state = ? WHERE conn_id = ?" (ratchetSyncState, connId)
@@ -2764,21 +2829,35 @@ checkRatchetKeyHashExists db connId hash =
       (connId, Binary hash)
 
 deleteRatchetKeyHashesExpired :: DB.Connection -> NominalDiffTime -> Int -> IO ()
-deleteRatchetKeyHashesExpired db ttl limit = do
+deleteRatchetKeyHashesExpired db ttl maxConnHashes = do
   cutoffTs <- addUTCTime (-ttl) <$> getCurrentTime
+#if defined(dbPostgres)
   DB.execute
     db
-    [sql|
-      DELETE FROM processed_ratchet_key_hashes
-      WHERE processed_ratchet_key_hash_id IN (
-        SELECT processed_ratchet_key_hash_id
-        FROM processed_ratchet_key_hashes
-        WHERE created_at < ?
-        ORDER BY created_at ASC
-        LIMIT ?
-      )
-    |]
-    (cutoffTs, limit)
+    ("DELETE FROM processed_ratchet_key_hashes h USING (" <> maxExcessIdsQuery <> ") e WHERE h.conn_id = e.conn_id AND h.processed_ratchet_key_hash_id <= e.max_excess_id AND h.created_at < ?")
+    (maxConnHashes, maxConnHashes, cutoffTs)
+#else
+  maxExcessIds <- DB.query db maxExcessIdsQuery (maxConnHashes, maxConnHashes)
+  DB.executeMany
+    db
+    "DELETE FROM processed_ratchet_key_hashes WHERE conn_id = ? AND processed_ratchet_key_hash_id <= ? AND created_at < ?"
+    (map (\(connId :: ConnId, maxExcessId :: Int64) -> (connId, maxExcessId, cutoffTs)) maxExcessIds)
+#endif
+  where
+    maxExcessIdsQuery :: Query
+    maxExcessIdsQuery =
+      [sql|
+        SELECT conn_id, (
+          SELECT processed_ratchet_key_hash_id
+          FROM processed_ratchet_key_hashes
+          WHERE conn_id = c.conn_id
+          ORDER BY processed_ratchet_key_hash_id DESC
+          LIMIT 1 OFFSET ?
+        ) AS max_excess_id
+        FROM processed_ratchet_key_hashes c
+        GROUP BY conn_id
+        HAVING COUNT(*) > ?
+      |]
 
 -- | returns all connection queues, the first queue is the primary one
 getRcvQueuesByConnId_ :: DB.Connection -> ConnId -> IO (Maybe (NonEmpty RcvQueue))
@@ -2835,17 +2914,32 @@ getRcvQueueSubsByConnId_ db connId =
 rcvQueueSubQuery :: Query
 rcvQueueSubQuery =
   [sql|
-    SELECT c.user_id, q.conn_id, q.host, q.port, COALESCE(q.server_key_hash, s.key_hash), q.rcv_id, q.rcv_private_key, q.status, c.enable_ntfs, q.client_notice_id,
+    SELECT q.host, q.port, COALESCE(q.server_key_hash, s.key_hash), c.user_id, q.conn_id, q.rcv_id, q.rcv_private_key, q.status, c.enable_ntfs, q.client_notice_id,
       q.rcv_queue_id, q.rcv_primary, q.replace_rcv_queue_id
     FROM rcv_queues q
     JOIN servers s ON q.host = s.host AND q.port = s.port
     JOIN connections c ON q.conn_id = c.conn_id
   |]
 
-toRcvQueueSub :: (UserId, ConnId, NonEmpty TransportHost, ServiceName, C.KeyHash, SMP.RecipientId, SMP.RcvPrivateAuthKey) :. (QueueStatus, Maybe BoolInt, Maybe NoticeId, Int64, BoolInt, Maybe Int64) -> RcvQueueSub
-toRcvQueueSub ((userId, connId, host, port, keyHash, rcvId, rcvPrivateKey) :. (status, enableNtfs_, clientNoticeId, dbQueueId, BI primary, dbReplaceQueueId)) =
+serverRcvQueueSubQuery :: Query
+serverRcvQueueSubQuery =
+  [sql|
+    SELECT c.user_id, q.conn_id, q.rcv_id, q.rcv_private_key, q.status, c.enable_ntfs, q.client_notice_id,
+      q.rcv_queue_id, q.rcv_primary, q.replace_rcv_queue_id
+    FROM rcv_queues q
+    JOIN servers s ON q.host = s.host AND q.port = s.port
+    JOIN connections c ON q.conn_id = c.conn_id
+  |]
+
+type RcvQueueSubRow = (UserId, ConnId, SMP.RecipientId, SMP.RcvPrivateAuthKey) :. (QueueStatus, Maybe BoolInt, Maybe NoticeId, Int64, BoolInt, Maybe Int64)
+
+toRcvQueueSub :: (NonEmpty TransportHost, ServiceName, C.KeyHash) :. RcvQueueSubRow -> RcvQueueSub
+toRcvQueueSub ((host, port, keyHash) :. r) = toServerRcvQueueSub (SMPServer host port keyHash) r
+
+toServerRcvQueueSub :: SMPServer -> RcvQueueSubRow -> RcvQueueSub
+toServerRcvQueueSub server ((userId, connId, rcvId, rcvPrivateKey) :. (status, enableNtfs_, clientNoticeId, dbQueueId, BI primary, dbReplaceQueueId)) =
   let enableNtfs = maybe True unBI enableNtfs_
-   in RcvQueueSub {userId, connId, server = SMPServer host port keyHash, rcvId, rcvPrivateKey, status, enableNtfs, clientNoticeId, dbQueueId, primary, dbReplaceQueueId}
+   in RcvQueueSub {userId, connId, server, rcvId, rcvPrivateKey, status, enableNtfs, clientNoticeId, dbQueueId, primary, dbReplaceQueueId}
 
 getRcvQueueById :: DB.Connection -> ConnId -> Int64 -> IO (Either StoreError RcvQueue)
 getRcvQueueById db connId dbRcvId =
@@ -2946,7 +3040,7 @@ insertRcvMsgDetails_ db connId RcvQueue {dbQueueId} RcvMsgData {msgMeta, interna
       VALUES
         (?,?,?,?,?,?,?,?,?,?)
     |]
-    (connId, dbQueueId, internalRcvId, fst recipient, sndMsgId, Binary (fst broker), snd broker, Binary internalHash, Binary externalPrevSndHash, integrity)
+    (connId, dbQueueId, internalRcvId, fst recipient, sndMsgId, fst broker, snd broker, Binary internalHash, Binary externalPrevSndHash, integrity)
   DB.execute db "INSERT INTO encrypted_rcv_message_hashes (conn_id, hash) VALUES (?,?)" (connId, Binary encryptedMsgHash)
 
 updateRcvMsgHash :: DB.Connection -> ConnId -> AgentMsgId -> InternalRcvId -> MsgHash -> IO ()
@@ -3094,7 +3188,8 @@ createRcvFile db gVar userId fd@FileDescription {chunks} prefixPath tmpPath file
 createRcvFileRedirect :: DB.Connection -> TVar ChaChaDRG -> UserId -> FileDescription 'FRecipient -> FilePath -> FilePath -> CryptoFile -> FilePath -> CryptoFile -> Bool -> IO (Either StoreError RcvFileId)
 createRcvFileRedirect _ _ _ FileDescription {redirect = Nothing} _ _ _ _ _ _ = pure $ Left $ SEInternal "createRcvFileRedirect called without redirect"
 createRcvFileRedirect db gVar userId redirectFd@FileDescription {chunks = redirectChunks, redirect = Just RedirectFileInfo {size, digest}} prefixPath redirectPath redirectFile dstPath dstFile approvedRelays = runExceptT $ do
-  (dstEntityId, dstId) <- ExceptT $ insertRcvFile db gVar userId dummyDst prefixPath dstPath dstFile Nothing Nothing approvedRelays
+  nonce <- atomically $ C.randomCbNonce gVar
+  (dstEntityId, dstId) <- ExceptT $ insertRcvFile db gVar userId (dummyDst nonce) prefixPath dstPath dstFile Nothing Nothing approvedRelays
   (_, redirectId) <- ExceptT $ insertRcvFile db gVar userId redirectFd prefixPath redirectPath redirectFile (Just dstId) (Just dstEntityId) approvedRelays
   liftIO $
     forM_ redirectChunks $ \fc@FileChunk {replicas} -> do
@@ -3102,7 +3197,7 @@ createRcvFileRedirect db gVar userId redirectFd@FileDescription {chunks = redire
       forM_ (zip [1 ..] replicas) $ \(rno, replica) -> insertRcvFileChunkReplica db rno replica chunkId
   pure dstEntityId
   where
-    dummyDst =
+    dummyDst nonce =
       FileDescription
         { party = SFRecipient,
           size,
@@ -3110,11 +3205,34 @@ createRcvFileRedirect db gVar userId redirectFd@FileDescription {chunks = redire
           redirect = Nothing,
           -- updated later with updateRcvFileRedirect
           key = C.unsafeSbKey $ B.replicate 32 '#',
-          nonce = C.cbNonce "",
+          nonce,
           chunkSize = FileSize 0,
           chunks = []
         }
 
+startPreparedRcvFile :: DB.Connection -> RcvFileId -> IO (Either StoreError [XFTPServer])
+startPreparedRcvFile db rcvFileEntityId = runExceptT $ do
+  rcvFileId <- ExceptT $ getRcvFileIdByEntityId_ db rcvFileEntityId
+  liftIO $ do
+    updatedAt <- getCurrentTime
+    DB.execute
+      db
+      "UPDATE rcv_files SET status = ?, updated_at = ? WHERE (rcv_file_id = ? OR redirect_id = ?) AND status = ?"
+      (RFSReceiving, updatedAt, rcvFileId, rcvFileId, RFSPrepared)
+    map toXFTPServer
+      <$> DB.query
+        db
+        [sql|
+          SELECT DISTINCT
+            s.xftp_host, s.xftp_port, s.xftp_key_hash
+          FROM rcv_file_chunk_replicas r
+          JOIN xftp_servers s ON s.xftp_server_id = r.xftp_server_id
+          JOIN rcv_file_chunks c ON c.rcv_file_chunk_id = r.rcv_file_chunk_id
+          JOIN rcv_files f ON f.rcv_file_id = c.rcv_file_id
+          WHERE (f.rcv_file_id = ? OR f.redirect_id = ?) AND r.replica_number = 1
+        |]
+        (rcvFileId, rcvFileId)
+
 insertRcvFile :: DB.Connection -> TVar ChaChaDRG -> UserId -> FileDescription 'FRecipient -> FilePath -> FilePath -> CryptoFile -> Maybe DBRcvFileId -> Maybe RcvFileId -> Bool -> IO (Either StoreError (RcvFileId, DBRcvFileId))
 insertRcvFile db gVar userId FileDescription {size, digest, key, nonce, chunkSize, redirect} prefixPath tmpPath (CryptoFile savePath cfArgs) redirectId_ redirectEntityId_ approvedRelays = runExceptT $ do
   let (redirectDigest_, redirectSize_) = case redirect of
@@ -3125,7 +3243,7 @@ insertRcvFile db gVar userId FileDescription {size, digest, key, nonce, chunkSiz
       DB.execute
         db
         "INSERT INTO rcv_files (rcv_file_entity_id, user_id, size, digest, key, nonce, chunk_size, prefix_path, tmp_path, save_path, save_file_key, save_file_nonce, status, redirect_id, redirect_entity_id, redirect_digest, redirect_size, approved_relays) VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)"
-        ((Binary rcvFileEntityId, userId, size, digest, key, nonce, chunkSize, prefixPath, tmpPath) :. (savePath, fileKey <$> cfArgs, fileNonce <$> cfArgs, RFSReceiving, redirectId_, Binary <$> redirectEntityId_, redirectDigest_, redirectSize_, BI approvedRelays))
+        ((Binary rcvFileEntityId, userId, size, digest, key, nonce, chunkSize, prefixPath, tmpPath) :. (savePath, fileKey <$> cfArgs, fileNonce <$> cfArgs, RFSPrepared, redirectId_, Binary <$> redirectEntityId_, redirectDigest_, redirectSize_, BI approvedRelays))
   rcvFileId <- liftIO $ insertedRowId db
   pure (rcvFileEntityId, rcvFileId)
 
@@ -3430,13 +3548,23 @@ createSndFile db gVar userId (CryptoFile path cfArgs) numRecipients prefixPath k
     DB.execute
       db
       "INSERT INTO snd_files (snd_file_entity_id, user_id, path, src_file_key, src_file_nonce, num_recipients, prefix_path, key, nonce, status, redirect_size, redirect_digest, storage_time) VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?)"
-      ((Binary sndFileEntityId, userId, path, fileKey <$> cfArgs, fileNonce <$> cfArgs, numRecipients) :. (prefixPath, key, nonce, SFSNew, redirectSize_, redirectDigest_, storageHours))
+      ((Binary sndFileEntityId, userId, path, fileKey <$> cfArgs, fileNonce <$> cfArgs, numRecipients) :. (prefixPath, key, nonce, SFSPrepared, redirectSize_, redirectDigest_, storageHours))
   where
     (redirectSize_, redirectDigest_) =
       case redirect_ of
         Nothing -> (Nothing, Nothing)
         Just RedirectFileInfo {size, digest} -> (Just size, Just digest)
 
+startPreparedSndFile :: DB.Connection -> SndFileId -> IO (Either StoreError ())
+startPreparedSndFile db sndFileEntityId = runExceptT $ do
+  sndFileId <- ExceptT $ getSndFileIdByEntityId_ db sndFileEntityId
+  liftIO $ do
+    updatedAt <- getCurrentTime
+    DB.execute
+      db
+      "UPDATE snd_files SET status = ?, updated_at = ? WHERE snd_file_id = ? AND status = ?"
+      (SFSNew, updatedAt, sndFileId, SFSPrepared)
+
 getSndFileByEntityId :: DB.Connection -> SndFileId -> IO (Either StoreError SndFile)
 getSndFileByEntityId db sndFileEntityId = runExceptT $ do
   sndFileId <- ExceptT $ getSndFileIdByEntityId_ db sndFileEntityId

src/Simplex/Messaging/Agent/Store/Postgres/DB.hs

@@ -13,6 +13,7 @@ module Simplex.Messaging.Agent.Store.Postgres.DB
     execute,
     execute_,
     executeMany,
+    returning,
     query,
     query_,
     blobFieldDecoder,
@@ -60,6 +61,10 @@ executeMany :: ToRow q => PSQL.Connection -> Query -> [q] -> IO ()
 executeMany db q qs = void $ PSQL.executeMany db q qs `E.catch` addSql q
 {-# INLINE executeMany #-}
 
+returning :: (ToRow q, FromRow r) => PSQL.Connection -> Query -> [q] -> IO [r]
+returning db q qs = PSQL.returning db q qs `E.catch` addSql q
+{-# INLINE returning #-}
+
 query :: (ToRow q, FromRow r) => PSQL.Connection -> Query -> q -> IO [r]
 query db q qs = PSQL.query db q qs `E.catch` addSql q
 {-# INLINE query #-}

src/Simplex/Messaging/Agent/Store/Postgres/Migrations/App.hs

@@ -15,6 +15,9 @@ import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260410_receive_attem
 import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260411_service_certs
 import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260712_address_dr_rpc
 import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260823_snd_files_entitlement
+import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260919_ratchet_verify_codes
+import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260929_ratchet_indexes
+import Simplex.Messaging.Agent.Store.Postgres.Migrations.M20261008_rcv_switch_status
 import Simplex.Messaging.Agent.Store.Shared (Migration (..))
 
 schemaMigrations :: [(String, Text, Maybe Text)]
@@ -29,7 +32,10 @@ schemaMigrations =
     ("20260410_receive_attempts", m20260410_receive_attempts, Just down_m20260410_receive_attempts),
     ("20260411_service_certs", m20260411_service_certs, Just down_m20260411_service_certs),
     ("20260712_address_dr_rpc", m20260712_address_dr_rpc, Just down_m20260712_address_dr_rpc),
-    ("20260823_snd_files_entitlement", m20260823_snd_files_entitlement, Just down_m20260823_snd_files_entitlement)
+    ("20260823_snd_files_entitlement", m20260823_snd_files_entitlement, Just down_m20260823_snd_files_entitlement),
+    ("20260919_ratchet_verify_codes", m20260919_ratchet_verify_codes, Just down_m20260919_ratchet_verify_codes),
+    ("20260929_ratchet_indexes", m20260929_ratchet_indexes, Just down_m20260929_ratchet_indexes),
+    ("20261008_rcv_switch_status", m20261008_rcv_switch_status, Just down_m20261008_rcv_switch_status)
   ]
 
 -- | The list of migrations in ascending order by date

src/Simplex/Messaging/Agent/Store/Postgres/Migrations/M20260919_ratchet_verify_codes.hs

@@ -0,0 +1,21 @@
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260919_ratchet_verify_codes where
+
+import Data.Text (Text)
+import Text.RawString.QQ (r)
+
+m20260919_ratchet_verify_codes :: Text
+m20260919_ratchet_verify_codes =
+  [r|
+ALTER TABLE ratchets ADD COLUMN rc_verify_code_ad BYTEA;
+ALTER TABLE ratchets ADD COLUMN rc_verify_code_pq BYTEA;
+|]
+
+down_m20260919_ratchet_verify_codes :: Text
+down_m20260919_ratchet_verify_codes =
+  [r|
+ALTER TABLE ratchets DROP COLUMN rc_verify_code_pq;
+ALTER TABLE ratchets DROP COLUMN rc_verify_code_ad;
+|]

src/Simplex/Messaging/Agent/Store/Postgres/Migrations/M20260929_ratchet_indexes.hs

@@ -0,0 +1,34 @@
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.Postgres.Migrations.M20260929_ratchet_indexes where
+
+import Data.Text (Text)
+import Text.RawString.QQ (r)
+
+m20260929_ratchet_indexes :: Text
+m20260929_ratchet_indexes =
+  [r|
+DROP INDEX idx_skipped_messages_conn_id;
+CREATE INDEX idx_skipped_messages_conn_id ON skipped_messages(conn_id, skipped_message_id);
+
+DELETE FROM processed_ratchet_key_hashes AS h
+WHERE EXISTS (
+  SELECT 1 FROM processed_ratchet_key_hashes d
+  WHERE d.conn_id = h.conn_id AND d.hash = h.hash AND d.processed_ratchet_key_hash_id < h.processed_ratchet_key_hash_id
+);
+
+DROP INDEX idx_processed_ratchet_key_hashes_hash;
+CREATE UNIQUE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(conn_id, hash);
+CREATE INDEX idx_processed_ratchet_key_hashes_conn_id ON processed_ratchet_key_hashes(conn_id, processed_ratchet_key_hash_id);
+|]
+
+down_m20260929_ratchet_indexes :: Text
+down_m20260929_ratchet_indexes =
+  [r|
+DROP INDEX idx_processed_ratchet_key_hashes_conn_id;
+DROP INDEX idx_processed_ratchet_key_hashes_hash;
+CREATE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(conn_id, hash);
+DROP INDEX idx_skipped_messages_conn_id;
+CREATE INDEX idx_skipped_messages_conn_id ON skipped_messages(conn_id);
+|]

src/Simplex/Messaging/Agent/Store/Postgres/Migrations/M20261008_rcv_switch_status.hs

@@ -0,0 +1,20 @@
+{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.Postgres.Migrations.M20261008_rcv_switch_status where
+
+import Data.Text (Text)
+import Text.RawString.QQ (r)
+
+m20261008_rcv_switch_status :: Text
+m20261008_rcv_switch_status =
+  [r|
+CREATE INDEX idx_rcv_queues_switch_status ON rcv_queues(switch_status);
+|]
+
+down_m20261008_rcv_switch_status :: Text
+down_m20261008_rcv_switch_status =
+  [r|
+UPDATE rcv_queues SET switch_status = NULL WHERE switch_status = 'received_qend';
+DROP INDEX idx_rcv_queues_switch_status;
+|]

src/Simplex/Messaging/Agent/Store/Postgres/Migrations/agent_postgres_schema.sql

@@ -448,7 +448,9 @@ CREATE TABLE smp_agent_test_protocol_schema.ratchets (
     x3dh_pub_key_1 bytea,
     x3dh_pub_key_2 bytea,
     pq_priv_kem bytea,
-    pq_pub_kem bytea
+    pq_pub_kem bytea,
+    rc_verify_code_ad bytea,
+    rc_verify_code_pq bytea
 );
 
 
@@ -1167,11 +1169,15 @@ CREATE INDEX idx_ntf_tokens_ntf_host_ntf_port ON smp_agent_test_protocol_schema.
 
 
 
+CREATE INDEX idx_processed_ratchet_key_hashes_conn_id ON smp_agent_test_protocol_schema.processed_ratchet_key_hashes USING btree (conn_id, processed_ratchet_key_hash_id);
+
+
+
 CREATE INDEX idx_processed_ratchet_key_hashes_created_at ON smp_agent_test_protocol_schema.processed_ratchet_key_hashes USING btree (created_at);
 
 
 
-CREATE INDEX idx_processed_ratchet_key_hashes_hash ON smp_agent_test_protocol_schema.processed_ratchet_key_hashes USING btree (conn_id, hash);
+CREATE UNIQUE INDEX idx_processed_ratchet_key_hashes_hash ON smp_agent_test_protocol_schema.processed_ratchet_key_hashes USING btree (conn_id, hash);
 
 
 
@@ -1227,6 +1233,10 @@ CREATE UNIQUE INDEX idx_rcv_queues_ntf ON smp_agent_test_protocol_schema.rcv_que
 
 
 
+CREATE INDEX idx_rcv_queues_switch_status ON smp_agent_test_protocol_schema.rcv_queues USING btree (switch_status);
+
+
+
 CREATE INDEX idx_rcv_queues_to_subscribe ON smp_agent_test_protocol_schema.rcv_queues USING btree (to_subscribe);
 
 
@@ -1239,7 +1249,7 @@ CREATE UNIQUE INDEX idx_server_certs_user_id_host_port ON smp_agent_test_protoco
 
 
 
-CREATE INDEX idx_skipped_messages_conn_id ON smp_agent_test_protocol_schema.skipped_messages USING btree (conn_id);
+CREATE INDEX idx_skipped_messages_conn_id ON smp_agent_test_protocol_schema.skipped_messages USING btree (conn_id, skipped_message_id);
 
 
 

src/Simplex/Messaging/Agent/Store/SQLite/Common.hs

@@ -100,7 +100,8 @@ withTransaction st = withTransactionPriority st False
 {-# INLINE withTransaction #-}
 
 withTransactionPriority :: DBStore -> Bool -> (DB.Connection -> IO a) -> IO a
-withTransactionPriority st priority action = withConnectionPriority st priority $ dbBusyLoop . transaction
+withTransactionPriority st@DBStore {dbClosed} priority action = withConnectionPriority st priority $ \db ->
+  ifM (readTVarIO dbClosed) (fail "withTransaction: database closed") (dbBusyLoop $ transaction db)
   where
     transaction db@DB.Connection {conn} = SQL.withImmediateTransaction conn $ action db
 

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/App.hs

@@ -51,6 +51,10 @@ import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260410_receive_attempt
 import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260411_service_certs
 import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260712_address_dr_rpc
 import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260823_snd_files_entitlement
+import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260919_ratchet_verify_codes
+import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260929_ratchet_indexes
+import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261005_commands_autoincrement
+import Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261008_rcv_switch_status
 import Simplex.Messaging.Agent.Store.Shared (Migration (..))
 
 schemaMigrations :: [(String, Query, Maybe Query)]
@@ -101,7 +105,11 @@ schemaMigrations =
     ("m20260410_receive_attempts", m20260410_receive_attempts, Just down_m20260410_receive_attempts),
     ("m20260411_service_certs", m20260411_service_certs, Just down_m20260411_service_certs),
     ("m20260712_address_dr_rpc", m20260712_address_dr_rpc, Just down_m20260712_address_dr_rpc),
-    ("m20260823_snd_files_entitlement", m20260823_snd_files_entitlement, Just down_m20260823_snd_files_entitlement)
+    ("m20260823_snd_files_entitlement", m20260823_snd_files_entitlement, Just down_m20260823_snd_files_entitlement),
+    ("m20260919_ratchet_verify_codes", m20260919_ratchet_verify_codes, Just down_m20260919_ratchet_verify_codes),
+    ("m20260929_ratchet_indexes", m20260929_ratchet_indexes, Just down_m20260929_ratchet_indexes),
+    ("m20261005_commands_autoincrement", m20261005_commands_autoincrement, Just down_m20261005_commands_autoincrement),
+    ("m20261008_rcv_switch_status", m20261008_rcv_switch_status, Just down_m20261008_rcv_switch_status)
   ]
 
 -- | The list of migrations in ascending order by date

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/M20260919_ratchet_verify_codes.hs

@@ -0,0 +1,20 @@
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260919_ratchet_verify_codes where
+
+import Database.SQLite.Simple (Query)
+import Database.SQLite.Simple.QQ (sql)
+
+m20260919_ratchet_verify_codes :: Query
+m20260919_ratchet_verify_codes =
+  [sql|
+ALTER TABLE ratchets ADD COLUMN rc_verify_code_ad BLOB;
+ALTER TABLE ratchets ADD COLUMN rc_verify_code_pq BLOB;
+  |]
+
+down_m20260919_ratchet_verify_codes :: Query
+down_m20260919_ratchet_verify_codes =
+  [sql|
+ALTER TABLE ratchets DROP COLUMN rc_verify_code_pq;
+ALTER TABLE ratchets DROP COLUMN rc_verify_code_ad;
+  |]

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/M20260929_ratchet_indexes.hs

@@ -0,0 +1,29 @@
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.SQLite.Migrations.M20260929_ratchet_indexes where
+
+import Database.SQLite.Simple (Query)
+import Database.SQLite.Simple.QQ (sql)
+
+-- idx_skipped_messages_conn_id is not changed: SQLite index entries include rowid (skipped_message_id)
+m20260929_ratchet_indexes :: Query
+m20260929_ratchet_indexes =
+  [sql|
+DELETE FROM processed_ratchet_key_hashes AS h
+WHERE EXISTS (
+  SELECT 1 FROM processed_ratchet_key_hashes d
+  WHERE d.conn_id = h.conn_id AND d.hash = h.hash AND d.processed_ratchet_key_hash_id < h.processed_ratchet_key_hash_id
+);
+
+DROP INDEX idx_processed_ratchet_key_hashes_hash;
+CREATE UNIQUE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(conn_id, hash);
+CREATE INDEX idx_processed_ratchet_key_hashes_conn_id ON processed_ratchet_key_hashes(conn_id, processed_ratchet_key_hash_id);
+  |]
+
+down_m20260929_ratchet_indexes :: Query
+down_m20260929_ratchet_indexes =
+  [sql|
+DROP INDEX idx_processed_ratchet_key_hashes_conn_id;
+DROP INDEX idx_processed_ratchet_key_hashes_hash;
+CREATE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(conn_id, hash);
+  |]

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/M20261005_commands_autoincrement.hs

@@ -0,0 +1,33 @@
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261005_commands_autoincrement where
+
+import Database.SQLite.Simple (Query)
+import Database.SQLite.Simple.QQ (sql)
+
+m20261005_commands_autoincrement :: Query
+m20261005_commands_autoincrement =
+  [sql|
+INSERT INTO sqlite_sequence (name, seq)
+SELECT 'commands', MAX(ROWID) FROM commands;
+
+PRAGMA writable_schema=1;
+
+UPDATE sqlite_master SET sql = replace(sql, 'command_id INTEGER PRIMARY KEY,', 'command_id INTEGER PRIMARY KEY AUTOINCREMENT,')
+WHERE name = 'commands' AND type = 'table';
+
+PRAGMA writable_schema=RESET;
+  |]
+
+down_m20261005_commands_autoincrement :: Query
+down_m20261005_commands_autoincrement =
+  [sql|
+DELETE FROM sqlite_sequence WHERE name = 'commands';
+
+PRAGMA writable_schema=1;
+
+UPDATE sqlite_master SET sql = replace(sql, 'command_id INTEGER PRIMARY KEY AUTOINCREMENT,', 'command_id INTEGER PRIMARY KEY,')
+WHERE name = 'commands' AND type = 'table';
+
+PRAGMA writable_schema=RESET;
+  |]

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/M20261008_rcv_switch_status.hs

@@ -0,0 +1,19 @@
+{-# LANGUAGE QuasiQuotes #-}
+
+module Simplex.Messaging.Agent.Store.SQLite.Migrations.M20261008_rcv_switch_status where
+
+import Database.SQLite.Simple (Query)
+import Database.SQLite.Simple.QQ (sql)
+
+m20261008_rcv_switch_status :: Query
+m20261008_rcv_switch_status =
+  [sql|
+CREATE INDEX idx_rcv_queues_switch_status ON rcv_queues(switch_status);
+|]
+
+down_m20261008_rcv_switch_status :: Query
+down_m20261008_rcv_switch_status =
+  [sql|
+UPDATE rcv_queues SET switch_status = NULL WHERE switch_status = 'received_qend';
+DROP INDEX idx_rcv_queues_switch_status;
+|]

src/Simplex/Messaging/Agent/Store/SQLite/Migrations/agent_schema.sql

@@ -182,7 +182,9 @@ CREATE TABLE ratchets(
   x3dh_pub_key_1 BLOB,
   x3dh_pub_key_2 BLOB,
   pq_priv_kem BLOB,
-  pq_pub_kem BLOB
+  pq_pub_kem BLOB,
+  rc_verify_code_ad BLOB,
+  rc_verify_code_pq BLOB
 ) WITHOUT ROWID, STRICT;
 CREATE TABLE skipped_messages(
   skipped_message_id INTEGER PRIMARY KEY,
@@ -245,7 +247,7 @@ CREATE TABLE ntf_subscriptions(
   ON DELETE RESTRICT ON UPDATE CASCADE
 ) WITHOUT ROWID, STRICT;
 CREATE TABLE commands(
-  command_id INTEGER PRIMARY KEY,
+  command_id INTEGER PRIMARY KEY AUTOINCREMENT,
   conn_id BLOB NOT NULL REFERENCES connections ON DELETE CASCADE,
   host TEXT,
   port TEXT,
@@ -259,6 +261,7 @@ CREATE TABLE commands(
   FOREIGN KEY(host, port) REFERENCES servers
   ON DELETE RESTRICT ON UPDATE CASCADE
 ) STRICT;
+CREATE TABLE sqlite_sequence(name,seq);
 CREATE TABLE snd_message_deliveries(
   snd_message_delivery_id INTEGER PRIMARY KEY AUTOINCREMENT,
   conn_id BLOB NOT NULL REFERENCES connections ON DELETE CASCADE,
@@ -267,7 +270,6 @@ CREATE TABLE snd_message_deliveries(
   failed INTEGER DEFAULT 0,
   FOREIGN KEY(conn_id, internal_id) REFERENCES messages ON DELETE CASCADE DEFERRABLE INITIALLY DEFERRED
 ) STRICT;
-CREATE TABLE sqlite_sequence(name,seq);
 CREATE TABLE users(
   user_id INTEGER PRIMARY KEY AUTOINCREMENT
   ,
@@ -571,10 +573,6 @@ CREATE INDEX idx_encrypted_rcv_message_hashes_hash ON encrypted_rcv_message_hash
   conn_id,
   hash
 );
-CREATE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(
-  conn_id,
-  hash
-);
 CREATE INDEX idx_snd_messages_rcpt_internal_id ON snd_messages(
   conn_id,
   rcpt_internal_id
@@ -639,6 +637,15 @@ CREATE INDEX idx_connections_deleted ON connections(deleted);
 CREATE INDEX idx_connections_service_request_expires_at ON connections(
   service_request_expires_at
 );
+CREATE UNIQUE INDEX idx_processed_ratchet_key_hashes_hash ON processed_ratchet_key_hashes(
+  conn_id,
+  hash
+);
+CREATE INDEX idx_processed_ratchet_key_hashes_conn_id ON processed_ratchet_key_hashes(
+  conn_id,
+  processed_ratchet_key_hash_id
+);
+CREATE INDEX idx_rcv_queues_switch_status ON rcv_queues(switch_status);
 CREATE TRIGGER tr_rcv_queue_insert
 AFTER INSERT ON rcv_queues
 FOR EACH ROW

src/Simplex/Messaging/Client.hs

@@ -67,6 +67,7 @@ module Simplex.Messaging.Client
     enableSMPQueuesNtfs,
     disableSMPQueuesNtfs,
     sendSMPMessage,
+    QueueDrained,
     ackSMPMessage,
     suspendSMPQueue,
     deleteSMPQueue,
@@ -195,6 +196,7 @@ data PClient v err msg = PClient
     transportHost :: TransportHost,
     tcpConnectTimeout :: NetworkTimeout,
     tcpTimeout :: NetworkTimeout,
+    proxiedRelayVRange :: VersionRange v,
     sendPings :: TVar Bool,
     lastReceived :: TVar UTCTime,
     timeoutErrorCount :: TVar Int,
@@ -240,6 +242,7 @@ smpClientStub g sessionId thVersion thAuth = do
               transportHost = "localhost",
               tcpConnectTimeout,
               tcpTimeout,
+              proxiedRelayVRange = supportedClientSMPRelayVRange,
               sendPings,
               lastReceived,
               timeoutErrorCount,
@@ -477,6 +480,7 @@ data ProtocolClientConfig v = ProtocolClientConfig
     serviceCredentials :: Maybe ServiceCredentials,
     -- | client-server protocol version range
     serverVRange :: VersionRange v,
+    proxiedRelayVRange :: VersionRange v,
     -- | agree shared session secret (used in SMP proxy for additional encryption layer)
     agreeSecret :: Bool,
     -- | Whether connecting client is a proxy server. See comment in ClientHandshake
@@ -495,6 +499,7 @@ defaultClientConfig clientALPN useSNI serverVRange =
       clientALPN,
       serviceCredentials = Nothing,
       serverVRange,
+      proxiedRelayVRange = serverVRange,
       agreeSecret = False,
       proxyServer = False,
       useSNI
@@ -505,6 +510,7 @@ defaultSMPClientConfig :: ProtocolClientConfig SMPVersion
 defaultSMPClientConfig =
   (defaultClientConfig (Just alpnSupportedSMPHandshakes) False supportedClientSMPRelayVRange)
     { defaultTransport = (show defaultSMPPort, transport @TLS),
+      proxiedRelayVRange = supportedClientSMPRelayVRange,
       agreeSecret = True
     }
 {-# INLINE defaultSMPClientConfig #-}
@@ -568,7 +574,7 @@ type SMPTransportSession = TransportSession BrokerMsg
 -- A single queue can be used for multiple 'SMPClient' instances,
 -- as 'SMPServerTransmission' includes server information.
 getProtocolClient :: forall v err msg. Protocol v err msg => TVar ChaChaDRG -> NetworkRequestMode -> TransportSession msg -> ProtocolClientConfig v -> [HostName] -> Maybe (TBQueue (ServerTransmissionBatch v err msg)) -> UTCTime -> (ProtocolClient v err msg -> IO ()) -> IO (Either (ProtocolClientError err) (ProtocolClient v err msg))
-getProtocolClient g nm transportSession@(_, srv, _) cfg@ProtocolClientConfig {qSize, networkConfig, clientALPN, serviceCredentials, serverVRange, agreeSecret, proxyServer, useSNI} presetDomains msgQ proxySessTs disconnected = do
+getProtocolClient g nm transportSession@(_, srv, _) cfg@ProtocolClientConfig {qSize, networkConfig, clientALPN, serviceCredentials, serverVRange, proxiedRelayVRange, agreeSecret, proxyServer, useSNI} presetDomains msgQ proxySessTs disconnected = do
   case chooseTransportHost networkConfig (host srv) of
     Right useHost ->
       (getCurrentTime >>= mkProtocolClient useHost >>= runClient useTransport useHost)
@@ -593,6 +599,7 @@ getProtocolClient g nm transportSession@(_, srv, _) cfg@ProtocolClientConfig {qS
             transportHost,
             tcpConnectTimeout,
             tcpTimeout,
+            proxiedRelayVRange,
             sendPings,
             lastReceived,
             timeoutErrorCount,
@@ -1081,15 +1088,17 @@ resolvedNameOrNotFound d NameResponse {registration} = case registration of
   NRRegistered {nameRecord} -> T.toLower (nrName nameRecord) == fullDomainName d
   _ -> True
 
+type QueueDrained = Bool
+
 -- | Acknowledge message delivery (server deletes the message).
 --
 -- https://github.com/simplex-chat/simplexmq/blob/master/protocol/simplex-messaging.md#acknowledge-message-delivery
 -- This command is always sent in background request mode
-ackSMPMessage :: SMPClient -> RcvPrivateAuthKey -> QueueId -> MsgId -> ExceptT SMPClientError IO ()
+ackSMPMessage :: SMPClient -> RcvPrivateAuthKey -> QueueId -> MsgId -> ExceptT SMPClientError IO QueueDrained
 ackSMPMessage c rpKey rId msgId =
   sendSMPCommand c NRMBackground (Just rpKey) rId (ACK msgId) >>= \case
-    OK -> return ()
-    cmd@MSG {} -> liftIO $ writeSMPMessage c rId cmd
+    OK -> pure True
+    cmd@MSG {} -> liftIO (writeSMPMessage c rId cmd) $> False
     r -> throwE $ unexpectedResponse r
 
 -- | Irreversibly suspend SMP queue.
@@ -1115,10 +1124,10 @@ deleteSMPQueues = okSMPCommands DEL
 -- send PRXY :: SMPServer -> Maybe BasicAuth -> Command Sender
 -- receives PKEY :: SessionId -> X.CertificateChain -> X.SignedExact X.PubKey -> BrokerMsg
 connectSMPProxiedRelay :: SMPClient -> NetworkRequestMode -> SMPServer -> Maybe BasicAuth -> ExceptT SMPClientError IO ProxiedRelay
-connectSMPProxiedRelay c@ProtocolClient {client_ = PClient {tcpConnectTimeout, tcpTimeout}} nm relayServ@ProtocolServer {port = relayPort, keyHash = C.KeyHash kh} proxyAuth =
+connectSMPProxiedRelay c@ProtocolClient {client_ = PClient {tcpConnectTimeout, tcpTimeout, proxiedRelayVRange}} nm relayServ@ProtocolServer {port = relayPort, keyHash = C.KeyHash kh} proxyAuth =
   sendProtocolCommand_ c nm Nothing tOut Nothing NoEntity (Cmd SProxiedClient (PRXY relayServ proxyAuth)) >>= \case
     PKEY sId vr (CertChainPubKey chain key) ->
-      case supportedClientSMPRelayVRange `compatibleVersion` vr of
+      case proxiedRelayVRange `compatibleVersion` vr of
         Nothing -> throwE $ transportErr TEVersion
         Just (Compatible v) -> do
           relayKey <- liftEitherWith (const $ transportErr $ TEHandshake IDENTITY) =<< liftIO (runExceptT $ validateRelay chain key)
@@ -1169,7 +1178,7 @@ instance StrEncoding ProxyClientError where
 -- consider how to process slow responses - is it handled somehow locally or delegated to the caller
 -- this method is used in the client
 -- sends PFWD :: C.PublicKeyX25519 -> EncTransmission -> Command Sender
--- receives PRES :: EncResponse -> BrokerMsg -- proxy to client
+-- receives PRES :: Maybe C.CbNonce -> EncResponse -> BrokerMsg -- proxy to client
 
 -- When client sends message via proxy, there may be one successful scenario and 9 error scenarios
 -- as shown below (WTF stands for unexpected response, ??? for response that failed to parse).
@@ -1222,20 +1231,20 @@ proxySMPCommand c@ProtocolClient {thParams = proxyThParams, client_ = PClient {c
   -- encode
   let TransmissionForAuth {tForAuth, tToSend} = encodeTransmissionForAuth serverThParams (CorrId corrId, sId, Cmd (sParty @p) command)
   -- serviceAuth is False here – proxied commands are not used with service certificates
-  auth <- liftEitherWith PCETransportError $ authTransmission serverThAuth False spKey nonce tForAuth
+  auth <- liftEitherWith PCETransportError $ authTransmission serverThAuth False spKey (Just nonce) tForAuth
   b <- case batchTransmissions serverThParams [Right (auth, tToSend)] of
     [] -> throwE $ PCETransportError TELargeMsg
     TBError e _ : _ -> throwE $ PCETransportError e
     TBTransmission s _ : _ -> pure s
     TBTransmissions s _ _ : _ -> pure s
-  et <- liftEitherWith PCECryptoError $ EncTransmission <$> C.cbEncrypt cmdSecret nonce b paddedProxiedTLength
+  et <- liftEitherWith PCECryptoError $ EncTransmission <$> C.cbEncrypt cmdSecret (encTransmissionNonce v nonce) b paddedProxiedTLength
   -- proxy interaction errors are wrapped
   let tOut = Just $ 2 * netTimeoutInt tcpTimeout nm
   tryE (sendProtocolCommand_ c nm (Just nonce) tOut Nothing (EntityId sessionId) (Cmd SProxiedClient (PFWD v cmdPubKey et))) >>= \case
     Right r -> case r of
-      PRES (EncResponse er) -> do
+      PRES nonce_ (EncResponse er) -> do
         -- server interaction errors are thrown directly
-        t' <- liftEitherWith PCECryptoError $ C.cbDecrypt cmdSecret (C.reverseNonce nonce) er
+        t' <- liftEitherWith PCECryptoError $ C.cbDecrypt cmdSecret (fromMaybe (C.reverseNonce nonce) nonce_) er
         case tParse serverThParams t' of
           t'' :| [] -> case tDecodeClient serverThParams t'' of
             (_, _, cmd) -> case cmd of
@@ -1253,10 +1262,10 @@ proxySMPCommand c@ProtocolClient {thParams = proxyThParams, client_ = PClient {c
 
 -- this method is used in the proxy
 -- sends RFWD :: EncFwdTransmission -> Command Sender
--- receives RRES :: EncFwdResponse -> BrokerMsg
+-- receives RRES :: Maybe C.CbNonce -> EncFwdResponse -> BrokerMsg
 -- proxy should send PRES to the client with EncResponse
 -- Always uses background timeout mode
-forwardSMPTransmission :: SMPClient -> CorrId -> VersionSMP -> C.PublicKeyX25519 -> EncTransmission -> ExceptT SMPClientError IO EncResponse
+forwardSMPTransmission :: SMPClient -> C.CorrCbNonce -> VersionSMP -> C.PublicKeyX25519 -> EncTransmission -> ExceptT SMPClientError IO (Maybe C.CbNonce, EncResponse)
 forwardSMPTransmission c@ProtocolClient {thParams, client_ = PClient {clientCorrId = g}} fwdCorrId fwdVersion fwdKey fwdTransmission = do
   -- prepare params
   sessSecret <- case thAuth thParams of
@@ -1268,11 +1277,11 @@ forwardSMPTransmission c@ProtocolClient {thParams, client_ = PClient {clientCorr
       eft = EncFwdTransmission $ C.cbEncryptNoPad sessSecret nonce (smpEncode fwdT)
   -- send
   sendProtocolCommand_ c NRMBackground (Just nonce) Nothing Nothing NoEntity (Cmd SProxyService (RFWD eft)) >>= \case
-    RRES (EncFwdResponse efr) -> do
+    RRES nonce_ (EncFwdResponse efr) -> do
       -- unwrap
       r' <- liftEitherWith PCECryptoError $ C.cbDecryptNoPad sessSecret (C.reverseNonce nonce) efr
       FwdResponse {fwdCorrId = _, fwdResponse} <- liftEitherWith (const $ PCEResponseError BLOCK) $ smpDecode r'
-      pure fwdResponse
+      pure (nonce_, fwdResponse)
     r -> throwE $ unexpectedResponse r
 
 -- get queue information - always sent interactively
@@ -1394,7 +1403,7 @@ mkTransmission_ :: forall v err msg. Protocol v err msg => ProtocolClient v err
 mkTransmission_ ProtocolClient {thParams, client_ = PClient {clientCorrId, sentCommands}} nonce_ (entityId, pKey_, command) = do
   nonce@(C.CbNonce corrId) <- maybe (atomically $ C.randomCbNonce clientCorrId) pure nonce_
   let TransmissionForAuth {tForAuth, tToSend} = encodeTransmissionForAuth thParams (CorrId corrId, entityId, command)
-      auth = authTransmission (thAuth thParams) (useServiceAuth command) pKey_ nonce tForAuth
+      auth = authTransmission (thAuth thParams) (useServiceAuth command) pKey_ (Just nonce) tForAuth
   r <- mkRequest (CorrId corrId)
   pure ((,tToSend) <$> auth, r)
   where
@@ -1413,14 +1422,14 @@ mkTransmission_ ProtocolClient {thParams, client_ = PClient {clientCorrId, sentC
       atomically $ TM.insert corrId r sentCommands
       pure r
 
-authTransmission :: Maybe (THandleAuth 'TClient) -> Bool -> Maybe C.APrivateAuthKey -> C.CbNonce -> ByteString -> Either TransportError (Maybe TAuthorizations)
-authTransmission thAuth serviceAuth pKey_ nonce t = traverse authenticate pKey_
+authTransmission :: Maybe (THandleAuth 'TClient) -> Bool -> Maybe C.APrivateAuthKey -> Maybe C.CbNonce -> ByteString -> Either TransportError (Maybe TAuthorizations)
+authTransmission thAuth serviceAuth pKey_ nonce_ t = traverse authenticate pKey_
   where
     authenticate :: C.APrivateAuthKey -> Either TransportError TAuthorizations
     authenticate (C.APrivateAuthKey a pk) = (,serviceSig) <$> case a of
-      C.SX25519 -> case thAuth of
-        Just THAuthClient {peerServerPubKey = k} -> Right $ TAAuthenticator $ C.cbAuthenticate k pk nonce t'
-        Nothing -> Left TENoServerAuth
+      C.SX25519 -> case (thAuth, nonce_) of
+        (Just THAuthClient {peerServerPubKey = k}, Just nonce) -> Right $ TAAuthenticator $ C.cbAuthenticate k pk nonce t'
+        _ -> Left TENoServerAuth
       C.SEd25519 -> sign pk
       C.SEd448 -> sign pk
     -- When command is signed by both entity key and service key,

src/Simplex/Messaging/Crypto.hs

@@ -140,8 +140,10 @@ module Simplex.Messaging.Crypto
     gcmIV,
 
     -- * NaCl crypto_box
-    CbNonce (unCbNonce),
+    CbNonceI (..),
+    CbNonce,
     pattern CbNonce,
+    CorrCbNonce,
     cbEncrypt,
     cbEncryptNoPad,
     cbEncryptMaxLenBS,
@@ -152,8 +154,12 @@ module Simplex.Messaging.Crypto
     sbEncryptNoPad,
     sbDecryptNoPad,
     cbNonce,
+    unsafeCbNonce,
+    corrCbNonce,
     randomCbNonce,
     reverseNonce,
+    xorNonce,
+    toCbNonce,
 
     -- * NaCl crypto_secretbox
     SbKey (unSbKey),
@@ -232,6 +238,7 @@ import Data.ASN1.Types
 import Data.Aeson (FromJSON (..), ToJSON (..))
 import qualified Data.Attoparsec.ByteString.Char8 as A
 import Data.Bifunctor (bimap, first)
+import Data.Bits (xor)
 import Data.ByteArray (ByteArrayAccess)
 import qualified Data.ByteArray as BA
 import Data.ByteString.Base64 (decode)
@@ -254,7 +261,7 @@ import Simplex.Messaging.Agent.Store.DB (Binary (..), FromField (..), ToField (.
 import Simplex.Messaging.Encoding
 import Simplex.Messaging.Encoding.String
 import Simplex.Messaging.Parsers (parseAll, parseString)
-import Simplex.Messaging.Util ((<$?>))
+import Simplex.Messaging.Util (packZipWith, (<$?>))
 
 -- | Cryptographic algorithms.
 data Algorithm = Ed25519 | Ed448 | X25519 | X448
@@ -931,6 +938,8 @@ data CryptoError
     CERatchetEarlierMessage Word32
   | -- | duplicate message number
     CERatchetDuplicateMessage
+  | -- | KEM key generation failed, indicating a broken RNG
+    CryptoKEMKeyGenError
   deriving (Eq, Show, Exception)
 
 aesKeySize :: Int
@@ -1045,9 +1054,7 @@ md5Hash = BA.convert . (hash :: ByteString -> Digest MD5)
 
 -- | AEAD-GCM encryption with associated data.
 --
--- Used as part of double ratchet encryption.
--- This function requires 16 bytes IV, it transforms IV in cryptonite_aes_gcm_init here:
--- https://github.com/haskell-crypto/cryptonite/blob/master/cbits/cryptonite_aes.c
+-- Used as part of double ratchet encryption, with a 16-byte IV (see @initAEAD@).
 encryptAEAD :: Key -> IV -> Int -> ByteString -> ByteString -> ExceptT CryptoError IO (AuthTag, ByteString)
 encryptAEAD aesKey ivBytes paddedLen ad msg = do
   aead <- initAEAD @AES256 aesKey ivBytes
@@ -1067,10 +1074,7 @@ encryptAEADNoPad aesKey ivBytes ad msg = do
 
 -- | AEAD-GCM decryption with associated data.
 --
--- Used as part of double ratchet encryption.
--- This function requires 16 bytes IV, it transforms IV in cryptonite_aes_gcm_init here:
--- https://github.com/haskell-crypto/cryptonite/blob/master/cbits/cryptonite_aes.c
--- To make it compatible with WebCrypto we will need to start using initAEADGCM.
+-- Used as part of double ratchet encryption, with a 16-byte IV (see @initAEAD@).
 decryptAEAD :: Key -> IV -> ByteString -> ByteString -> AuthTag -> ExceptT CryptoError IO ByteString
 decryptAEAD aesKey ivBytes ad msg (AuthTag authTag) = do
   aead <- initAEAD @AES256 aesKey ivBytes
@@ -1148,9 +1152,9 @@ maxLength :: forall i. KnownNat i => Int
 maxLength = fromIntegral (natVal $ Proxy @i)
 {-# INLINE maxLength #-}
 
--- this function requires 16 bytes IV, it transforms IV in cryptonite_aes_gcm_init here:
--- https://github.com/haskell-crypto/cryptonite/blob/master/cbits/cryptonite_aes.c
--- This is used for double ratchet encryption, so to make it compatible with WebCrypto we will need to deprecate it and start using initAEADGCM
+-- The 16-byte double ratchet IV is intentionally not the 96-bit IV recommended by NIST SP 800-38D, so GCM derives J0 = GHASH(IV || 0^64 || [128]_64),
+-- as in crypton_aes_gcm_init: https://hackage.haskell.org/package/crypton-0.34/src/cbits/crypton_aes.c
+-- WebCrypto and other SP 800-38D implementations interoperate only when given all 16 IV bytes.
 initAEAD :: forall c. AES.BlockCipher c => Key -> IV -> ExceptT CryptoError IO (AES.AEAD c)
 initAEAD (Key aesKey) (IV ivBytes) = do
   iv <- makeIV @c ivBytes
@@ -1327,8 +1331,8 @@ sbEncryptNoPad (SbKey key) (CbNonce nonce) = cryptoBox key nonce
 {-# INLINE sbEncryptNoPad #-}
 
 -- | NaCl @crypto_box@ encrypt with a shared DH secret and 192-bit nonce.
-cbEncryptMaxLenBS :: KnownNat i => DhSecret X25519 -> CbNonce -> MaxLenBS i -> ByteString
-cbEncryptMaxLenBS (DhSecretX25519 secret) (CbNonce nonce) = cryptoBox secret nonce . unMaxLenBS . padMaxLenBS
+cbEncryptMaxLenBS :: (KnownNat i, CbNonceI n) => DhSecret X25519 -> n -> MaxLenBS i -> ByteString
+cbEncryptMaxLenBS (DhSecretX25519 secret) nonce = cryptoBox secret (unCbNonce nonce) . unMaxLenBS . padMaxLenBS
 {-# INLINE cbEncryptMaxLenBS #-}
 
 cryptoBox :: ByteArrayAccess key => key -> ByteString -> ByteString -> ByteString
@@ -1338,8 +1342,8 @@ cryptoBox secret nonce s = BA.convert tag <> c
     tag = Poly1305.auth rs c
 
 -- | NaCl @crypto_box@ decrypt with a shared DH secret and 192-bit nonce.
-cbDecrypt :: DhSecret X25519 -> CbNonce -> ByteString -> Either CryptoError ByteString
-cbDecrypt (DhSecretX25519 secret) = sbDecrypt_ secret
+cbDecrypt :: CbNonceI n => DhSecret X25519 -> n -> ByteString -> Either CryptoError ByteString
+cbDecrypt (DhSecretX25519 secret) = sbDecrypt_ secret . toCbNonce
 {-# INLINE cbDecrypt #-}
 
 -- | NaCl @crypto_box@ decrypt with a shared DH secret and 192-bit nonce (without unpadding).
@@ -1386,9 +1390,16 @@ cbAuthenticate k pk nonce msg = CbAuthenticator $ cbEncryptNoPad (dh' k pk) nonc
 cbVerify :: PublicKeyX25519 -> PrivateKeyX25519 -> CbNonce -> CbAuthenticator -> ByteString -> Bool
 cbVerify k pk nonce (CbAuthenticator s) authorized = cbDecryptNoPad (dh' k pk) nonce s == Right (sha512Hash authorized)
 
-newtype CbNonce = CryptoBoxNonce {unCbNonce :: ByteString}
+class CbNonceI n where
+  unCbNonce :: n -> ByteString
+
+newtype CbNonce = CryptoBoxNonce ByteString
   deriving (Eq, Show)
-  deriving newtype (FromField)
+
+instance CbNonceI CbNonce where
+  unCbNonce (CryptoBoxNonce s) = s
+
+instance FromField CbNonce where fromField = blobFieldDecoder cbNonce
 
 instance ToField CbNonce where toField (CryptoBoxNonce s) = toField $ Binary s
 
@@ -1399,7 +1410,7 @@ pattern CbNonce s <- CryptoBoxNonce s
 
 instance StrEncoding CbNonce where
   strEncode (CbNonce s) = strEncode s
-  strP = cbNonce <$> strP
+  strP = cbNonce <$?> strP
 
 instance ToJSON CbNonce where
   toJSON = strToJSON
@@ -1408,13 +1419,16 @@ instance ToJSON CbNonce where
 instance FromJSON CbNonce where
   parseJSON = strParseJSON "CbNonce"
 
-cbNonce :: ByteString -> CbNonce
+cbNonce :: ByteString -> Either String CbNonce
 cbNonce s
-  | len == 24 = CryptoBoxNonce s
-  | len > 24 = CryptoBoxNonce . fst $ B.splitAt 24 s
-  | otherwise = CryptoBoxNonce $ s <> B.replicate (24 - len) (toEnum 0)
-  where
-    len = B.length s
+  | B.length s == 24 = Right $ CryptoBoxNonce s
+  | otherwise = Left "CbNonce: invalid length"
+
+unsafeCbNonce :: ByteString -> CbNonce
+unsafeCbNonce s = either error id $ cbNonce s
+
+corrCbNonce :: ByteString -> Either String CorrCbNonce
+corrCbNonce s = CorrCbNonce . unCbNonce <$> cbNonce s
 
 randomCbNonce :: TVar ChaChaDRG -> STM CbNonce
 randomCbNonce = fmap CryptoBoxNonce . randomBytes 24
@@ -1422,13 +1436,30 @@ randomCbNonce = fmap CryptoBoxNonce . randomBytes 24
 randomBytes :: Int -> TVar ChaChaDRG -> STM ByteString
 randomBytes n gVar = stateTVar gVar $ randomBytesGenerate n
 
-reverseNonce :: CbNonce -> CbNonce
-reverseNonce (CryptoBoxNonce s) = CryptoBoxNonce (B.reverse s)
+reverseNonce :: CbNonceI n => n -> CbNonce
+reverseNonce = CryptoBoxNonce . B.reverse . unCbNonce
+
+xorNonce :: CbNonceI n => ByteString -> n -> CbNonce
+xorNonce s nonce = CryptoBoxNonce $ packZipWith xor s n <> B.drop (B.length s) n
+  where
+    n = unCbNonce nonce
+
+toCbNonce :: CbNonceI n => n -> CbNonce
+toCbNonce = CryptoBoxNonce . unCbNonce
 
 instance Encoding CbNonce where
   smpEncode = unCbNonce
   smpP = CryptoBoxNonce <$> A.take 24
 
+newtype CorrCbNonce = CorrCbNonce ByteString
+
+instance CbNonceI CorrCbNonce where
+  unCbNonce (CorrCbNonce s) = s
+
+instance Encoding CorrCbNonce where
+  smpEncode = smpEncode . unCbNonce
+  smpP = corrCbNonce <$?> smpP
+
 newtype SbKey = SecretBoxKey {unSbKey :: ByteString}
   deriving (Eq, Show)
   deriving newtype (FromField)

src/Simplex/Messaging/Crypto/Ratchet.hs

@@ -72,6 +72,7 @@ module Simplex.Messaging.Crypto.Ratchet
     rcEncryptHeader,
     rcEncryptMsg,
     rcDecrypt,
+    maxSkippedMsgKeys,
     -- used in tests
     MsgHeader (..),
     RatchetInitParams (..),
@@ -119,7 +120,7 @@ import Simplex.Messaging.Crypto.SNTRUP761.Bindings
 import Simplex.Messaging.Encoding
 import Simplex.Messaging.Encoding.String
 import Simplex.Messaging.Parsers (defaultJSON, parseE, parseE')
-import Simplex.Messaging.Util (($>>=), (<$?>))
+import Simplex.Messaging.Util ((<$?>))
 import Simplex.Messaging.Version
 import Simplex.Messaging.Version.Internal
 import UnliftIO.STM
@@ -327,9 +328,13 @@ instance StrEncoding AnyE2ERatchetParamsUri where
             Nothing -> pure $ AnyE2ERatchetParamsUri SRKSProposed a $ E2ERatchetParamsUri vr k1 k2 Nothing
       _ -> fail "bad e2e params"
     where
-      kemP query =
-        queryParam_ "kem_key" query
-          $>>= \k -> Just . kemParams k <$> queryParam_ "kem_ct" query
+      kemP query = do
+        k_ <- queryParam_ "kem_key" query
+        ct_ <- queryParam_ "kem_ct" query
+        case (k_, ct_) of
+          (Just k, _) -> pure $ Just $ kemParams k ct_
+          (Nothing, Nothing) -> pure Nothing
+          (Nothing, Just _) -> fail "bad e2e params: kem_ct without kem_key"
       kemParams k = \case
         Nothing -> ARKP SRKSProposed $ RKParamsProposed k
         Just ct -> ARKP SRKSAccepted $ RKParamsAccepted ct k
@@ -451,6 +456,7 @@ generateSndE2EParams g v = \case
 
 data RatchetInitParams = RatchetInitParams
   { assocData :: Str,
+    rcVerifyCodePQ :: Str,
     ratchetKey :: RatchetKey,
     sndHK :: HeaderKey,
     rcvNextHK :: HeaderKey,
@@ -493,14 +499,14 @@ pqX3dhRcv (rpk1, rpk2, rpKem_) (E2ERatchetParams _ sk1 sk2 sKem_) = do
 
 pqX3dh :: DhAlgorithm a => (PublicKey a, PublicKey a) -> DhSecret a -> DhSecret a -> DhSecret a -> Maybe RatchetKEMAccepted -> RatchetInitParams
 pqX3dh (sk1, rk1) dh1 dh2 dh3 kemAccepted =
-  RatchetInitParams {assocData, ratchetKey = RatchetKey sk, sndHK = Key hk, rcvNextHK = Key nhk, kemAccepted}
+  RatchetInitParams {assocData, rcVerifyCodePQ = Str vcPQ, ratchetKey = RatchetKey sk, sndHK = Key hk, rcvNextHK = Key nhk, kemAccepted}
   where
     assocData = Str $ pubKeyBytes sk1 <> pubKeyBytes rk1
     dhs = dhBytes' dh1 <> dhBytes' dh2 <> dhBytes' dh3 <> pq
     pq = maybe "" (\RatchetKEMAccepted {rcPQRss = KEMSharedKey ss} -> BA.convert ss) kemAccepted
-    (hk, nhk, sk) =
-      let salt = B.replicate 64 '\0'
-       in hkdf3 salt dhs "SimpleXX3DH"
+    salt = B.replicate 64 '\0'
+    (hk, nhk, sk) = hkdf3 salt dhs "SimpleXX3DH"
+    vcPQ = hkdf salt dhs "SimpleXVerifyCode" 32
 
 type RatchetX448 = Ratchet 'X448
 
@@ -509,6 +515,8 @@ data Ratchet a = Ratchet
     rcVersion :: RatchetVersions,
     -- associated data - must be the same in both parties ratchets
     rcAD :: Str,
+    -- verification code covering all handshake keys, absent in the ratchets created before it was added
+    rcVCPQ :: Maybe Str,
     rcDHRs :: PrivateKey a,
     rcKEM :: Maybe RatchetKEM,
     rcSupportKEM :: PQSupport, -- defines header size, can only be enabled once
@@ -637,13 +645,14 @@ instance FromField MessageKey where fromField = blobFieldDecoder smpDecode
 -- @
 initSndRatchet ::
   forall a. (AlgorithmI a, DhAlgorithm a) => RatchetVersions -> PublicKey a -> PrivateKey a -> (RatchetInitParams, Maybe KEMKeyPair) -> Ratchet a
-initSndRatchet rcVersion rcDHRr rcDHRs (RatchetInitParams {assocData, ratchetKey, sndHK, rcvNextHK, kemAccepted}, rcPQRs_) = do
+initSndRatchet rcVersion rcDHRr rcDHRs (RatchetInitParams {assocData, rcVerifyCodePQ, ratchetKey, sndHK, rcvNextHK, kemAccepted}, rcPQRs_) = do
   -- state.RK, state.CKs, state.NHKs = KDF_RK_HE(SK, DH(state.DHRs, state.DHRr) || state.PQRss)
   let (rcRK, rcCKs, rcNHKs) = rootKdf ratchetKey rcDHRr rcDHRs (rcPQRss <$> kemAccepted)
       pqOn = isJust rcPQRs_
    in Ratchet
         { rcVersion,
           rcAD = assocData,
+          rcVCPQ = Just rcVerifyCodePQ,
           rcDHRs,
           rcKEM = (`RatchetKEM` kemAccepted) <$> rcPQRs_,
           rcSupportKEM = PQSupport pqOn,
@@ -668,10 +677,11 @@ initSndRatchet rcVersion rcDHRr rcDHRs (RatchetInitParams {assocData, ratchetKey
 -- as part of the connection request and random salt was received from the sender.
 initRcvRatchet ::
   forall a. (AlgorithmI a, DhAlgorithm a) => RatchetVersions -> PrivateKey a -> (RatchetInitParams, Maybe KEMKeyPair) -> PQSupport -> Ratchet a
-initRcvRatchet rcVersion rcDHRs (RatchetInitParams {assocData, ratchetKey, sndHK, rcvNextHK, kemAccepted}, rcPQRs_) pqSupport =
+initRcvRatchet rcVersion rcDHRs (RatchetInitParams {assocData, rcVerifyCodePQ, ratchetKey, sndHK, rcvNextHK, kemAccepted}, rcPQRs_) pqSupport =
   Ratchet
     { rcVersion,
       rcAD = assocData,
+      rcVCPQ = Just rcVerifyCodePQ,
       rcDHRs,
       -- rcKEM:
       -- state.PQRs = bob_pq_kem_key_pair
@@ -953,6 +963,9 @@ type DecryptResult a = (Either CryptoError ByteString, Ratchet a, SkippedMsgDiff
 maxSkip :: Word32
 maxSkip = 512
 
+maxSkippedMsgKeys :: Int
+maxSkippedMsgKeys = 2000
+
 rcDecrypt ::
   forall a.
   (AlgorithmI a, DhAlgorithm a) =>

src/Simplex/Messaging/Crypto/SNTRUP761/Bindings.hs

@@ -16,6 +16,8 @@ module Simplex.Messaging.Crypto.SNTRUP761.Bindings
   ) where
 
 import Control.Concurrent.STM
+import Control.Exception (throwIO)
+import Control.Monad (when)
 import Crypto.Random (ChaChaDRG)
 import Data.Aeson (FromJSON (..), ToJSON (..))
 import Data.Bifunctor (bimap)
@@ -23,6 +25,7 @@ import Data.ByteArray (ScrubbedBytes)
 import qualified Data.ByteArray as BA
 import Data.ByteString (ByteString)
 import Simplex.Messaging.Agent.Store.DB (FromField (..), ToField (..))
+import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings.Defines
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings.FFI
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings.RNG (rngFuncPtr, withDRG)
@@ -55,14 +58,13 @@ pattern KEMSharedKey s <- KEMSharedKey_ s
 type KEMKeyPair = (KEMPublicKey, KEMSecretKey)
 
 sntrup761Keypair :: TVar ChaChaDRG -> IO KEMKeyPair
-sntrup761Keypair drg =
-  bimap KEMPublicKey_ KEMSecretKey
-    <$> BA.allocRet
-      c_SNTRUP761_SECRETKEY_SIZE
-      ( \skPtr ->
-          BA.alloc c_SNTRUP761_PUBLICKEY_SIZE $ \pkPtr ->
-            withDRG drg $ \cxtPtr -> c_sntrup761_keypair pkPtr skPtr cxtPtr rngFuncPtr
-      )
+sntrup761Keypair drg = do
+  ((r, pk), sk) <-
+    BA.allocRet c_SNTRUP761_SECRETKEY_SIZE $ \skPtr ->
+      BA.allocRet c_SNTRUP761_PUBLICKEY_SIZE $ \pkPtr ->
+        withDRG drg $ \cxtPtr -> c_sntrup761_keypair pkPtr skPtr cxtPtr rngFuncPtr
+  when (r /= 0) $ throwIO C.CryptoKEMKeyGenError
+  pure (KEMPublicKey_ pk, KEMSecretKey sk)
 
 sntrup761Enc :: TVar ChaChaDRG -> KEMPublicKey -> IO (KEMCiphertext, KEMSharedKey)
 sntrup761Enc drg (KEMPublicKey pk) =

src/Simplex/Messaging/Crypto/SNTRUP761/Bindings/FFI.hs

@@ -10,9 +10,9 @@ import Foreign
 import Foreign.C
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings.RNG (RNGContext, RNGFunc)
 
--- void sntrup761_keypair (uint8_t *pk, uint8_t *sk, void *random_ctx, sntrup761_random_func *random);
+-- int sntrup761_keypair (uint8_t *pk, uint8_t *sk, void *random_ctx, sntrup761_random_func *random);
 foreign import ccall "sntrup761_keypair"
-  c_sntrup761_keypair :: Ptr Word8 -> Ptr Word8 -> Ptr RNGContext -> FunPtr RNGFunc -> IO ()
+  c_sntrup761_keypair :: Ptr Word8 -> Ptr Word8 -> Ptr RNGContext -> FunPtr RNGFunc -> IO CInt
 
 -- void sntrup761_enc (uint8_t *c, uint8_t *k, const uint8_t *pk, void *random_ctx, sntrup761_random_func *random);
 foreign import ccall "sntrup761_enc"

src/Simplex/Messaging/Crypto/ShortLink.hs

@@ -17,6 +17,7 @@ module Simplex.Messaging.Crypto.ShortLink
     encodeSignUserData,
     newOwnerAuth,
     encryptLinkData,
+    encryptFixedData,
     encryptUserData,
     decryptLinkData,
   )
@@ -84,9 +85,10 @@ newOwnerAuth g ownerId signingKey = do
   pure (ownerPrivKey, OwnerAuth {ownerId, ownerKey, authOwnerSig})
 
 encryptLinkData :: TVar ChaChaDRG -> C.SbKey -> (ByteString, ByteString) -> ExceptT AgentErrorType IO QueueLinkData
-encryptLinkData g k = bimapM (encrypt fixedDataPaddedLength) (encrypt userDataPaddedLength)
-  where
-    encrypt len = encryptData g k len
+encryptLinkData g k = bimapM (encryptFixedData g k) (encryptUserData g k)
+
+encryptFixedData :: TVar ChaChaDRG -> C.SbKey -> ByteString -> ExceptT AgentErrorType IO EncDataBytes
+encryptFixedData g k s = encryptData g k fixedDataPaddedLength s
 
 encryptUserData :: TVar ChaChaDRG -> C.SbKey -> ByteString -> ExceptT AgentErrorType IO EncDataBytes
 encryptUserData g k s = encryptData g k userDataPaddedLength s

src/Simplex/Messaging/Notifications/Server/Store/ntf_server_schema.sql

@@ -1,5 +1,6 @@
 
 
+
 SET statement_timeout = 0;
 SET lock_timeout = 0;
 SET idle_in_transaction_session_timeout = 0;
@@ -312,3 +313,4 @@ ALTER TABLE ONLY ntf_server.subscriptions
 
 
 
+

src/Simplex/Messaging/Protocol.hs

@@ -145,7 +145,10 @@ module Simplex.Messaging.Protocol
     RcvNtfDhSecret,
     Message (..),
     RcvMessage (..),
-    MsgId,
+    MsgId (unMsgId),
+    mkMsgId,
+    unsafeMsgId,
+    randomMsgId,
     MsgBody,
     IdsHash (..),
     ServiceSub (..),
@@ -168,6 +171,7 @@ module Simplex.Messaging.Protocol
     EncFwdTransmission (..),
     EncResponse (..),
     EncTransmission (..),
+    encTransmissionNonce,
     FwdResponse (..),
     FwdTransmission (..),
     NameRecord (..),
@@ -232,8 +236,10 @@ module Simplex.Messaging.Protocol
 where
 
 import Control.Applicative (optional, (<|>))
+import Control.Concurrent.STM (STM, TVar)
 import Control.Exception (Exception, SomeException, displayException, fromException)
 import Control.Monad.Except
+import Crypto.Random (ChaChaDRG)
 import Data.Aeson (FromJSON (..), ToJSON (..))
 import qualified Data.Aeson as J
 import qualified Data.Aeson.TH as J
@@ -241,6 +247,7 @@ import Data.Attoparsec.ByteString.Char8 (Parser, (<?>))
 import qualified Data.Attoparsec.ByteString.Char8 as A
 import Data.Bifunctor (bimap, first)
 import Data.Bits (xor)
+import qualified Data.ByteArray as BA
 import qualified Data.ByteString as BS
 import qualified Data.ByteString.Base64 as B64
 import Data.ByteString.Char8 (ByteString)
@@ -267,7 +274,7 @@ import qualified GHC.TypeLits as TE
 import qualified GHC.TypeLits as Type
 import Network.Socket (ServiceName)
 import qualified Network.TLS as TLS
-import Simplex.Messaging.Agent.Store.DB (Binary (..), FromField (..), ToField (..))
+import Simplex.Messaging.Agent.Store.DB (Binary (..), FromField (..), ToField (..), blobFieldDecoder)
 import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Encoding
 import Simplex.Messaging.Encoding.String
@@ -701,19 +708,24 @@ instance Encoding NewNtfCreds where
 newtype EncTransmission = EncTransmission ByteString
   deriving (Show)
 
+encTransmissionNonce :: C.CbNonceI n => VersionSMP -> n -> C.CbNonce
+encTransmissionNonce v nonce
+  | v >= fwdNoncesSMPVersion = C.xorNonce (smpEncode v) nonce
+  | otherwise = C.toCbNonce nonce
+
 data FwdTransmission = FwdTransmission
-  { fwdCorrId :: CorrId,
+  { fwdCorrId :: C.CorrCbNonce,
     fwdVersion :: VersionSMP,
     fwdKey :: C.PublicKeyX25519,
     fwdTransmission :: EncTransmission
   }
 
 instance Encoding FwdTransmission where
-  smpEncode FwdTransmission {fwdCorrId = CorrId corrId, fwdVersion, fwdKey, fwdTransmission = EncTransmission t} =
-    smpEncode (corrId, fwdVersion, fwdKey, Tail t)
+  smpEncode FwdTransmission {fwdCorrId, fwdVersion, fwdKey, fwdTransmission = EncTransmission t} =
+    smpEncode (fwdCorrId, fwdVersion, fwdKey, Tail t)
   smpP = do
-    (corrId, fwdVersion, fwdKey, Tail t) <- smpP
-    pure FwdTransmission {fwdCorrId = CorrId corrId, fwdVersion, fwdKey, fwdTransmission = EncTransmission t}
+    (fwdCorrId, fwdVersion, fwdKey, Tail t) <- smpP
+    pure FwdTransmission {fwdCorrId, fwdVersion, fwdKey, fwdTransmission = EncTransmission t}
 
 newtype EncFwdTransmission = EncFwdTransmission ByteString
   deriving (Show)
@@ -736,8 +748,8 @@ data BrokerMsg where
   NMSG :: C.CbNonce -> EncNMsgMeta -> BrokerMsg
   -- Should include certificate chain
   PKEY :: SessionId -> VersionRangeSMP -> CertChainPubKey -> BrokerMsg -- TLS-signed server key for proxy shared secret and initial sender key
-  RRES :: EncFwdResponse -> BrokerMsg -- relay to proxy
-  PRES :: EncResponse -> BrokerMsg -- proxy to client
+  RRES :: Maybe C.CbNonce -> EncFwdResponse -> BrokerMsg -- relay to proxy
+  PRES :: Maybe C.CbNonce -> EncResponse -> BrokerMsg -- proxy to client
   END :: BrokerMsg
   ENDS :: Int64 -> IdsHash -> BrokerMsg
   DELD :: BrokerMsg
@@ -759,16 +771,16 @@ newtype EncFwdResponse = EncFwdResponse ByteString
   deriving (Eq, Show)
 
 data FwdResponse = FwdResponse
-  { fwdCorrId :: CorrId,
+  { fwdCorrId :: C.CorrCbNonce,
     fwdResponse :: EncResponse
   }
 
 instance Encoding FwdResponse where
-  smpEncode FwdResponse {fwdCorrId = CorrId corrId, fwdResponse = EncResponse t} =
-    smpEncode (corrId, Tail t)
+  smpEncode FwdResponse {fwdCorrId, fwdResponse = EncResponse t} =
+    smpEncode (fwdCorrId, Tail t)
   smpP = do
-    (corrId, Tail t) <- smpP
-    pure FwdResponse {fwdCorrId = CorrId corrId, fwdResponse = EncResponse t}
+    (fwdCorrId, Tail t) <- smpP
+    pure FwdResponse {fwdCorrId, fwdResponse = EncResponse t}
 
 newtype EncResponse = EncResponse ByteString
   deriving (Eq, Show)
@@ -791,7 +803,7 @@ toMsgInfo = \case
   Message {msgId, msgTs} -> msgInfo msgId msgTs MTMessage
   MessageQuota {msgId, msgTs} -> msgInfo msgId msgTs MTQuota
   where
-    msgInfo msgId msgTs msgType = MsgInfo {msgId = decodeLatin1 $ B64.encode msgId, msgTs = systemToUTCTime msgTs, msgType}
+    msgInfo msgId msgTs msgType = MsgInfo {msgId = decodeLatin1 $ B64.encode $ unMsgId msgId, msgTs = systemToUTCTime msgTs, msgType}
 
 messageId :: Message -> MsgId
 messageId = \case
@@ -1323,7 +1335,10 @@ instance ProtocolTypeI p => FromJSON (ProtocolServer p) where
   parseJSON = strParseJSON "ProtocolServer"
 
 newtype BasicAuth = BasicAuth {unBasicAuth :: ByteString}
-  deriving (Eq, Ord, Show)
+  deriving (Ord, Show)
+
+instance Eq BasicAuth where
+  BasicAuth s == BasicAuth s' = BA.constEq s s'
 
 instance IsString BasicAuth where fromString = BasicAuth . B.pack
 
@@ -1489,7 +1504,34 @@ type RcvNtfPublicDhKey = C.PublicKeyX25519
 type RcvNtfDhSecret = C.DhSecretX25519
 
 -- | SMP message server ID.
-type MsgId = ByteString
+newtype MsgId = MsgId {unMsgId :: ByteString}
+  deriving (Eq, Show)
+
+instance C.CbNonceI MsgId where
+  unCbNonce (MsgId s) = B.take 24 s
+
+instance Encoding MsgId where
+  smpEncode (MsgId s) = smpEncode s
+  smpP = mkMsgId <$?> smpP
+
+instance StrEncoding MsgId where
+  strEncode (MsgId s) = strEncode s
+  strP = mkMsgId <$?> strP
+
+instance FromField MsgId where fromField = blobFieldDecoder mkMsgId
+
+instance ToField MsgId where toField (MsgId s) = toField $ Binary s
+
+mkMsgId :: ByteString -> Either String MsgId
+mkMsgId s
+  | B.length s >= 24 = Right $ MsgId s
+  | otherwise = Left "MsgId: invalid length"
+
+unsafeMsgId :: ByteString -> MsgId
+unsafeMsgId s = either error id $ mkMsgId s
+
+randomMsgId :: Int -> TVar ChaChaDRG -> STM MsgId
+randomMsgId n = fmap MsgId . C.randomBytes (max 24 n)
 
 -- | SMP message body.
 type MsgBody = ByteString
@@ -1814,8 +1856,7 @@ instance PartyI p => ProtocolEncoding SMPVersion ErrorType (Command p) where
   encodeProtocol v = \case
     NEW NewQueueReq {rcvAuthKey = rKey, rcvDhKey = dhKey, auth_, subMode, queueReqData, ntfCreds}
       | v >= newNtfCredsSMPVersion -> new <> e (subMode, queueReqData, ntfCreds)
-      | v >= shortLinksSMPVersion -> new <> e (subMode, queueReqData)
-      | otherwise -> new <> e (subMode, senderCanSecure (queueReqMode <$> queueReqData))
+      | otherwise -> new <> e (subMode, queueReqData)
       where
         new = e (NEW_, ' ', rKey, dhKey, auth_)
     SUB -> e SUB_
@@ -1902,20 +1943,18 @@ instance ProtocolEncoding SMPVersion ErrorType Cmd where
     CT SCreator NEW_ -> Cmd SCreator <$> newCmd
       where
         newCmd
-          | v >= newNtfCredsSMPVersion = new smpP smpP
-          | v >= shortLinksSMPVersion = new smpP nothing
-          | otherwise = new (qReq <$> smpP) nothing
+          | v >= newNtfCredsSMPVersion = new smpP
+          | otherwise = new nothing
           where
             nothing = pure Nothing
-            new p2 p3 = NEW <$> do
+            new p3 = NEW <$> do
               rcvAuthKey <- _smpP
               rcvDhKey <- smpP
               auth_ <- smpP
               subMode <- smpP
-              queueReqData <- p2
+              queueReqData <- smpP
               ntfCreds <- p3
               pure NewQueueReq {rcvAuthKey, rcvDhKey, auth_, subMode, queueReqData, ntfCreds}
-            qReq sndSecure = Just $ if sndSecure then QRMessaging Nothing else QRContact Nothing
     CT SRecipient tag ->
       Cmd SRecipient <$> case tag of
         SUB_ -> pure SUB
@@ -1966,8 +2005,7 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
     IDS QIK {rcvId, sndId, rcvPublicDhKey = srvDh, queueMode, linkId, serviceId, serverNtfCreds}
       | v >= newNtfCredsSMPVersion -> ids <> e (queueMode, linkId, serviceId, serverNtfCreds)
       | v >= serviceCertsSMPVersion -> ids <> e (queueMode, linkId, serviceId)
-      | v >= shortLinksSMPVersion -> ids <> e (queueMode, linkId)
-      | otherwise -> ids <> e (senderCanSecure queueMode)
+      | otherwise -> ids <> e (queueMode, linkId)
       where
         ids = e (IDS_, ' ', rcvId, sndId, srvDh)
     LNK sId d -> e (LNK_, ' ', sId, d)
@@ -1981,8 +2019,8 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
     NID nId srvNtfDh -> e (NID_, ' ', nId, srvNtfDh)
     NMSG nmsgNonce encNMsgMeta -> e (NMSG_, ' ', nmsgNonce, encNMsgMeta)
     PKEY sid vr certKey -> e (PKEY_, ' ', sid, vr, certKey)
-    RRES (EncFwdResponse encBlock) -> e (RRES_, ' ', Tail encBlock)
-    PRES (EncResponse encBlock) -> e (PRES_, ' ', Tail encBlock)
+    RRES nonce_ (EncFwdResponse encBlock) -> fwdResp RRES_ nonce_ encBlock
+    PRES nonce_ (EncResponse encBlock) -> fwdResp PRES_ nonce_ encBlock
     END -> e END_
     ENDS n idsHash -> serviceResp ENDS_ n idsHash
     DELD -> e DELD_
@@ -2006,6 +2044,9 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
       serviceResp tag n idsHash
         | v >= rcvServiceSMPVersion = e (tag, ' ', n, idsHash)
         | otherwise = e (tag, ' ', n)
+      fwdResp tag nonce_ encBlock
+        | v >= fwdNoncesSMPVersion = e (tag, ' ', nonce_, Tail encBlock)
+        | otherwise = e (tag, ' ', Tail encBlock)
 
   protocolP v = \case
     MSG_ -> do
@@ -2015,19 +2056,17 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
         bodyP = EncRcvMsgBody . unTail <$> smpP
     ALLS_ -> pure ALLS
     IDS_
-      | v >= newNtfCredsSMPVersion -> ids smpP smpP smpP smpP
-      | v >= serviceCertsSMPVersion -> ids smpP smpP smpP nothing
-      | v >= shortLinksSMPVersion -> ids smpP smpP nothing nothing
-      | otherwise -> ids (qm <$> smpP) nothing nothing nothing
+      | v >= newNtfCredsSMPVersion -> ids smpP smpP
+      | v >= serviceCertsSMPVersion -> ids smpP nothing
+      | otherwise -> ids nothing nothing
       where
-        qm sndSecure = Just $ if sndSecure then QMMessaging else QMContact
         nothing = pure Nothing
-        ids p1 p2 p3 p4 = do
+        ids p3 p4 = do
           rcvId <- _smpP
           sndId <- smpP
           rcvPublicDhKey <- smpP
-          queueMode <- p1
-          linkId <- p2
+          queueMode <- smpP
+          linkId <- smpP
           serviceId <- p3
           serverNtfCreds <- p4
           pure $ IDS QIK {rcvId, sndId, rcvPublicDhKey, queueMode, linkId, serviceId, serverNtfCreds}
@@ -2037,8 +2076,8 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
     NID_ -> NID <$> _smpP <*> smpP
     NMSG_ -> NMSG <$> _smpP <*> smpP
     PKEY_ -> PKEY <$> _smpP <*> smpP <*> smpP
-    RRES_ -> RRES <$> (EncFwdResponse . unTail <$> _smpP)
-    PRES_ -> PRES <$> (EncResponse . unTail <$> _smpP)
+    RRES_ -> fwdRespP RRES EncFwdResponse
+    PRES_ -> fwdRespP PRES EncResponse
     END_ -> pure END
     ENDS_ -> serviceRespP ENDS
     DELD_ -> pure DELD
@@ -2055,6 +2094,9 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
       serviceRespP resp
         | v >= rcvServiceSMPVersion = resp <$> _smpP <*> smpP
         | otherwise = resp <$> _smpP <*> pure mempty
+      fwdRespP :: (Maybe C.CbNonce -> a -> BrokerMsg) -> (ByteString -> a) -> Parser BrokerMsg
+      fwdRespP resp enc = resp <$> (A.space *> nonceP) <*> (enc <$> A.takeByteString)
+      nonceP = if v >= fwdNoncesSMPVersion then smpP else pure Nothing
 
   fromProtocolError = \case
     PECmdSyntax -> CMD SYNTAX
@@ -2071,7 +2113,7 @@ instance ProtocolEncoding SMPVersion ErrorType BrokerMsg where
     -- PONG response must not have queue ID
     PONG -> noEntityMsg
     PKEY {} -> noEntityMsg
-    RRES _ -> noEntityMsg
+    RRES {} -> noEntityMsg
     ALLS -> noEntityMsg
     RNAME {} -> noEntityMsg
     -- other broker responses must have queue ID
@@ -2407,7 +2449,9 @@ tDecodeServer THandleParams {sessionId, thVersion = v, implySessId} = \case
           where
             cmdOrErr = parseProtocol @v @err @cmd v command >>= checkCredentials tAuth entityId
             t :: a -> (CorrId, EntityId, a)
-            t = (corrId,entityId,)
+            -- IDs are slices of the ~16 KB received block and are kept as subscription keys,
+            -- so without a copy one live key retains the whole block
+            t = (CorrId $ B.copy $ bs corrId,EntityId $ B.copy $ unEntityId entityId,)
         Left _ -> tError corrId PEBlock
     | otherwise -> tError corrId PESession
   Left _ -> tError "" PEBlock

src/Simplex/Messaging/Server.hs

@@ -1092,6 +1092,7 @@ controlPortAuth h user admin role auth = do
   readTVarIO role >>= \case
     CPRNone -> do
       atomically $ writeTVar role $! newRole
+      when (newRole == CPRNone) $ logWarn "ControlPort: failed auth"
       hPutStrLn h $ currentRole newRole
     r -> hPutStrLn h $ currentRole r <> if r == newRole then "" else ", start new session to change."
   where
@@ -1144,14 +1145,16 @@ cancelSub s = case subThread s of
 type VerifiedTransmissionOrError s = Either (Transmission BrokerMsg) (VerifiedTransmission s)
 
 receive :: forall c s. (Transport c, MsgStoreClass s) => THandleSMP c 'TServer -> s -> Client s -> M s ()
-receive h@THandle {params = THandleParams {thAuth, sessionId}} ms Client {rcvQ, sndQ, rcvActiveAt} = do
+receive h@THandle {params = THandleParams {thAuth, sessionId}} ms c@Client {rcvQ, rcvActiveAt} = do
   labelMyThread . B.unpack $ "client $" <> encode sessionId <> " receive"
   sa <- asks serverActive
   stats <- asks serverStats
+  authDelay <- asks $ authErrorDelay . config
   liftIO $ forever $ do
     ts <- tGetServer h
     unlessM (readTVarIO sa) $ throwIO $ userError "server stopped"
-    atomically . (writeTVar rcvActiveAt $!) =<< getSystemTime
+    receivedAt <- getSystemTime
+    atomically $ writeTVar rcvActiveAt $! receivedAt
     let (es, ts') = partitionEithers $ L.toList ts
         errs = map (second ERR) es
     errs' <- case ts' of
@@ -1167,7 +1170,7 @@ receive h@THandle {params = THandleParams {thAuth, sessionId}} ms Client {rcvQ,
         mapM_ (atomically . writeTBQueue rcvQ) $ L.nonEmpty cmds
         pure $ errs ++ errs'
       [] -> pure errs
-    mapM_ (atomically . writeTBQueue sndQ . (,[])) $ L.nonEmpty errs'
+    mapM_ (sendResponses c stats authDelay receivedAt . (,[])) $ L.nonEmpty errs'
   where
     sameParty :: SParty p -> SignedTransmission Cmd -> Bool
     sameParty p (_, _, (_, _, Cmd p' _)) = isJust $ testEquality p p'
@@ -1190,6 +1193,27 @@ receive h@THandle {params = THandleParams {thAuth, sessionId}} ms Client {rcvQ,
             GET -> incStat $ msgGetAuth stats
             _ -> pure ()
 
+isAuthError :: Transmission BrokerMsg -> Bool
+isAuthError (_, _, r) = case r of
+  ERR AUTH -> True
+  _ -> False
+
+sendResponses :: Client s -> ServerStats -> Int64 -> SystemTime -> (NonEmpty (Transmission BrokerMsg), [Transmission BrokerMsg]) -> IO ()
+sendResponses c@Client {sndQ} stats authDelay since rs@(ts, _)
+  | any isAuthError ts = sendAfterDelay c stats authDelay since rs
+  | otherwise = atomically $ writeTBQueue sndQ rs
+
+sendAfterDelay :: Client s -> ServerStats -> Int64 -> SystemTime -> (NonEmpty (Transmission BrokerMsg), [Transmission BrokerMsg]) -> IO ()
+sendAfterDelay c@Client {sndQ, clientTHParams = THandleParams {sessionId}} stats delay since rs
+  | delay > 0 = do
+      elapsed <- subtract (microseconds since) . microseconds <$> getSystemTime
+      when (elapsed >= delay) $ incStat $ authDelayExceeded stats
+      forkClient c (B.unpack $ "client $" <> encode sessionId <> " auth") $ threadDelay' (delay - elapsed `mod` delay) >> write
+  | otherwise = write
+  where
+    write = atomically $ writeTBQueue sndQ rs
+    microseconds (MkSystemTime s ns) = s * 1000000 + fromIntegral (ns `div` 1000)
+
 send :: Transport c => MVar (THandleSMP c 'TServer) -> Client s -> IO ()
 send th c@Client {sndQ, msgQ, clientTHParams = THandleParams {sessionId}} = do
   labelMyThread . B.unpack $ "client $" <> encode sessionId <> " send"
@@ -1318,11 +1342,6 @@ isContactQueue QueueRec {queueMode, senderKey} = case queueMode of
   Just QMContact -> True
   Nothing -> isNothing senderKey -- for backward compatibility with pre-SKEY contact addresses
 
-isSecuredMsgQueue :: QueueRec -> Bool
-isSecuredMsgQueue QueueRec {queueMode, senderKey} = case queueMode of
-  Just QMContact -> False
-  _ -> isJust senderKey
-
 -- Random correlation ID is used as a nonce in case crypto_box authenticator is used to authorize transmission
 verifyCmdAuthorization :: Maybe (THandleAuth 'TServer) -> Maybe TAuthorizations -> ByteString -> CorrId -> C.APublicAuthKey -> Bool
 verifyCmdAuthorization thAuth tAuth authorized corrId key = maybe False (verify key) tAuth
@@ -1338,7 +1357,7 @@ verifyCmdAuthorization thAuth tAuth authorized corrId key = maybe False (verify
 
 verifyCmdAuth :: Maybe (THandleAuth 'TServer) -> C.PublicKeyX25519 -> C.CbAuthenticator -> ByteString -> CorrId -> Bool
 verifyCmdAuth thAuth k authenticator authorized (CorrId corrId) = case thAuth of
-  Just THAuthServer {serverPrivKey = pk} -> C.cbVerify k pk (C.cbNonce corrId) authenticator authorized
+  Just THAuthServer {serverPrivKey = pk} -> either (const False) (\nonce -> C.cbVerify k pk nonce authenticator authorized) $ C.cbNonce corrId
   Nothing -> False
 
 dummyVerifyCmd :: Maybe (THandleAuth 'TServer) -> Maybe TAuthorizations -> ByteString -> CorrId -> Bool
@@ -1378,15 +1397,18 @@ client
   ms
   clnt@Client {clientId, rcvQ, sndQ, msgQ, clientTHParams = thParams'@THandleParams {sessionId}, procThreads} = do
     labelMyThread . B.unpack $ "client $" <> encode sessionId <> " commands"
+    stats <- asks serverStats
+    authDelay <- asks $ authErrorDelay . config
     let clntServiceId = (\THClientService {serviceId} -> serviceId) <$> (peerClientService =<< thAuth thParams')
         process batchSubs t acc@(rs, msgs) =
           (maybe acc (\(!r, !msg_) -> (r : rs, maybe msgs (: msgs) msg_)))
             <$> processCommand clntServiceId batchSubs t
     forever $ do
       batch <- atomically (readTBQueue rcvQ)
+      processedAt <- liftIO getSystemTime
       batchSubs <- prepareBatchSubs clntServiceId batch
       foldrM (process batchSubs) ([], []) batch
-        >>= \(rs_, msgs) -> mapM_ (atomically . writeTBQueue sndQ . (,msgs)) (L.nonEmpty rs_)
+        >>= \(rs_, msgs) -> mapM_ (liftIO . sendResponses clnt stats authDelay processedAt . (,msgs)) (L.nonEmpty rs_)
   where
     prepareBatchSubs ::
       Maybe ServiceId ->
@@ -1457,14 +1479,15 @@ client
         ServerStats {pMsgFwds, pMsgFwdsOwn} <- asks serverStats
         let inc = mkIncProxyStats pMsgFwds pMsgFwdsOwn
         liftIO (lookupSMPServerClient a sessId) >>= \case
-          Just (own, smp) -> do
+          Just (own, smp) | Right fwdCorrId <- C.corrCbNonce (bs corrId) -> do
             inc own pRequests
             forkProxiedCmd $ do
-              liftIO (runExceptT (forwardSMPTransmission smp corrId fwdV pubKey encBlock) `E.catches` clientHandlers)  >>= \case
-                Right r -> PRES r <$ inc own pSuccesses
+              liftIO (runExceptT (forwardSMPTransmission smp fwdCorrId fwdV pubKey encBlock) `E.catches` clientHandlers)  >>= \case
+                Right (nonce_, r) -> PRES nonce_ r <$ inc own pSuccesses
                 Left e -> ERR (smpProxyError e) <$ case e of
                   PCEProtocolError {} -> inc own pSuccesses
                   _ -> inc own pErrorsOther
+          Just _ -> pure $ Just $ ERR $ CMD SYNTAX
           Nothing -> inc False pRequests >> inc False pErrorsConnect $> Just (ERR $ PROXY NO_SESSION)
       where
         forkProxiedCmd :: M s BrokerMsg -> M s (Maybe BrokerMsg)
@@ -1954,7 +1977,7 @@ client
             getDelivered :: Sub -> STM (Maybe (ServerSub, SystemSeconds))
             getDelivered Sub {delivered, subThread} = do
               readTVar delivered $>>= \(msgId', ts) ->
-                if msgId == msgId' || B.null msgId
+                if msgId == msgId'
                   then writeTVar delivered Nothing $> Just (subThread, ts)
                   else pure Nothing
             updateStats :: ServerStats -> Bool -> SystemSeconds -> Message -> IO ()
@@ -2000,7 +2023,7 @@ client
                       when (isJust (queueData qr) && isSecuredMsgQueue qr) $ void $ liftIO $
                         deleteQueueLinkData (queueStore ms) q
                       ServerConfig {messageExpiration, expireMessagesOnSend, msgIdBytes} <- asks config
-                      msgId <- randomId' msgIdBytes
+                      msgId <- atomically . randomMsgId msgIdBytes =<< asks random
                       msg_ <- liftIO $ runExceptT $ do
                         when expireMessagesOnSend $ mapM_ (expireMessages stats) messageExpiration
                         msg <- liftIO $ mkMessage msgId body
@@ -2113,7 +2136,7 @@ client
               liftIO $ storeNtf ns nId ntf
               incStat . ntfCount =<< asks serverStats
 
-            mkMessageNotification :: ByteString -> SystemTime -> RcvNtfDhSecret -> M s MsgNtf
+            mkMessageNotification :: MsgId -> SystemTime -> RcvNtfDhSecret -> M s MsgNtf
             mkMessageNotification msgId msgTs rcvNtfDhSecret = do
               ntfNonce <- atomically . C.randomCbNonce =<< asks random
               let msgMeta = NMsgMeta {msgId, msgTs}
@@ -2122,14 +2145,16 @@ client
 
         processForwardedCommand :: EncFwdTransmission -> M s (Maybe BrokerMsg)
         processForwardedCommand (EncFwdTransmission s) = fmap (either (Just . ERR) id) . runExceptT $ do
+          startedAt <- liftIO getSystemTime
           THAuthServer {serverPrivKey, sessSecret'} <- maybe (throwE $ transportErr TENoServerAuth) pure (thAuth thParams')
           sessSecret <- maybe (throwE $ transportErr TENoServerAuth) pure sessSecret'
-          let proxyNonce = C.cbNonce $ bs corrId
+          proxyNonce <- liftEitherWith (const CRYPTO) $ C.cbNonce $ bs corrId
           s' <- liftEitherWith (const CRYPTO) $ C.cbDecryptNoPad sessSecret proxyNonce s
           FwdTransmission {fwdCorrId, fwdVersion, fwdKey, fwdTransmission = EncTransmission et} <- liftEitherWith (const $ CMD SYNTAX) $ smpDecode s'
+          unless (fwdVersion `isCompatible` thServerVRange thParams') $ throwE $ transportErr TEVersion
           let clientSecret = C.dh' fwdKey serverPrivKey
-              clientNonce = C.cbNonce $ bs fwdCorrId
-          b <- liftEitherWith (const CRYPTO) $ C.cbDecrypt clientSecret clientNonce et
+              clientNonce = fwdCorrId
+          b <- liftEitherWith (const CRYPTO) $ C.cbDecrypt clientSecret (encTransmissionNonce fwdVersion clientNonce) et
           let clntTHParams = smpTHParamsSetVersion fwdVersion thParams'
           -- only allowing single forwarded transactions
           t' <- case tParse clntTHParams b of
@@ -2142,9 +2167,13 @@ client
                   TBError _ _ : _ -> throwE BLOCK
                   TBTransmission b' _ : _ -> pure b'
                   TBTransmissions b' _ _ : _ -> pure b'
-                r2 <- liftEitherWith (const BLOCK) $ EncResponse <$> C.cbEncrypt clientSecret (C.reverseNonce clientNonce) r' paddedProxiedTLength
+                nonce_ <-
+                  if fwdVersion >= fwdNoncesSMPVersion
+                    then Just <$> (atomically . C.randomCbNonce =<< asks random)
+                    else pure Nothing
+                r2 <- liftEitherWith (const BLOCK) $ EncResponse <$> C.cbEncrypt clientSecret (fromMaybe (C.reverseNonce clientNonce) nonce_) r' paddedProxiedTLength
                 let fr = FwdResponse {fwdCorrId, fwdResponse = r2}
-                pure $ RRES $ EncFwdResponse $ C.cbEncryptNoPad sessSecret (C.reverseNonce proxyNonce) (smpEncode fr)
+                pure $ RRES nonce_ $ EncFwdResponse $ C.cbEncryptNoPad sessSecret (C.reverseNonce proxyNonce) (smpEncode fr)
           -- the inner response, or Nothing if forked (RSLV).
           r_ <- lift (rejectOrVerify clntThAuth t') >>= \case
             -- rejectOrVerify filters allowed commands, no need to repeat it here.
@@ -2160,7 +2189,11 @@ client
               _ -> Just . maybe (corrId', entId', ERR INTERNAL) fst <$> lift (processCommand Nothing (Right (M.empty, M.empty, M.empty)) t'')
           stats <- asks serverStats
           incStat $ pMsgFwdsRecv stats
-          traverse encodeResp r_
+          traverse encodeResp r_ >>= \case
+            Just rres | any isAuthError r_ -> do
+              authDelay <- asks $ authErrorDelay . config
+              Nothing <$ liftIO (sendAfterDelay clnt stats authDelay startedAt ([(corrId, NoEntity, rres)], []))
+            rres_ -> pure rres_
           where
             rejectOrVerify :: Maybe (THandleAuth 'TServer) -> SignedTransmissionOrError ErrorType Cmd -> M s (VerifiedTransmissionOrError s)
             rejectOrVerify clntThAuth = \case
@@ -2186,7 +2219,7 @@ client
           MessageQuota {} -> RcvMsgQuota msgTs'
           where
             encrypt :: KnownNat i => C.MaxLenBS i -> RcvMessage
-            encrypt body = RcvMessage msgId' . EncRcvMsgBody $ C.cbEncryptMaxLenBS (rcvDhSecret qr) (C.cbNonce msgId') body
+            encrypt body = RcvMessage msgId' . EncRcvMsgBody $ C.cbEncryptMaxLenBS (rcvDhSecret qr) msgId' body
             msgId' = messageId msg
             msgTs' = messageTs msg
 
@@ -2235,7 +2268,7 @@ client
                     SubPending -> QSubPending
                     SubThread _ -> QSubThread
                 ProhibitSub -> pure QProhibitSub
-              qDelivered <- decodeLatin1 . encode . fst <$$> readTVarIO delivered
+              qDelivered <- decodeLatin1 . encode . unMsgId . fst <$$> readTVarIO delivered
               pure QSub {qSubThread, qDelivered}
 
         ok :: Transmission BrokerMsg

src/Simplex/Messaging/Server/Env/STM.hs

@@ -68,6 +68,7 @@ module Simplex.Messaging.Server.Env.STM
     defaultInactiveClientExpiration,
     defaultProxyClientConcurrency,
     defaultNameResolverConcurrency,
+    defaultAuthErrorDelayMs,
     defaultMaxJournalMsgCount,
     defaultMaxJournalStateLines,
     defaultIdleQueueInterval,
@@ -187,6 +188,7 @@ data ServerConfig s = ServerConfig
     ntfDeliveryInterval :: Int,
     -- | interval between sending pending END events to unsubscribed clients, seconds
     pendingENDInterval :: Int,
+    authErrorDelay :: Int64,
     smpCredentials :: ServerCredentials,
     httpCredentials :: Maybe ServerCredentials,
     -- | SMP client-server protocol version range
@@ -255,6 +257,9 @@ defaultProxyClientConcurrency = 32
 defaultNameResolverConcurrency :: Int
 defaultNameResolverConcurrency = 1000
 
+defaultAuthErrorDelayMs :: Int64
+defaultAuthErrorDelayMs = 50
+
 journalMsgStoreDepth :: Int
 journalMsgStoreDepth = 5
 
@@ -571,7 +576,10 @@ newProhibitedSub = do
   return Sub {subThread = ProhibitSub, delivered}
 
 newEnv :: ServerConfig s -> IO (Env s)
-newEnv config@ServerConfig {smpCredentials, httpCredentials, serverStoreCfg, smpAgentCfg, information, messageExpiration, idleQueueInterval, msgQueueQuota, maxJournalMsgCount, maxJournalStateLines, namesConfig} = do
+newEnv config@ServerConfig {smpCredentials, httpCredentials, serverStoreCfg, smpAgentCfg, information, messageExpiration, idleQueueInterval, msgQueueQuota, maxJournalMsgCount, maxJournalStateLines, namesConfig, msgIdBytes} = do
+  when (msgIdBytes < 24) $ do
+    logError "Error: msgIdBytes must be at least 24"
+    exitFailure
   serverActive <- newTVarIO True
   server <- newServer
   msgStore_ <- case serverStoreCfg of

src/Simplex/Messaging/Server/Main.hs

@@ -580,6 +580,7 @@ smpServerCLI_ generateSite serveStaticFiles attachStaticFiles cfgPath logPath =
               prometheusMetricsFile = combine logPath "smp-server-metrics.txt",
               pendingENDInterval = 15000000, -- 15 seconds
               ntfDeliveryInterval = 1500000, -- 1.5 second
+              authErrorDelay = 1000 * readIniDefault defaultAuthErrorDelayMs "TRANSPORT" "auth_error_delay_ms" ini,
               smpServerVRange = supportedServerSMPRelayVRange,
               transportConfig =
                 mkTransportServerConfig

src/Simplex/Messaging/Server/Main/Init.hs

@@ -130,7 +130,9 @@ iniFileContent cfgPath logPath opts host basicAuth controlPortPwds =
     <> ("host = " <> T.pack host <> "\n")
     <> ("port = " <> defaultServerPorts <> "\n")
     <> "log_tls_errors = off\n\n\
-        \# Use `websockets = 443` to run websockets server in addition to plain TLS.\n\
+        \# Minimum time to respond with AUTH error, milliseconds.\n"
+    <> ("# auth_error_delay_ms = " <> tshow defaultAuthErrorDelayMs <> "\n\n")
+    <> "# Use `websockets = 443` to run websockets server in addition to plain TLS.\n\
         \# This option is deprecated and should be used for testing only.\n\
         \# , port 443 should be specified in port above\n\
         \websockets = off\n"

src/Simplex/Messaging/Server/MsgStore/Postgres.hs

@@ -114,7 +114,9 @@ instance MsgStoreClass PostgresMsgStore where
     where
       st = dbStore $ queueStore_ ms
       oldMsg = now - ttl
-      batchSize = 10000 :: Int
+      -- expired messages read per page in expire_old_messages, and the page is one
+      -- transaction: queues in it stay row-locked against SEND and ACK until it commits.
+      batchSize = 100 :: Int
       toMessageStats (expiredMsgsCount, storedMsgsCount, storedQueues) =
         MessageStats {expiredMsgsCount, storedMsgsCount, storedQueues}
 
@@ -200,7 +202,7 @@ instance MsgStoreClass PostgresMsgStore where
           DB.query
             db
             "SELECT quota_written, was_empty FROM write_message(?,?,?,?,?,?,?)"
-            (recipientId' q, Binary (messageId msg), systemSeconds (messageTs msg), msgQuota, ntf, Binary body, quota)
+            (recipientId' q, messageId msg, systemSeconds (messageTs msg), msgQuota, ntf, Binary body, quota)
     where
       toResult = \case
         ((msgQuota, wasEmpty) : _) -> if msgQuota then Nothing else Just (msg, wasEmpty)
@@ -269,14 +271,14 @@ instance MsgStoreClass PostgresMsgStore where
     uninterruptibleMask_ $
       withDB' "tryDelMsg" (queueStore_ ms) $ \db ->
         maybeFirstRow toMessage $
-          DB.query db "SELECT r_msg_id, r_msg_ts, r_msg_quota, r_msg_ntf_flag, r_msg_body FROM try_del_msg(?, ?)" (recipientId' q, Binary msgId)
+          DB.query db "SELECT r_msg_id, r_msg_ts, r_msg_quota, r_msg_ntf_flag, r_msg_body FROM try_del_msg(?, ?)" (recipientId' q, msgId)
 
   tryDelPeekMsg :: PostgresMsgStore -> PostgresQueue -> MsgId -> ExceptT ErrorType IO (Maybe Message, Maybe Message)
   tryDelPeekMsg ms q msgId =
     uninterruptibleMask_ $
       withDB' "tryDelPeekMsg" (queueStore_ ms) $ \db ->
         toResult . map toMessage
-          <$> DB.query db "SELECT r_msg_id, r_msg_ts, r_msg_quota, r_msg_ntf_flag, r_msg_body FROM try_del_peek_msg(?, ?)" (recipientId' q, Binary msgId)
+          <$> DB.query db "SELECT r_msg_id, r_msg_ts, r_msg_quota, r_msg_ntf_flag, r_msg_body FROM try_del_peek_msg(?, ?)" (recipientId' q, msgId)
     where
       toResult = \case
         [] -> (Nothing, Nothing)
@@ -295,13 +297,13 @@ uninterruptibleMask_ :: ExceptT ErrorType IO a -> ExceptT ErrorType IO a
 uninterruptibleMask_ = ExceptT . E.uninterruptibleMask_ . runExceptT
 {-# INLINE uninterruptibleMask_ #-}
 
-toMaybeMessage :: (Maybe (Binary MsgId), Maybe Int64, Maybe Bool, Maybe Bool, Maybe (Binary MsgBody)) -> Maybe Message
+toMaybeMessage :: (Maybe MsgId, Maybe Int64, Maybe Bool, Maybe Bool, Maybe (Binary MsgBody)) -> Maybe Message
 toMaybeMessage = \case
   (Just msgId, Just ts, Just msgQuota, Just ntf, Just body) -> Just $ toMessage (msgId, ts, msgQuota, ntf, body)
   _ -> Nothing
 
-toMessage :: (Binary MsgId, Int64, Bool, Bool, Binary MsgBody) -> Message
-toMessage (Binary msgId, ts, msgQuota, ntf, Binary body)
+toMessage :: (MsgId, Int64, Bool, Bool, Binary MsgBody) -> Message
+toMessage (msgId, ts, msgQuota, ntf, Binary body)
   | msgQuota = MessageQuota {msgId, msgTs}
   | otherwise = Message {msgId, msgTs, msgFlags = MsgFlags ntf, msgBody = C.unsafeMaxLenBS body} -- TODO [messages] unsafeMaxLenBS?
   where
@@ -363,8 +365,8 @@ deleteAllMessages ms =
       db
       [sql|
         UPDATE msg_queues
-        SET msg_queue_size = 0, msg_can_write = TRUE, msg_queue_expire = FALSE
-        WHERE msg_queue_size != 0 OR msg_can_write = FALSE OR msg_queue_expire = TRUE
+        SET msg_queue_size = 0, msg_can_write = TRUE
+        WHERE msg_queue_size != 0 OR msg_can_write = FALSE
       |]
 
 updateQueueCounts :: PostgresMsgStore -> IO ()
@@ -384,16 +386,15 @@ updateQueueCounts ms =
       db
       [sql|
         UPDATE msg_queues
-        SET msg_queue_size = 0, msg_can_write = TRUE, msg_queue_expire = FALSE
-        WHERE msg_queue_size != 0 OR msg_can_write = FALSE OR msg_queue_expire = TRUE
+        SET msg_queue_size = 0, msg_can_write = TRUE
+        WHERE msg_queue_size != 0 OR msg_can_write = FALSE
       |]
     void $ DB.execute_
       db
       [sql|
         UPDATE msg_queues q
         SET msg_queue_size = s.size,
-            msg_can_write = s.quota_count = 0,
-            msg_queue_expire = s.size > s.quota_count
+            msg_can_write = s.quota_count = 0
         FROM queue_stats s
         WHERE q.recipient_id = s.recipient_id
       |]
@@ -428,7 +429,7 @@ messageRecToText rId msg =
     tabFields = BB.char7 ',' `intersperse` fields
     fields =
       [ renderField (toField rId),
-        renderField (toField $ Binary (messageId msg)),
+        renderField (toField $ messageId msg),
         renderField (toField $ systemSeconds (messageTs msg)),
         renderField (toField msgQuota),
         renderField (toField ntf),

src/Simplex/Messaging/Server/Prometheus.hs

@@ -126,6 +126,7 @@ prometheusMetrics sm rtm ts =
         _ntfServices,
         _rcvServicesSubMsg,
         _rcvServicesSubDuplicate,
+        _authDelayExceeded,
         _qCount,
         _msgCount,
         _ntfCount,
@@ -192,6 +193,10 @@ prometheusMetrics sm rtm ts =
       \# HELP simplex_smp_queues_notify_monthly Monthly active queues with notifications.\n\
       \# TYPE simplex_smp_queues_notify_monthly gauge\n\
       \simplex_smp_queues_notify_monthly " <> mstr (monthCount psNtf) <> "\n# monthCountNtf\n\
+      \\n\
+      \# HELP simplex_smp_auth_delay_exceeded Responses with AUTH errors sent without delay, as processing took longer than the delay.\n\
+      \# TYPE simplex_smp_auth_delay_exceeded counter\n\
+      \simplex_smp_auth_delay_exceeded " <> mshow _authDelayExceeded <> "\n# authDelayExceeded\n\
       \\n"
     subscriptions =
       "# Subscriptions\n\

src/Simplex/Messaging/Server/QueueStore.hs

@@ -13,12 +13,14 @@ module Simplex.Messaging.Server.QueueStore
     ServiceRec (..),
     CertFingerprint,
     ServerEntityStatus (..),
+    isSecuredMsgQueue,
   ) where
 
 import Control.Applicative (optional, (<|>))
 import qualified Data.ByteString.Char8 as B
 import Data.Functor (($>))
 import Data.List.NonEmpty (NonEmpty)
+import Data.Maybe (isJust)
 import qualified Data.X509 as X
 import qualified Data.X509.Validation as XV
 import Simplex.Messaging.Encoding
@@ -48,6 +50,11 @@ data QueueRec = QueueRec
   }
   deriving (Show)
 
+isSecuredMsgQueue :: QueueRec -> Bool
+isSecuredMsgQueue QueueRec {queueMode, senderKey} = case queueMode of
+  Just QMContact -> False
+  _ -> isJust senderKey
+
 data NtfCreds = NtfCreds
   { notifierId :: NotifierId,
     notifierKey :: NtfPublicAuthKey,

src/Simplex/Messaging/Server/QueueStore/Postgres.hs

@@ -314,9 +314,9 @@ instance StoreQueueClass q => QueueStoreClass q (PostgresQueueStore q) where
   addQueueLinkData st sq lnkId d =
     withQueueRec sq "addQueueLinkData" $ \q -> case queueData q of
       Nothing ->
-        addLink q $ \db -> DB.execute db qry (d :. (lnkId, rId))
+        addLink q $ \db -> DB.execute db qry (d :. (lnkId, rId, QMContact))
       Just (lnkId', _) | lnkId' == lnkId ->
-        addLink q $ \db -> DB.execute db (qry <> " AND (fixed_data IS NULL OR fixed_data = ?)") (d :. (lnkId, rId, fst d))
+        addLink q $ \db -> DB.execute db (qry <> " AND (fixed_data IS NULL OR fixed_data = ?)") (d :. (lnkId, rId, QMContact, fst d))
       _ -> throwE AUTH
     where
       rId = recipientId sq
@@ -324,7 +324,8 @@ instance StoreQueueClass q => QueueStoreClass q (PostgresQueueStore q) where
         assertUpdated $ withDB' "addQueueLinkData" st update
         atomically $ writeTVar (queueRec sq) $ Just q {queueData = Just (lnkId, d)}
         withLog "addQueueLinkData" st $ \s -> logCreateLink s rId lnkId d
-      qry = "UPDATE msg_queues SET fixed_data = ?, user_data = ?, link_id = ? WHERE recipient_id = ? AND deleted_at IS NULL"
+      -- the sender key condition is checked in SQL because without cache each command reads its own copy of the queue record
+      qry = "UPDATE msg_queues SET fixed_data = ?, user_data = ?, link_id = ? WHERE recipient_id = ? AND deleted_at IS NULL AND (sender_key IS NULL OR queue_mode = ?)"
 
   deleteQueueLinkData :: PostgresQueueStore q -> q -> IO (Either ErrorType ())
   deleteQueueLinkData st sq =
@@ -824,6 +825,10 @@ instance ToField EncDataBytes where toField (EncDataBytes s) = toField (Binary s
 
 deriving newtype instance FromField EncDataBytes
 
+instance ToField MsgId where toField = toField . Binary . unMsgId
+
+instance FromField MsgId where fromField = blobFieldDecoder mkMsgId
+
 deriving newtype instance ToField (RoundedSystemTime t)
 
 deriving newtype instance FromField (RoundedSystemTime t)

src/Simplex/Messaging/Server/QueueStore/Postgres/Migrations.hs

@@ -22,7 +22,8 @@ serverSchemaMigrations =
     ("20250903_store_messages", m20250903_store_messages, Just down_m20250903_store_messages),
     ("20250915_queue_ids_hash", m20250915_queue_ids_hash, Just down_m20250915_queue_ids_hash),
     ("20260916_prometheus_indexes", m20260916_prometheus_indexes, Just down_m20260916_prometheus_indexes),
-    ("20260917_msg_queues_hot", m20260917_msg_queues_hot, Just down_m20260917_msg_queues_hot)
+    ("20260917_msg_queues_hot", m20260917_msg_queues_hot, Just down_m20260917_msg_queues_hot),
+    ("20260918_expire_messages", m20260918_expire_messages, Just down_m20260918_expire_messages)
   ]
 
 -- | The list of migrations in ascending order by date
@@ -734,3 +735,504 @@ ALTER TABLE msg_queues RESET (fillfactor, autovacuum_vacuum_scale_factor, autova
 ALTER TABLE messages RESET (autovacuum_vacuum_scale_factor, autovacuum_analyze_scale_factor, toast.autovacuum_vacuum_scale_factor);
 ALTER TABLE services RESET (fillfactor, autovacuum_vacuum_threshold, autovacuum_vacuum_scale_factor);
     |]
+
+m20260918_expire_messages :: Text
+m20260918_expire_messages =
+  [r|
+CREATE INDEX idx_messages_expire ON messages (msg_ts, recipient_id) WHERE NOT msg_quota;
+
+DROP INDEX idx_messages_recipient_id_msg_ts;
+
+DROP INDEX idx_msg_queues_expire;
+
+DROP PROCEDURE expire_old_messages(bigint, integer);
+
+CREATE PROCEDURE expire_old_messages(IN p_old_ts bigint, IN batch_size integer, OUT r_expired_msgs_count bigint, OUT r_stored_msgs_count bigint, OUT r_stored_queues bigint)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  rids BYTEA[];
+  rid BYTEA;
+  last_ts BIGINT := -1;
+  last_rid BYTEA := '\x';
+  next_ts BIGINT;
+  next_rid BYTEA;
+  del_count BIGINT;
+  total_deleted BIGINT := 0;
+BEGIN
+  LOOP
+    -- The page is scanned in (msg_ts, recipient_id) order, so its last row is the next
+    -- keyset cursor. Advancing it past every row read, including queues left unexpired
+    -- because delete_expired_msgs skipped a locked row or raised, is what terminates the
+    -- loop; re-reading from the start instead would repeat those queues forever.
+    SELECT array_agg(DISTINCT recipient_id),
+           (array_agg(msg_ts ORDER BY msg_ts DESC, recipient_id DESC))[1],
+           (array_agg(recipient_id ORDER BY msg_ts DESC, recipient_id DESC))[1]
+    INTO rids, next_ts, next_rid
+    FROM (
+      SELECT msg_ts, recipient_id
+      FROM messages
+      WHERE NOT msg_quota
+        AND msg_ts < p_old_ts
+        AND (msg_ts, recipient_id) > (last_ts, last_rid)
+      ORDER BY msg_ts ASC, recipient_id ASC
+      LIMIT batch_size
+    ) m;
+
+    EXIT WHEN rids IS NULL;
+
+    FOREACH rid IN ARRAY rids
+    LOOP
+      BEGIN
+        del_count := delete_expired_msgs(rid, p_old_ts);
+        total_deleted := total_deleted + del_count;
+      EXCEPTION WHEN OTHERS THEN
+        RAISE WARNING 'STORE, expire_old_messages, error expiring queue %: %', encode(rid, 'base64'), SQLERRM;
+        CONTINUE;
+      END;
+    END LOOP;
+    COMMIT;
+
+    last_ts := next_ts;
+    last_rid := next_rid;
+  END LOOP;
+
+  r_expired_msgs_count := total_deleted;
+  r_stored_msgs_count := (SELECT COUNT(1) FROM messages);
+  r_stored_queues := (SELECT COUNT(1) FROM msg_queues WHERE deleted_at IS NULL);
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION delete_expired_msgs(p_recipient_id bytea, p_old_ts bigint) RETURNS bigint
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  keep_min_id BIGINT;
+  del_count BIGINT;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE SKIP LOCKED;
+
+  IF NOT FOUND OR q_size = 0 THEN
+    RETURN 0;
+  END IF;
+
+  SELECT MIN(message_id) INTO keep_min_id
+  FROM messages WHERE recipient_id = p_recipient_id AND msg_ts >= p_old_ts AND msg_quota = FALSE;
+
+  IF keep_min_id IS NULL THEN
+    DELETE FROM messages WHERE recipient_id = p_recipient_id AND msg_quota = FALSE;
+  ELSE
+    DELETE FROM messages WHERE recipient_id = p_recipient_id AND message_id < keep_min_id AND msg_quota = FALSE;
+  END IF;
+
+  GET DIAGNOSTICS del_count = ROW_COUNT;
+  IF del_count > 0 THEN
+    UPDATE msg_queues
+    SET msg_can_write = msg_can_write OR msg_queue_size <= del_count,
+        msg_queue_size = GREATEST(msg_queue_size - del_count, 0)
+    WHERE recipient_id = p_recipient_id;
+  END IF;
+  RETURN del_count;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION write_message(p_recipient_id bytea, p_msg_id bytea, p_msg_ts bigint, p_msg_quota boolean, p_msg_ntf_flag boolean, p_msg_body bytea, p_quota integer) RETURNS TABLE(quota_written boolean, was_empty boolean)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_can_write BOOLEAN;
+  q_size BIGINT;
+BEGIN
+  SELECT msg_can_write, msg_queue_size INTO q_can_write, q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF q_can_write OR q_size = 0 THEN
+    quota_written := p_msg_quota OR q_size >= p_quota;
+    was_empty := q_size = 0;
+
+    INSERT INTO messages(recipient_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body)
+    VALUES (p_recipient_id, p_msg_id, p_msg_ts, quota_written, p_msg_ntf_flag AND NOT quota_written, CASE WHEN quota_written THEN '' :: BYTEA ELSE p_msg_body END);
+
+    UPDATE msg_queues
+    SET msg_can_write = NOT quota_written,
+        msg_queue_size = msg_queue_size + 1
+    WHERE recipient_id = p_recipient_id;
+
+    RETURN QUERY VALUES (quota_written, was_empty);
+  END IF;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION try_del_msg(p_recipient_id bytea, p_msg_id bytea) RETURNS TABLE(r_msg_id bytea, r_msg_ts bigint, r_msg_quota boolean, r_msg_ntf_flag boolean, r_msg_body bytea)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  msg RECORD;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF NOT FOUND THEN
+    RETURN;
+  END IF;
+
+  SELECT message_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+  INTO msg
+  FROM messages
+  WHERE recipient_id = p_recipient_id
+  ORDER BY message_id ASC LIMIT 1;
+
+  IF NOT FOUND THEN
+    IF q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+    RETURN;
+  END IF;
+
+  IF msg.msg_id = p_msg_id THEN
+    DELETE FROM messages WHERE message_id = msg.message_id;
+    IF FOUND THEN
+      UPDATE msg_queues
+      SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
+          msg_queue_size = GREATEST(msg_queue_size - 1, 0)
+      WHERE recipient_id = p_recipient_id;
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+    END IF;
+  END IF;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION try_del_peek_msg(p_recipient_id bytea, p_msg_id bytea) RETURNS TABLE(r_msg_id bytea, r_msg_ts bigint, r_msg_quota boolean, r_msg_ntf_flag boolean, r_msg_body bytea)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  msg RECORD;
+  msg_deleted BOOLEAN;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF NOT FOUND THEN
+    RETURN;
+  END IF;
+
+  SELECT message_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+  INTO msg
+  FROM messages
+  WHERE recipient_id = p_recipient_id
+  ORDER BY message_id ASC LIMIT 1;
+
+  IF NOT FOUND THEN
+    IF q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+    RETURN;
+  END IF;
+
+  IF msg.msg_id = p_msg_id THEN
+    DELETE FROM messages WHERE message_id = msg.message_id;
+
+    msg_deleted := FOUND;
+    IF msg_deleted THEN
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+    END IF;
+
+    SELECT msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+    INTO msg
+    FROM messages
+    WHERE recipient_id = p_recipient_id
+    ORDER BY message_id ASC LIMIT 1;
+
+    IF FOUND THEN
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+      IF msg_deleted THEN
+        UPDATE msg_queues
+        SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
+            msg_queue_size = GREATEST(msg_queue_size - 1, 0)
+        WHERE recipient_id = p_recipient_id;
+      END IF;
+    ELSIF msg_deleted OR q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+  ELSE
+    RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+  END IF;
+END;
+$$;
+
+ALTER TABLE msg_queues DROP COLUMN msg_queue_expire;
+    |]
+
+down_m20260918_expire_messages :: Text
+down_m20260918_expire_messages =
+  [r|
+ALTER TABLE msg_queues ADD COLUMN msg_queue_expire boolean NOT NULL DEFAULT FALSE;
+
+-- ADD COLUMN already defaulted every row to FALSE, so the backfill only needs the rows
+-- that become TRUE, which avoids rewriting the whole table. The flag is only read to pick
+-- queues for delete_expired_msgs, and that returns 0 without deleting when msg_queue_size
+-- is 0, so excluding empty queues here cannot change what expires.
+UPDATE msg_queues q
+SET msg_queue_expire = TRUE
+WHERE msg_queue_size > 0
+  AND EXISTS (SELECT 1 FROM messages m WHERE m.recipient_id = q.recipient_id AND NOT m.msg_quota);
+
+CREATE INDEX idx_msg_queues_expire ON msg_queues (recipient_id) WHERE deleted_at IS NULL AND msg_queue_expire;
+
+CREATE INDEX idx_messages_recipient_id_msg_ts ON messages (recipient_id, msg_ts);
+
+DROP INDEX idx_messages_expire;
+
+CREATE OR REPLACE FUNCTION delete_expired_msgs(p_recipient_id bytea, p_old_ts bigint) RETURNS bigint
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  keep_min_id BIGINT;
+  del_count BIGINT;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE SKIP LOCKED;
+
+  IF NOT FOUND OR q_size = 0 THEN
+    RETURN 0;
+  END IF;
+
+  SELECT MIN(message_id) INTO keep_min_id
+  FROM messages WHERE recipient_id = p_recipient_id AND msg_ts >= p_old_ts AND msg_quota = FALSE;
+
+  IF keep_min_id IS NULL THEN
+    DELETE FROM messages WHERE recipient_id = p_recipient_id AND msg_quota = FALSE;
+  ELSE
+    DELETE FROM messages WHERE recipient_id = p_recipient_id AND message_id < keep_min_id AND msg_quota = FALSE;
+  END IF;
+
+  GET DIAGNOSTICS del_count = ROW_COUNT;
+  IF del_count > 0 THEN
+    UPDATE msg_queues
+    SET msg_can_write = msg_can_write OR msg_queue_size <= del_count,
+        msg_queue_expire = msg_queue_size > del_count AND keep_min_id IS NOT NULL,
+        msg_queue_size = GREATEST(msg_queue_size - del_count, 0)
+    WHERE recipient_id = p_recipient_id;
+  END IF;
+  RETURN del_count;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION write_message(p_recipient_id bytea, p_msg_id bytea, p_msg_ts bigint, p_msg_quota boolean, p_msg_ntf_flag boolean, p_msg_body bytea, p_quota integer) RETURNS TABLE(quota_written boolean, was_empty boolean)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_can_write BOOLEAN;
+  q_size BIGINT;
+BEGIN
+  SELECT msg_can_write, msg_queue_size INTO q_can_write, q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF q_can_write OR q_size = 0 THEN
+    quota_written := p_msg_quota OR q_size >= p_quota;
+    was_empty := q_size = 0;
+
+    INSERT INTO messages(recipient_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body)
+    VALUES (p_recipient_id, p_msg_id, p_msg_ts, quota_written, p_msg_ntf_flag AND NOT quota_written, CASE WHEN quota_written THEN '' :: BYTEA ELSE p_msg_body END);
+
+    UPDATE msg_queues
+    SET msg_can_write = NOT quota_written,
+        msg_queue_expire = TRUE,
+        msg_queue_size = msg_queue_size + 1
+    WHERE recipient_id = p_recipient_id;
+
+    RETURN QUERY VALUES (quota_written, was_empty);
+  END IF;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION try_del_msg(p_recipient_id bytea, p_msg_id bytea) RETURNS TABLE(r_msg_id bytea, r_msg_ts bigint, r_msg_quota boolean, r_msg_ntf_flag boolean, r_msg_body bytea)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  msg RECORD;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF NOT FOUND THEN
+    RETURN;
+  END IF;
+
+  SELECT message_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+  INTO msg
+  FROM messages
+  WHERE recipient_id = p_recipient_id
+  ORDER BY message_id ASC LIMIT 1;
+
+  IF NOT FOUND THEN
+    IF q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_expire = FALSE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+    RETURN;
+  END IF;
+
+  IF msg.msg_id = p_msg_id THEN
+    DELETE FROM messages WHERE message_id = msg.message_id;
+    IF FOUND THEN
+      UPDATE msg_queues
+      SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
+          msg_queue_expire = msg_queue_size > 1,
+          msg_queue_size = GREATEST(msg_queue_size - 1, 0)
+      WHERE recipient_id = p_recipient_id;
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+    END IF;
+  END IF;
+END;
+$$;
+
+CREATE OR REPLACE FUNCTION try_del_peek_msg(p_recipient_id bytea, p_msg_id bytea) RETURNS TABLE(r_msg_id bytea, r_msg_ts bigint, r_msg_quota boolean, r_msg_ntf_flag boolean, r_msg_body bytea)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  q_size BIGINT;
+  msg RECORD;
+  msg_deleted BOOLEAN;
+BEGIN
+  SELECT msg_queue_size INTO q_size
+  FROM msg_queues
+  WHERE recipient_id = p_recipient_id AND deleted_at IS NULL
+  FOR UPDATE;
+
+  IF NOT FOUND THEN
+    RETURN;
+  END IF;
+
+  SELECT message_id, msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+  INTO msg
+  FROM messages
+  WHERE recipient_id = p_recipient_id
+  ORDER BY message_id ASC LIMIT 1;
+
+  IF NOT FOUND THEN
+    IF q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_expire = FALSE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+    RETURN;
+  END IF;
+
+  IF msg.msg_id = p_msg_id THEN
+    DELETE FROM messages WHERE message_id = msg.message_id;
+
+    msg_deleted := FOUND;
+    IF msg_deleted THEN
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+    END IF;
+
+    SELECT msg_id, msg_ts, msg_quota, msg_ntf_flag, msg_body
+    INTO msg
+    FROM messages
+    WHERE recipient_id = p_recipient_id
+    ORDER BY message_id ASC LIMIT 1;
+
+    IF FOUND THEN
+      RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+      IF msg_deleted THEN
+        UPDATE msg_queues
+        SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
+            msg_queue_expire = msg_queue_size > 1,
+            msg_queue_size = GREATEST(msg_queue_size - 1, 0)
+        WHERE recipient_id = p_recipient_id;
+      END IF;
+    ELSIF msg_deleted OR q_size != 0 THEN
+      UPDATE msg_queues
+      SET msg_can_write = TRUE,
+          msg_queue_expire = FALSE,
+          msg_queue_size = 0
+      WHERE recipient_id = p_recipient_id;
+    END IF;
+  ELSE
+    RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
+  END IF;
+END;
+$$;
+
+DROP PROCEDURE expire_old_messages(bigint, integer);
+
+CREATE PROCEDURE expire_old_messages(IN p_old_ts bigint, IN batch_size integer, OUT r_expired_msgs_count bigint, OUT r_stored_msgs_count bigint, OUT r_stored_queues bigint)
+    LANGUAGE plpgsql
+    AS $$
+DECLARE
+  rids BYTEA[];
+  rid BYTEA;
+  last_rid BYTEA := '\x';
+  del_count BIGINT;
+  total_deleted BIGINT := 0;
+BEGIN
+  LOOP
+    SELECT array_agg(recipient_id)
+    INTO rids
+    FROM (
+      SELECT recipient_id
+      FROM msg_queues
+      WHERE deleted_at IS NULL
+        AND msg_queue_expire = TRUE
+        AND recipient_id > last_rid
+      ORDER BY recipient_id ASC
+      LIMIT batch_size
+    ) qs;
+
+    EXIT WHEN rids IS NULL OR cardinality(rids) = 0;
+
+    FOREACH rid IN ARRAY rids
+    LOOP
+      BEGIN
+        del_count := delete_expired_msgs(rid, p_old_ts);
+        total_deleted := total_deleted + del_count;
+      EXCEPTION WHEN OTHERS THEN
+        RAISE WARNING 'STORE, expire_old_messages, error expiring queue %: %', encode(rid, 'base64'), SQLERRM;
+        CONTINUE;
+      END;
+      COMMIT;
+    END LOOP;
+    last_rid := rids[cardinality(rids)];
+  END LOOP;
+
+  r_expired_msgs_count := total_deleted;
+  r_stored_msgs_count := (SELECT COUNT(1) FROM messages);
+  r_stored_queues := (SELECT COUNT(1) FROM msg_queues WHERE deleted_at IS NULL);
+END;
+$$;
+    |]

src/Simplex/Messaging/Server/QueueStore/Postgres/server_schema.sql

@@ -47,7 +47,6 @@ BEGIN
   IF del_count > 0 THEN
     UPDATE msg_queues
     SET msg_can_write = msg_can_write OR msg_queue_size <= del_count,
-        msg_queue_expire = msg_queue_size > del_count AND keep_min_id IS NOT NULL,
         msg_queue_size = GREATEST(msg_queue_size - del_count, 0)
     WHERE recipient_id = p_recipient_id;
   END IF;
@@ -63,24 +62,33 @@ CREATE PROCEDURE smp_server.expire_old_messages(IN p_old_ts bigint, IN batch_siz
 DECLARE
   rids BYTEA[];
   rid BYTEA;
+  last_ts BIGINT := -1;
   last_rid BYTEA := '\x';
+  next_ts BIGINT;
+  next_rid BYTEA;
   del_count BIGINT;
   total_deleted BIGINT := 0;
 BEGIN
   LOOP
-    SELECT array_agg(recipient_id)
-    INTO rids
+    -- The page is scanned in (msg_ts, recipient_id) order, so its last row is the next
+    -- keyset cursor. Advancing it past every row read, including queues left unexpired
+    -- because delete_expired_msgs skipped a locked row or raised, is what terminates the
+    -- loop; re-reading from the start instead would repeat those queues forever.
+    SELECT array_agg(DISTINCT recipient_id),
+           (array_agg(msg_ts ORDER BY msg_ts DESC, recipient_id DESC))[1],
+           (array_agg(recipient_id ORDER BY msg_ts DESC, recipient_id DESC))[1]
+    INTO rids, next_ts, next_rid
     FROM (
-      SELECT recipient_id
-      FROM msg_queues
-      WHERE deleted_at IS NULL
-        AND msg_queue_expire = TRUE
-        AND recipient_id > last_rid
-      ORDER BY recipient_id ASC
+      SELECT msg_ts, recipient_id
+      FROM messages
+      WHERE NOT msg_quota
+        AND msg_ts < p_old_ts
+        AND (msg_ts, recipient_id) > (last_ts, last_rid)
+      ORDER BY msg_ts ASC, recipient_id ASC
       LIMIT batch_size
-    ) qs;
+    ) m;
 
-    EXIT WHEN rids IS NULL OR cardinality(rids) = 0;
+    EXIT WHEN rids IS NULL;
 
     FOREACH rid IN ARRAY rids
     LOOP
@@ -91,9 +99,11 @@ BEGIN
         RAISE WARNING 'STORE, expire_old_messages, error expiring queue %: %', encode(rid, 'base64'), SQLERRM;
         CONTINUE;
       END;
-      COMMIT;
     END LOOP;
-    last_rid := rids[cardinality(rids)];
+    COMMIT;
+
+    last_ts := next_ts;
+    last_rid := next_rid;
   END LOOP;
 
   r_expired_msgs_count := total_deleted;
@@ -195,7 +205,6 @@ BEGIN
     IF q_size != 0 THEN
       UPDATE msg_queues
       SET msg_can_write = TRUE,
-          msg_queue_expire = FALSE,
           msg_queue_size = 0
       WHERE recipient_id = p_recipient_id;
     END IF;
@@ -207,7 +216,6 @@ BEGIN
     IF FOUND THEN
       UPDATE msg_queues
       SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
-          msg_queue_expire = msg_queue_size > 1,
           msg_queue_size = GREATEST(msg_queue_size - 1, 0)
       WHERE recipient_id = p_recipient_id;
       RETURN QUERY VALUES (msg.msg_id, msg.msg_ts, msg.msg_quota, msg.msg_ntf_flag, msg.msg_body);
@@ -245,7 +253,6 @@ BEGIN
     IF q_size != 0 THEN
       UPDATE msg_queues
       SET msg_can_write = TRUE,
-          msg_queue_expire = FALSE,
           msg_queue_size = 0
       WHERE recipient_id = p_recipient_id;
     END IF;
@@ -271,14 +278,12 @@ BEGIN
       IF msg_deleted THEN
         UPDATE msg_queues
         SET msg_can_write = msg_can_write OR msg_queue_size <= 1,
-            msg_queue_expire = msg_queue_size > 1,
             msg_queue_size = GREATEST(msg_queue_size - 1, 0)
         WHERE recipient_id = p_recipient_id;
       END IF;
     ELSIF msg_deleted OR q_size != 0 THEN
       UPDATE msg_queues
       SET msg_can_write = TRUE,
-          msg_queue_expire = FALSE,
           msg_queue_size = 0
       WHERE recipient_id = p_recipient_id;
     END IF;
@@ -348,7 +353,6 @@ BEGIN
 
     UPDATE msg_queues
     SET msg_can_write = NOT quota_written,
-        msg_queue_expire = TRUE,
         msg_queue_size = msg_queue_size + 1
     WHERE recipient_id = p_recipient_id;
 
@@ -440,7 +444,6 @@ CREATE TABLE smp_server.msg_queues (
     rcv_service_id bytea,
     ntf_service_id bytea,
     msg_can_write boolean DEFAULT true NOT NULL,
-    msg_queue_expire boolean DEFAULT false NOT NULL,
     msg_queue_size bigint DEFAULT 0 NOT NULL
 )
 WITH (fillfactor='80', autovacuum_vacuum_scale_factor='0.02', autovacuum_analyze_scale_factor='0.01', autovacuum_vacuum_cost_limit='1000');
@@ -485,6 +488,10 @@ ALTER TABLE ONLY smp_server.services
 
 
 
+CREATE INDEX idx_messages_expire ON smp_server.messages USING btree (msg_ts, recipient_id) WHERE (NOT msg_quota);
+
+
+
 CREATE INDEX idx_messages_recipient_id_message_id ON smp_server.messages USING btree (recipient_id, message_id);
 
 
@@ -493,14 +500,6 @@ CREATE INDEX idx_messages_recipient_id_msg_quota ON smp_server.messages USING bt
 
 
 
-CREATE INDEX idx_messages_recipient_id_msg_ts ON smp_server.messages USING btree (recipient_id, msg_ts);
-
-
-
-CREATE INDEX idx_msg_queues_expire ON smp_server.msg_queues USING btree (recipient_id) WHERE ((deleted_at IS NULL) AND msg_queue_expire);
-
-
-
 CREATE UNIQUE INDEX idx_msg_queues_link_id ON smp_server.msg_queues USING btree (link_id);
 
 

src/Simplex/Messaging/Server/QueueStore/STM.hs

@@ -173,6 +173,7 @@ instance StoreQueueClass q => QueueStoreClass q (STMQueueStore q) where
       rId = recipientId sq
       qr = queueRec sq
       add q = case queueData q of
+        _ | isSecuredMsgQueue q -> pure $ Left AUTH
         Nothing -> addLink
         Just (lnkId', d') | lnkId' == lnkId && fst d' == fst d -> addLink
         _ -> pure $ Left AUTH

src/Simplex/Messaging/Server/Stats.hs

@@ -128,6 +128,7 @@ data ServerStats = ServerStats
     ntfServices :: ServiceStats,
     rcvServicesSubMsg :: IORef Int,
     rcvServicesSubDuplicate :: IORef Int,
+    authDelayExceeded :: IORef Int,
     qCount :: IORef Int,
     msgCount :: IORef Int,
     ntfCount :: IORef Int,
@@ -190,6 +191,7 @@ data ServerStatsData = ServerStatsData
     _rcvServices :: ServiceStatsData,
     _rcvServicesSubMsg :: Int,
     _rcvServicesSubDuplicate :: Int,
+    _authDelayExceeded :: Int,
     _qCount :: Int,
     _msgCount :: Int,
     _ntfCount :: Int,
@@ -254,6 +256,7 @@ newServerStats ts = do
   ntfServices <- newServiceStats
   rcvServicesSubMsg <- newIORef 0
   rcvServicesSubDuplicate <- newIORef 0
+  authDelayExceeded <- newIORef 0
   qCount <- newIORef 0
   msgCount <- newIORef 0
   ntfCount <- newIORef 0
@@ -315,6 +318,7 @@ newServerStats ts = do
         ntfServices,
         rcvServicesSubMsg,
         rcvServicesSubDuplicate,
+        authDelayExceeded,
         qCount,
         msgCount,
         ntfCount,
@@ -378,6 +382,7 @@ getServerStatsData s = do
   _ntfServices <- getServiceStatsData $ ntfServices s
   _rcvServicesSubMsg <- readIORef $ rcvServicesSubMsg s
   _rcvServicesSubDuplicate <- readIORef $ rcvServicesSubDuplicate s
+  _authDelayExceeded <- readIORef $ authDelayExceeded s
   _qCount <- readIORef $ qCount s
   _msgCount <- readIORef $ msgCount s
   _ntfCount <- readIORef $ ntfCount s
@@ -439,6 +444,7 @@ getServerStatsData s = do
         _ntfServices,
         _rcvServicesSubMsg,
         _rcvServicesSubDuplicate,
+        _authDelayExceeded,
         _qCount,
         _msgCount,
         _ntfCount,
@@ -503,6 +509,7 @@ setServerStats s d = do
   setServiceStats (ntfServices s) $! _ntfServices d
   writeIORef (rcvServicesSubMsg s) $! _rcvServicesSubMsg d
   writeIORef (rcvServicesSubDuplicate s) $! _rcvServicesSubDuplicate d
+  writeIORef (authDelayExceeded s) $! _authDelayExceeded d
   writeIORef (qCount s) $! _qCount d
   writeIORef (msgCount s) $! _msgCount d
   writeIORef (ntfCount s) $! _ntfCount d
@@ -705,6 +712,7 @@ instance StrEncoding ServerStatsData where
           _ntfServices,
           _rcvServicesSubMsg = 0,
           _rcvServicesSubDuplicate = 0,
+          _authDelayExceeded = 0,
           _qCount,
           _msgCount = 0,
           _ntfCount = 0,

src/Simplex/Messaging/Transport.hs

@@ -46,7 +46,6 @@ module Simplex.Messaging.Transport
     minServerSMPRelayVersion,
     currentClientSMPRelayVersion,
     currentServerSMPRelayVersion,
-    shortLinksSMPVersion,
     serviceCertsSMPVersion,
     newNtfCredsSMPVersion,
     clientNoticesSMPVersion,
@@ -54,6 +53,7 @@ module Simplex.Messaging.Transport
     namesSMPVersion,
     serverInfoSMPVersion,
     nameAvailSMPVersion,
+    fwdNoncesSMPVersion,
     simplexMQVersion,
     smpBlockSize,
     TransportConfig (..),
@@ -177,6 +177,7 @@ smpBlockSize = 16384
 -- 20 - public namespaces resolver, RSLV command (6/20/2026)
 -- 21 - server public information in handshake (7/5/2026)
 -- 22 - RNAME answers name availability as well as the record (7/25/2026)
+-- 23 - version in forwarded command nonce, random nonce in forwarded responses (10/2/2026)
 
 data SMPVersion
 
@@ -189,11 +190,8 @@ type VersionRangeSMP = VersionRange SMPVersion
 pattern VersionSMP :: Word16 -> VersionSMP
 pattern VersionSMP v = Version v
 
-_proxyServerHandshakeSMPVersion :: VersionSMP
-_proxyServerHandshakeSMPVersion = VersionSMP 14
-
-shortLinksSMPVersion :: VersionSMP
-shortLinksSMPVersion = VersionSMP 15
+_shortLinksSMPVersion :: VersionSMP
+_shortLinksSMPVersion = VersionSMP 15
 
 serviceCertsSMPVersion :: VersionSMP
 serviceCertsSMPVersion = VersionSMP 16
@@ -218,29 +216,32 @@ serverInfoSMPVersion = VersionSMP 21
 nameAvailSMPVersion :: VersionSMP
 nameAvailSMPVersion = VersionSMP 22
 
+fwdNoncesSMPVersion :: VersionSMP
+fwdNoncesSMPVersion = VersionSMP 23
+
 minClientSMPRelayVersion :: VersionSMP
-minClientSMPRelayVersion = VersionSMP 14
+minClientSMPRelayVersion = VersionSMP 15
 
 minServerSMPRelayVersion :: VersionSMP
-minServerSMPRelayVersion = VersionSMP 14
+minServerSMPRelayVersion = VersionSMP 15
 
 currentClientSMPRelayVersion :: VersionSMP
-currentClientSMPRelayVersion = VersionSMP 22
+currentClientSMPRelayVersion = VersionSMP 23
 
 currentServerSMPRelayVersion :: VersionSMP
-currentServerSMPRelayVersion = VersionSMP 22
+currentServerSMPRelayVersion = VersionSMP 23
 
 -- Max SMP protocol version to be used in e2e encrypted connection between
 -- client and server, as defined by SMP proxy. Normally set below the current
 -- version to prevent client version fingerprinting by the destination relays
--- when clients upgrade at different times. Pinned to the current version (22)
--- for this release because a proxied RSLV only carries availability from
--- nameAvailSMPVersion (22), so the one-version anti-fingerprinting buffer does
--- not apply yet; it reappears once the current version advances past 22.
+-- when clients upgrade at different times. Pinned to the current version (23)
+-- for this release because forwarded commands use the nonces from
+-- fwdNoncesSMPVersion (23), so the one-version anti-fingerprinting buffer does
+-- not apply yet; it reappears once the current version advances past 23.
 proxiedSMPRelayVersion :: VersionSMP
-proxiedSMPRelayVersion = VersionSMP 22
+proxiedSMPRelayVersion = VersionSMP 23
 
--- minimal supported protocol version is 14
+-- minimal supported protocol version is 15
 supportedClientSMPRelayVRange :: VersionRangeSMP
 supportedClientSMPRelayVRange = mkVersionRange minClientSMPRelayVersion currentClientSMPRelayVersion
 
@@ -865,7 +866,6 @@ forceCertChain :: CertChainPubKey -> CertChainPubKey
 forceCertChain cert@(CertChainPubKey (X.CertificateChain cc) signedKey) = length (show cc) `seq` show signedKey `seq` cert
 {-# INLINE forceCertChain #-}
 
--- This function is only used with v >= 8, so currently it's a simple record update.
 -- * Note: it requires updating version-based parameters, to be consistent with smpTHandle_.
 smpTHParamsSetVersion :: VersionSMP -> THandleParams SMPVersion p -> THandleParams SMPVersion p
 smpTHParamsSetVersion v params =

src/Simplex/RemoteControl/Client.hs

@@ -26,6 +26,7 @@ module Simplex.RemoteControl.Client
     -- for tests only
     sendRCPacket,
     receiveRCPacket,
+    findRCCtrlPairing,
   ) where
 
 import Control.Applicative ((<|>))
@@ -46,7 +47,7 @@ import Data.List.NonEmpty (NonEmpty (..))
 import qualified Data.List.NonEmpty as L
 import Data.Maybe (isNothing)
 import qualified Data.Text as T
-import Data.Time.Clock.System (getSystemTime)
+import Data.Time.Clock.System (SystemTime (..), getSystemTime)
 import Data.Tuple (swap)
 import Data.Word (Word16)
 import qualified Data.X509 as X
@@ -395,8 +396,10 @@ findRCCtrlPairing :: NonEmpty RCCtrlPairing -> RCEncInvitation -> ExceptT RCErro
 findRCCtrlPairing pairings RCEncInvitation {dhPubKey, nonce, encInvitation} = do
   (pairing, signedInvStr) <- liftEither $ decrypt (L.toList pairings)
   signedInv <- liftEitherWith RCESyntax $ strDecode signedInvStr
-  inv@(RCVerifiedInvitation RCInvitation {dh = invDh}) <- maybe (throwE RCEInvitation) pure $ verifySignedInvitation signedInv
+  inv@(RCVerifiedInvitation RCInvitation {dh = invDh, ts}) <- maybe (throwE RCEInvitation) pure $ verifySignedInvitation signedInv
   unless (invDh == dhPubKey) $ throwE RCEInvitation
+  now <- systemSeconds <$> liftIO getSystemTime
+  unless (now - 3660 <= systemSeconds ts && systemSeconds ts <= now + 3600) $ throwE RCEInvitation
   pure (pairing, inv)
   where
     decrypt :: [RCCtrlPairing] -> Either RCErrorType (RCCtrlPairing, ByteString)

src/Simplex/RemoteControl/Discovery.hs

@@ -122,18 +122,22 @@ closeListener subscribers sock =
 joinMulticast :: TMVar Int -> N.Socket -> N.HostAddress -> IO ()
 joinMulticast subscribers sock group = do
   now <- atomically $ takeTMVar subscribers
-  when (now == 0) $ do
-    setMembership sock group True >>= \case
-      Left e -> atomically (putTMVar subscribers now) >> logError ("setMembership failed " <> tshow e)
-      Right () -> atomically $ putTMVar subscribers (now + 1)
+  if now == 0
+    then
+      setMembership sock group True >>= \case
+        Left e -> atomically (putTMVar subscribers now) >> logError ("setMembership failed " <> tshow e)
+        Right () -> atomically $ putTMVar subscribers (now + 1)
+    else atomically $ putTMVar subscribers (now + 1)
 
 partMulticast :: TMVar Int -> N.Socket -> N.HostAddress -> IO ()
 partMulticast subscribers sock group = do
   now <- atomically $ takeTMVar subscribers
-  when (now == 1) $
-    setMembership sock group False >>= \case
-      Left e -> atomically (putTMVar subscribers now) >> logError ("setMembership failed " <> tshow e)
-      Right () -> atomically $ putTMVar subscribers (now - 1)
+  if now == 1
+    then
+      setMembership sock group False >>= \case
+        Left e -> atomically (putTMVar subscribers (now - 1)) >> logError ("setMembership failed " <> tshow e)
+        Right () -> atomically $ putTMVar subscribers (now - 1)
+    else atomically $ putTMVar subscribers (max 0 (now - 1))
 
 listenerHostAddr4 :: UDP.ListenSocket -> N.HostAddress
 listenerHostAddr4 sock = case UDP.mySockAddr sock of

tests/AgentTests/ConnectionRequestTests.hs

@@ -20,6 +20,8 @@ module AgentTests.ConnectionRequestTests
 
 import AgentTests.EqInstances ()
 import Data.ByteString (ByteString)
+import qualified Data.ByteString.Char8 as B
+import Data.Either (isLeft)
 import Network.HTTP.Types (urlEncode)
 import Simplex.Messaging.Agent.Protocol
 import qualified Simplex.Messaging.Crypto as C
@@ -285,6 +287,9 @@ connectionRequestTests =
       contactAddressV6 #== ("https://simplex.chat/contact#/?v=1-2&smp=" <> url queueStr) -- adjusted to v6
       contactAddressV6 #== ("https://simplex.chat/contact#/?v=2-2&smp=" <> url queueStr)
       contactAddressClientData #==# ("simplex:/contact#/?v=6-8&smp=" <> url queueStr <> "&data=" <> url "{\"type\":\"group_link\", \"group_link_id\":\"abc\"}")
+    it "should reject KEM ciphertext without KEM key in e2e params" $
+      strDecode @(RcvE2ERatchetParamsUri 'C.X448) (strEncode testE2ERatchetParams <> "&kem_ct=" <> strEncode (B.replicate 1039 '\0'))
+        `shouldSatisfy` isLeft
     it "should serialize / parse queue address, connection invitations and contact addresses as binary" $ do
       smpEncodingTest queue
       smpEncodingTest queueNoQM -- this passes, no queue mode patch in SMPQueueUri encoding
@@ -357,6 +362,10 @@ connectionRequestTests =
       smpEncodingTest $ AgentServiceRequest [qInfo] Nothing "service request payload"
       smpEncodingTest $ AgentServiceResponse "service response payload"
       smpEncodingTest $ AgentRejection "rejected: not allowed"
+    it "should serialize and parse ratchet key info" $ do
+      smpDecode "R" `shouldBe` Right (AgentRatchetInfo RatchetInfo {answeredKeyHash = Nothing})
+      smpDecode "R1\3abcdef" `shouldBe` Right (AgentRatchetInfo RatchetInfo {answeredKeyHash = Just "abc"})
+      smpEncodingTest $ AgentRatchetInfo RatchetInfo {answeredKeyHash = Just "0123456789abcdef0123456789abcdef"}
   where
     smpEncodingTest :: (Encoding a, Eq a, Show a, HasCallStack) => a -> Expectation
     smpEncodingTest a = smpDecode (smpEncode a) `shouldBe` Right a

tests/AgentTests/DoubleRatchetTests.hs

@@ -53,6 +53,7 @@ doubleRatchetTests = do
       it "should propose KEM during agreement, but no shared secret" $ testAlgs testPqX3dhProposeInReply
       it "should agree shared secret using KEM" $ testAlgs testPqX3dhProposeAccept
       it "should reject proposed KEM in reply" $ testAlgs testPqX3dhProposeReject
+      it "should agree different PQ associated data with substituted KEM key" $ testAlgs testPqX3dhSubstitutedKem
       it "should allow second proposal in reply" $ testAlgs testPqX3dhProposeAgain
     describe "hybrid KEM key agreement errors" $ do
       it "should fail if reply contains acceptance without proposal" $ testAlgs testPqX3dhAcceptWithoutProposalError
@@ -362,6 +363,7 @@ testDecodeV2RatchetJSON :: IO ()
 testDecodeV2RatchetJSON = do
   let v2RatchetJSON = "{\"rcVersion\":[2,2],\"rcAD\":\"2GEJrq48TmQse6NR16I-hrI0tSySZQ57E_g46nDceAPRAiF6j0drq26RTE7be6X7uiB4RaGJGf4QRXzcYuVtWw==\",\"rcDHRs\":\"TUM0Q0FRQXdCUVlESzJWdUJDSUVJRkNYbUxtSHQ3SUNfeHpGTi1Qb3ZqTVQ3S2p6XzZlZlBjOG9fRFY2RWxKOQ==\",\"rcRK\":\"BOX2X7YW5qDSp2XknY_lqacSrtDqQNPvS6iJlZIs3G0=\",\"rcNs\":0,\"rcNr\":0,\"rcPN\":0,\"rcNHKs\":\"IMouSkXUvzT_mo0WM-pqEUK09-HTLk9WOTCFQglyQxU=\",\"rcNHKr\":\"g-tus1clYPV0rGlzkf5a959tUqDYQVZ1FpcPeXdKwxI=\"}"
   Right (r :: Ratchet X25519) <- pure $ J.eitherDecodeStrict' v2RatchetJSON
+  rcVCPQ r `shouldBe` Nothing
   rcSupportKEM r `shouldBe` PQSupportOff
   rcEnableKEM r `shouldBe` PQEncOff
   rcSndKEM r `shouldBe` PQEncOff
@@ -417,6 +419,18 @@ testPqX3dhProposeAccept _ = do
   Right paramsAlice <- runExceptT $ pqX3dhRcv pksAlice e2eBob
   paramsAlice `compatibleRatchets` paramsBob
 
+testPqX3dhSubstitutedKem :: forall a. (AlgorithmI a, DhAlgorithm a) => C.SAlgorithm a -> IO ()
+testPqX3dhSubstitutedKem _ = do
+  g <- C.newRandom
+  let v = currentE2EEncryptVersion
+  (pksAlice@(_, _, Just _), e2eAlice) <- liftIO $ generateRcvE2EParams @a g v PQSupportOn
+  (_, E2ERatchetParams _ _ _ (Just (RKParamsProposed mallorysKem))) <- liftIO $ generateRcvE2EParams @a g v PQSupportOn
+  (pksBob@(_, _, Just _), AE2ERatchetParams _ e2eBob) <- liftIO $ generateSndE2EParams @a g v (Just $ AUseKEM SRKSAccepted $ AcceptKEM mallorysKem)
+  Right (paramsBob, _) <- pure $ pqX3dhSnd pksBob e2eAlice
+  Right (paramsAlice, _) <- runExceptT $ pqX3dhRcv pksAlice e2eBob
+  assocData paramsAlice `shouldBe` assocData paramsBob
+  rcVerifyCodePQ paramsAlice `shouldNotBe` rcVerifyCodePQ paramsBob
+
 testPqX3dhProposeReject :: forall a. (AlgorithmI a, DhAlgorithm a) => C.SAlgorithm a -> IO ()
 testPqX3dhProposeReject _ = do
   g <- C.newRandom
@@ -459,9 +473,9 @@ testPqX3dhProposeAgain _ = do
 
 compatibleRatchets :: (RatchetInitParams, x) -> (RatchetInitParams, x) -> Expectation
 compatibleRatchets
-  (RatchetInitParams {assocData, ratchetKey, sndHK, rcvNextHK, kemAccepted}, _)
-  (RatchetInitParams {assocData = ad, ratchetKey = rk, sndHK = shk, rcvNextHK = rnhk, kemAccepted = ka}, _) = do
-    assocData == ad && ratchetKey == rk && sndHK == shk && rcvNextHK == rnhk `shouldBe` True
+  (RatchetInitParams {assocData, rcVerifyCodePQ, ratchetKey, sndHK, rcvNextHK, kemAccepted}, _)
+  (RatchetInitParams {assocData = ad, rcVerifyCodePQ = vcPQ, ratchetKey = rk, sndHK = shk, rcvNextHK = rnhk, kemAccepted = ka}, _) = do
+    assocData == ad && rcVerifyCodePQ == vcPQ && ratchetKey == rk && sndHK == shk && rcvNextHK == rnhk `shouldBe` True
     case (kemAccepted, ka) of
       (Just RatchetKEMAccepted {rcPQRr, rcPQRss, rcPQRct}, Just RatchetKEMAccepted {rcPQRr = pqk, rcPQRss = pqss, rcPQRct = pqct}) ->
         pqk /= rcPQRr && pqss == rcPQRss && pqct == rcPQRct `shouldBe` True

tests/AgentTests/FunctionalAPITests.hs

@@ -70,8 +70,8 @@ import Data.ByteString.Char8 (ByteString)
 import qualified Data.ByteString.Char8 as B
 import Data.Either (isRight)
 import Data.Int (Int64)
-import Data.List (find, isPrefixOf, isSuffixOf)
-import Data.List.NonEmpty (NonEmpty)
+import Data.List (find, isInfixOf, isPrefixOf, isSuffixOf)
+import Data.List.NonEmpty (NonEmpty (..))
 import qualified Data.Map as M
 import Data.Maybe (isJust, isNothing)
 import qualified Data.Set as S
@@ -87,12 +87,12 @@ import SMPAgentClient
 import SMPClient
 import Simplex.Messaging.Agent hiding (acceptContact, createConnection, deleteConnection, deleteConnections, getConnShortLink, joinConnection, sendMessage, setConnShortLink, subscribeConnection, suspendConnection)
 import qualified Simplex.Messaging.Agent as A
-import Simplex.Messaging.Agent.Client (ProtocolTestFailure (..), ProtocolTestStep (..), ServerQueueInfo (..), UserNetworkInfo (..), UserNetworkType (..), waitForUserNetwork)
+import Simplex.Messaging.Agent.Client (ProtocolTestFailure (..), ProtocolTestStep (..), ServerQueueInfo (..), UserNetworkInfo (..), UserNetworkType (..), sendAgentMessage, waitForUserNetwork)
 import Simplex.Messaging.Agent.Env.SQLite (AgentConfig (..), Env (..), InitialAgentServers (..), createAgentStore)
 import Simplex.Messaging.Agent.Protocol hiding (CON, CONF, INFO, REQ, SENT)
 import qualified Simplex.Messaging.Agent.Protocol as A
 import Simplex.Messaging.Agent.Store (Connection' (..), SomeConn' (..), StoredRcvQueue (..))
-import Simplex.Messaging.Agent.Store.AgentStore (getConn)
+import Simplex.Messaging.Agent.Store.AgentStore (deleteRatchetKeyHashesExpired, getConn, getConnIds, getRatchetX3dhKeys)
 import Simplex.Messaging.Agent.Store.Common (DBStore (..), withTransaction)
 import Simplex.Messaging.Agent.Store.Interface
 import qualified Simplex.Messaging.Agent.Store.DB as DB
@@ -112,7 +112,7 @@ import Simplex.Messaging.Server.Information (ServerPublicInfo (..))
 import Simplex.Messaging.Server.MsgStore.Types (SMSType (..), SQSType (..))
 import Simplex.Messaging.Server.QueueStore.QueueInfo
 import Simplex.Messaging.Server.StoreLog (StoreLogRecord (..))
-import Simplex.Messaging.Transport (ASrvTransport, SMPVersion, VersionSMP, currentServerSMPRelayVersion, minClientSMPRelayVersion, minServerSMPRelayVersion, alpnSupportedSMPHandshakes, supportedServerSMPRelayVRange)
+import Simplex.Messaging.Transport (ASrvTransport, SMPVersion, VersionSMP, currentServerSMPRelayVersion, minClientSMPRelayVersion, minServerSMPRelayVersion, alpnSupportedSMPHandshakes)
 import Simplex.Messaging.Transport.Server (TransportServerConfig (..))
 import Simplex.Messaging.Util (bshow, diffToMicroseconds)
 import Simplex.Messaging.Version (VersionRange (..))
@@ -204,7 +204,7 @@ pattern INFO :: ConnInfo -> AEvent 'AEConn
 pattern INFO connInfo = A.INFO PQSupportOn connInfo
 
 pattern REQ :: InvitationId -> NonEmpty SMPServer -> ConnInfo -> AEvent e
-pattern REQ invId srvs connInfo <- A.REQ invId PQSupportOn srvs connInfo _
+pattern REQ invId srvs connInfo <- A.REQ invId PQSupportOn srvs connInfo _ _
 
 pattern CON :: AEvent 'AEConn
 pattern CON = A.CON PQEncOn
@@ -231,7 +231,11 @@ pattern Rcvd' :: AgentMsgId -> AgentMsgId -> AEvent 'AEConn
 pattern Rcvd' aMsgId rcvdMsgId <- RCVD MsgMeta {integrity = MsgOk, recipient = (aMsgId, _)} [MsgReceipt {agentMsgId = rcvdMsgId, msgRcptStatus = MROk}]
 
 smpCfgVPrev :: ProtocolClientConfig SMPVersion
-smpCfgVPrev = (smpCfg agentCfg) {serverVRange = prevRange $ serverVRange $ smpCfg agentCfg}
+smpCfgVPrev =
+  (smpCfg agentCfg)
+    { serverVRange = prevRange $ serverVRange $ smpCfg agentCfg,
+      proxiedRelayVRange = prevRange $ proxiedRelayVRange $ smpCfg agentCfg
+    }
 
 -- ntfCfgVPrev :: ProtocolClientConfig NTFVersion
 -- ntfCfgVPrev = (ntfCfg agentCfg) {clientALPN = Nothing, serverVRange = V.mkVersionRange (VersionNTF 1) (VersionNTF 1)}
@@ -293,7 +297,7 @@ createConnection c userId enableNtfs cMode clientData subMode = do
 
 joinConnection :: AgentClient -> UserId -> Bool -> ConnectionRequestUri c -> ConnInfo -> SubscriptionMode -> AE (ConnId, SndQueueSecured)
 joinConnection c userId enableNtfs cReq connInfo subMode = do
-  connId <- A.prepareConnectionToJoin c userId enableNtfs cReq PQSupportOn
+  (connId, _) <- A.prepareConnectionToJoin c userId enableNtfs cReq PQSupportOn
   sndSecure <- A.joinConnection c NRMInteractive userId connId enableNtfs cReq connInfo PQSupportOn subMode
   pure (connId, sndSecure)
 
@@ -367,6 +371,10 @@ functionalAPITests ps = do
       testAddDRViaSetConnShortLink ps
     it "should resume DR accept after a transient failure (reuse the send queue and ratchet)" $
       testAcceptContactDRResumeAfterOffline ps
+    it "should store the ratchet when DR accept is prepared" $
+      withSmpServer ps $ testAcceptContactDRPrepared False
+    it "should create the ratchet at DR accept when the prepared connection has none" $
+      withSmpServer ps $ testAcceptContactDRPrepared True
     it "should support rejecting contact request" $
       withSmpServer ps testRejectContactRequest
     it "should communicate rejection reason via double ratchet" $
@@ -411,13 +419,16 @@ functionalAPITests ps = do
     describe "should connect via 1-time short link with async join" $ testProxyMatrix ps testInvitationShortLinkAsync
     describe "should connect via contact short link" $ testProxyMatrix ps testContactShortLink
     describe "should add short link to existing contact and connect" $ testProxyMatrix ps testAddContactShortLink
-    xdescribe "try to create 1-time short link with prev versions" $ testProxyMatrixWithPrev ps testInvitationShortLinkPrev
     describe "server restart" $ do
       it "should get 1-time link data after restart" $ testInvitationShortLinkRestart ps
       it "should connect via contact short link after restart" $ testContactShortLinkRestart ps
       it "should connect via added contact short link after restart" $ testAddContactShortLinkRestart ps
     it "should create and get short links with the old contact queues" $ testOldContactQueueShortLink ps
     it "should connect via prepared connection link" $ testPrepareCreateConnectionLink ps
+    it "should connect via prepared invitation link" $ testPrepareCreateInvitationLink ps
+    it "should not create connection for prepared invitation link with too large data" $ testPrepareInvitationLinkTooLarge ps
+    it "should not create connection for prepared contact link with too large data" $ testPrepareContactLinkTooLarge ps
+    it "should connect via short link prepared for existing contact connection" $ testPrepareConnShortLink ps
   describe "Message delivery" $ do
     describe "update connection agent version on received messages" $ do
       it "should increase if compatible, shouldn'ps decrease" $
@@ -442,6 +453,9 @@ functionalAPITests ps = do
       it "should expire multiple messages" $ testExpireManyMessages ps
       it "should expire one message if quota is exceeded" $ testExpireMessageQuota ps
       it "should expire multiple messages if quota is exceeded" $ testExpireManyMessagesQuota ps
+    describe "joining a full contact address" $ do
+      it "should retry joining until the address has space" $ testJoinFullContactAsync ps
+      it "should fail joining after quota exceeded timeout" $ testJoinFullContactAsyncExpire ps
     it "should drop message after too many receive attempts" $ testDropMsgAfterRcvAttempts ps
 #if !defined(dbPostgres)
     -- TODO [postgres] restore from outdated db backup (we use copyFile/renameFile for sqlite)
@@ -456,6 +470,16 @@ functionalAPITests ps = do
         testRatchetSyncSuspendForeground ps
       it "should synchronize ratchets when clients start synchronization simultaneously" $
         testRatchetSyncSimultaneous ps
+      it "should ignore replayed ratchet key after expired hashes are deleted" $
+        testRatchetSyncReplayedKey ps
+      it "should synchronize ratchets when synchronization is forced again" $
+        testRatchetSyncRepeated ps
+      it "should not mark ratchet key as processed when ratchet recreation fails" $
+        testRatchetSyncFailedKeyNotProcessed ps
+      it "should not store reply ratchet key when ratchet recreation fails" $
+        testRatchetSyncFailedRecreationNoReply ps
+      it "should not store ratchet key when starting synchronization fails" $
+        testRatchetSyncStartFailedNoKey ps
 #endif
     describe "Subscription mode OnlyCreate" $ do
       it "messages delivered only when polled" $
@@ -534,6 +558,10 @@ functionalAPITests ps = do
       testServerMatrix2 ps testFastSwitchConnection
     it "should switch delivery to the new queue when the old server is down" $
       testFastSwitchDeadOldServer ps
+    it "should drain the old queue after QEND on the new queue" $
+      testFastSwitchDrainOldQueue ps
+    it "should drain the old queue after QEND on the new queue and restart" $
+      testFastSwitchDrainOldQueueRestart ps
     describe "should switch delivery to the new queue" $
       testServerMatrix2 ps testSwitchConnection
     describe "should switch to new queue asynchronously" $
@@ -581,9 +609,9 @@ functionalAPITests ps = do
       it "should pass with correct password" $ testSMPServerConnectionTest ps auth (srv auth) `shouldReturn` Right (Just (Right testServerInformation))
       it "should fail without password" $ testSMPServerConnectionTest ps auth (srv Nothing) `shouldReturn` Left authErr
       it "should fail with incorrect password" $ testSMPServerConnectionTest ps auth (srv $ Just "wrong") `shouldReturn` Left authErr
-  describe "getRatchetAdHash" $
-    it "should return the same data for both peers" $
-      withSmpServer ps testRatchetAdHash
+  describe "getConnectionVerifyCodes" $
+    it "should return the same codes for both peers" $
+      withSmpServer ps testConnectionVerifyCodes
   describe "Delivery receipts" $ do
     it "should send and receive delivery receipt" $ withSmpServer ps testDeliveryReceipts
     it "send delivery receipts concurrently with messages" $ testDeliveryReceiptsConcurrent ps
@@ -660,19 +688,6 @@ testProxyMatrix ps runTest = do
   it "2 servers, directly" $ withSmpServers2 ps $ withAgentClientsServers2 (agentCfg, initAgentServers) (agentCfg, initAgentServers2) $ runTest False
   it "2 servers, via proxy" $ withSmpServersProxy2 ps $ withAgentClientsServers2 (agentCfg, initAgentServersProxy) (agentCfg, initAgentServersProxy2) $ runTest True
 
-testProxyMatrixWithPrev :: HasCallStack => (ASrvTransport, AStoreType) -> (Bool -> Bool -> AgentClient -> AgentClient -> IO ()) -> Spec
-testProxyMatrixWithPrev ps@(t, msType@(ASType qs _ms)) runTest = do
-  it "2 servers, directly, curr clients, prev servers" $ withSmpServers2Prev $ withAgentClientsServers2 (agentCfg, initAgentServers) (agentCfg, initAgentServers2) $ runTest False True
-  it "2 servers, via proxy, curr clients, prev servers" $ withSmpServersProxy2Prev $ withAgentClientsServers2 (agentCfg, initAgentServersProxy) (agentCfg, initAgentServersProxy2) $ runTest True True
-  it "2 servers, directly, prev clients, curr servers" $ withSmpServers2 ps $ withAgentClientsServers2 (agentCfgVPrevPQ, initAgentServers) (agentCfgVPrevPQ, initAgentServers2) $ runTest False False
-  it "2 servers, via proxy, prev clients, curr servers" $ withSmpServersProxy2 ps $ withAgentClientsServers2 (agentCfgVPrevPQ, initAgentServersProxy) (agentCfgVPrevPQ, initAgentServersProxy2) $ runTest True False
-  where
-    prev cfg' = updateCfg cfg' $ \cfg_ -> cfg_ {smpServerVRange = prevRange supportedServerSMPRelayVRange}
-    withSmpServers2Prev a = withServers2 (prev $ cfgMS msType) (prev $ cfgJ2QS qs) a
-    withSmpServersProxy2Prev a = withServers2 (prev $ proxyCfgMS msType) (prev $ proxyCfgJ2QS qs) a
-    withServers2 cfg1 cfg2 a =
-      withSmpServerConfigOn t cfg1 testPort $ \_ -> withSmpServerConfigOn t cfg2 testPort2 $ \_ -> a
-
 testPQMatrix2 :: HasCallStack => (ASrvTransport, AStoreType) -> (HasCallStack => (AgentClient, InitialKeys) -> (AgentClient, PQSupport) -> AgentMsgId -> IO ()) -> Spec
 testPQMatrix2 = pqMatrix2_ True
 
@@ -755,7 +770,7 @@ runAgentClientTestPQ :: HasCallStack => Bool -> (AgentClient, InitialKeys) -> (A
 runAgentClientTestPQ viaProxy (alice, aPQ) (bob, bPQ) baseId =
   runRight_ $ do
     (bobId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMInvitation Nothing Nothing aPQ False SMSubscribe
-    aliceId <- A.prepareConnectionToJoin bob 1 True qInfo bPQ
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo bPQ
     sqSecured' <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" bPQ SMSubscribe
     liftIO $ sqSecured' `shouldBe` True
     ("", _, A.CONF confId pqSup' _ "bob's connInfo") <- get alice
@@ -984,12 +999,12 @@ runAgentClientContactTestPQ :: HasCallStack => SndQueueSecured -> Bool -> (Agent
 runAgentClientContactTestPQ sqSecured viaProxy (alice, aPQ) (bob, bPQ) baseId =
   runRight_ $ do
     (_, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMContact Nothing Nothing aPQ False SMSubscribe
-    aliceId <- A.prepareConnectionToJoin bob 1 True qInfo bPQ
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo bPQ
     sqSecuredJoin <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" bPQ SMSubscribe
     liftIO $ sqSecuredJoin `shouldBe` False -- joining via contact address connection
-    ("", _, A.REQ invId pqSup' _ "bob's connInfo" _) <- get alice
+    ("", _, A.REQ invId pqSup' _ "bob's connInfo" _ _) <- get alice
     liftIO $ pqSup' `shouldBe` PQSupportOn
-    bobId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption aPQ)
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption aPQ)
     sqSecured' <- acceptContact alice 1 bobId True invId "alice's connInfo" (CR.connPQEncryption aPQ) SMSubscribe
     liftIO $ sqSecured' `shouldBe` sqSecured
     ("", _, A.CONF confId pqSup'' _ "alice's connInfo") <- get bob
@@ -1053,7 +1068,7 @@ runAgentClientContactDRTest_ asyncAccept asyncJoin addrIK useDR bPQ ps = withSmp
       Nothing -> expectationFailure "address must advertise DR ratchet keys"
     -- classic (non-DR) join drops the address keys from the request
     let connReqJoin = if useDR then connReq' else case connReq' of CRContactUri d _ -> CRContactUri d Nothing
-    aliceId <- A.prepareConnectionToJoin bob 1 True connReqJoin bPQ
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True connReqJoin bPQ
     if asyncJoin
       then do
         A.joinConnectionAsync bob "join" False aliceId True connReqJoin "bob's connInfo" bPQ SMSubscribe
@@ -1061,9 +1076,9 @@ runAgentClientContactDRTest_ asyncAccept asyncJoin addrIK useDR bPQ ps = withSmp
       else do
         sqSecuredJoin <- A.joinConnection bob NRMInteractive 1 aliceId True connReqJoin "bob's connInfo" bPQ SMSubscribe
         liftIO $ sqSecuredJoin `shouldBe` False
-    ("", _, A.REQ invId reqPQ _ "bob's connInfo" _) <- get alice
+    ("", _, A.REQ invId reqPQ _ "bob's connInfo" _ _) <- get alice
     liftIO $ reqPQ `shouldBe` PQSupportOn
-    bobId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
     if asyncAccept
       then do
         acceptContactAsync alice "accept" bobId True invId "alice's connInfo" (CR.connPQEncryption addrIK) SMSubscribe
@@ -1093,10 +1108,10 @@ runCreateConnectionDRTest_ asyncNew addrIK bPQ ps = withSmpServer ps $ withAgent
     liftIO $ case addrKeys_ of
       Just (_, CR.E2ERatchetParamsUri _ _ _ kem_) -> isJust kem_ `shouldBe` supportPQ (CR.initialPQEncryption False addrIK)
       Nothing -> expectationFailure "createConnection must advertise DR ratchet keys"
-    aliceId <- A.prepareConnectionToJoin bob 1 True connReq bPQ
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True connReq bPQ
     void $ A.joinConnection bob NRMInteractive 1 aliceId True connReq "bob's connInfo" bPQ SMSubscribe
-    ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
-    bobId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
+    ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
     void $ acceptContact alice 1 bobId True invId "alice's connInfo" (CR.connPQEncryption addrIK) SMSubscribe
     ("", _, A.CONF confId _ _ "alice's connInfo") <- get bob
     allowConfirmGreet alice bobId bob aliceId confId addrIK pqEnc
@@ -1118,10 +1133,10 @@ testContactDRMatrix ps = do
 
 joinContactDR :: HasCallStack => AgentClient -> AgentClient -> ConnectionRequestUri 'CMContact -> InitialKeys -> PQEncryption -> ExceptT AgentErrorType IO ()
 joinContactDR alice requester connReq addrIK pqEnc = do
-  aliceId <- A.prepareConnectionToJoin requester 1 True connReq PQSupportOn
+  (aliceId, _) <- A.prepareConnectionToJoin requester 1 True connReq PQSupportOn
   void $ A.joinConnection requester NRMInteractive 1 aliceId True connReq "bob's connInfo" PQSupportOn SMSubscribe
-  ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
-  reqId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
+  ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
+  (reqId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
   void $ acceptContact alice 1 reqId True invId "alice's connInfo" (CR.connPQEncryption addrIK) SMSubscribe
   ("", _, A.CONF confId _ _ "alice's connInfo") <- get requester
   allowConfirmGreet alice reqId requester aliceId confId addrIK pqEnc
@@ -1149,7 +1164,7 @@ testAddressKeyRotation ps = withSmpServer ps $ withAgentClients3 $ \alice bob ca
     joinContactDR alice carol connReq2 addrIK pqEnc
     void $ A.setConnShortLink alice NRMInteractive addrConnId SCMContact userLinkData Nothing True (Just addrIK)
     void $ A.setConnShortLink alice NRMInteractive addrConnId SCMContact userLinkData Nothing True (Just addrIK)
-    aId <- A.prepareConnectionToJoin bob 1 True connReq1 PQSupportOn
+    (aId, _) <- A.prepareConnectionToJoin bob 1 True connReq1 PQSupportOn
     void $ A.joinConnection bob NRMInteractive 1 aId True connReq1 "bob's connInfo" PQSupportOn SMSubscribe
     get alice =##> \case ("", _, A.ERR _) -> True; _ -> False
 
@@ -1192,11 +1207,11 @@ testAddressUpdatePreservesDRKeys ps = withSmpServer ps $ withAgentClients2 $ \al
     liftIO $ shortLink' `shouldBe` shortLink
     (_, ContactLinkData _ updated, connReq') <- getConnShortLink bob 1 shortLink
     liftIO $ ratchetKeys updated `shouldBe` ratchetKeys published
-    aliceId <- A.prepareConnectionToJoin bob 1 True connReq' bPQ
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True connReq' bPQ
     sqSecuredJoin <- A.joinConnection bob NRMInteractive 1 aliceId True connReq' "bob's connInfo" bPQ SMSubscribe
     liftIO $ sqSecuredJoin `shouldBe` False
-    ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
-    bobId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
+    ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
     _ <- acceptContact alice 1 bobId True invId "alice's connInfo" (CR.connPQEncryption addrIK) SMSubscribe
     ("", _, A.CONF confId _ _ "alice's connInfo") <- get bob
     allowConfirmGreet alice bobId bob aliceId confId addrIK pqEnc
@@ -1216,10 +1231,10 @@ testAcceptContactDRResumeAfterOffline ps = withAgentClients2 $ \alice bob -> do
     (ccLink@(CCLink _ (Just shortLink)), preparedParams) <- A.prepareConnectionLink alice 1 rootKey linkEntId True Nothing connIK True Nothing
     _ <- A.createConnectionForLink alice NRMInteractive 1 True ccLink preparedParams (UserContactLinkData userCtData) SMSubscribe
     (_, _, connReq') <- getConnShortLink bob 1 shortLink
-    aId <- A.prepareConnectionToJoin bob 1 True connReq' PQSupportOn
+    (aId, _) <- A.prepareConnectionToJoin bob 1 True connReq' PQSupportOn
     _ <- A.joinConnection bob NRMInteractive 1 aId True connReq' "bob's connInfo" PQSupportOn SMSubscribe
-    ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
-    bId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
+    ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
+    (bId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption addrIK)
     pure (bId, invId, aId)
   ("", "", DOWN _ _) <- nGet alice
   ("", "", DOWN _ _) <- nGet bob
@@ -1234,6 +1249,30 @@ testAcceptContactDRResumeAfterOffline ps = withAgentClients2 $ \alice bob -> do
     ("", _, A.CONF confId _ _ "alice's connInfo") <- get bob
     allowConfirmGreet alice bobId bob aliceId confId addrIK pqEnc
 
+testAcceptContactDRPrepared :: HasCallStack => Bool -> IO ()
+testAcceptContactDRPrepared ratchetDeleted =
+  withAgentClients2 $ \alice bob -> runRight_ $ do
+    let userLinkData = UserContactLinkData UserContactData {direct = True, owners = [], relays = [], userData = UserLinkData "test user data", ratchetKeys = Nothing}
+        pqEnc = PQEncryption $ pqConnectionMode IKPQOn PQSupportOn
+    (_, CCLink connReq _) <- A.createConnection alice NRMInteractive 1 True True SCMContact (Just userLinkData) Nothing IKPQOn True SMSubscribe
+    (aliceId, CRBRatchet codes) <- A.prepareConnectionToJoin bob 1 True connReq PQSupportOn
+    void $ A.joinConnection bob NRMInteractive 1 aliceId True connReq "bob's connInfo" PQSupportOn SMSubscribe
+    ("", _, A.REQ invId _ _ "bob's connInfo" (CRBRatchet reqCodes) _) <- get alice
+    (bobId, CRBRatchet acceptCodes) <- A.prepareConnectionToAccept alice 1 True invId PQSupportOn
+    liftIO $ do
+      reqCodes `shouldBe` codes
+      acceptCodes `shouldBe` codes
+    if ratchetDeleted
+      then do
+        liftIO $ withTransaction (store $ agentEnv alice) (`DB.execute_` "DELETE FROM ratchets")
+        Left (CONN NOT_FOUND _) <- tryError $ getConnectionVerifyCodes alice bobId
+        pure ()
+      else getConnectionVerifyCodes alice bobId >>= liftIO . (`shouldBe` codes)
+    void $ acceptContact alice 1 bobId True invId "alice's connInfo" PQSupportOn SMSubscribe
+    ("", _, A.CONF confId _ _ "alice's connInfo") <- get bob
+    allowConfirmGreet alice bobId bob aliceId confId IKPQOn pqEnc
+    getConnectionVerifyCodes alice bobId >>= liftIO . (`shouldBe` codes)
+
 runAgentClientContactTestPQ3 :: HasCallStack => Bool -> (AgentClient, InitialKeys) -> (AgentClient, PQSupport) -> (AgentClient, PQSupport) -> AgentMsgId -> IO ()
 runAgentClientContactTestPQ3 viaProxy (alice, aPQ) (bob, bPQ) (tom, tPQ) baseId = runRight_ $ do
   (_, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMContact Nothing Nothing aPQ False SMSubscribe
@@ -1244,12 +1283,12 @@ runAgentClientContactTestPQ3 viaProxy (alice, aPQ) (bob, bPQ) (tom, tPQ) baseId
   where
     msgId = subtract baseId . fst
     connectViaContact b pq qInfo = do
-      aId <- A.prepareConnectionToJoin b 1 True qInfo pq
+      (aId, _) <- A.prepareConnectionToJoin b 1 True qInfo pq
       sqSecuredJoin <- A.joinConnection b NRMInteractive 1 aId True qInfo "bob's connInfo" pq SMSubscribe
       liftIO $ sqSecuredJoin `shouldBe` False -- joining via contact address connection
-      ("", _, A.REQ invId pqSup' _ "bob's connInfo" _) <- get alice
+      ("", _, A.REQ invId pqSup' _ "bob's connInfo" _ _) <- get alice
       liftIO $ pqSup' `shouldBe` PQSupportOn
-      bId <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption aPQ)
+      (bId, _) <- A.prepareConnectionToAccept alice 1 True invId (CR.connPQEncryption aPQ)
       sqSecuredAccept <- acceptContact alice 1 bId True invId "alice's connInfo" (CR.connPQEncryption aPQ) SMSubscribe
       liftIO $ sqSecuredAccept `shouldBe` True
       ("", _, A.CONF confId pqSup'' _ "alice's connInfo") <- get b
@@ -1290,10 +1329,10 @@ testRejectContactRequest :: HasCallStack => IO ()
 testRejectContactRequest =
   withAgentClients2 $ \alice bob -> runRight_ $ do
     (_addrConnId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMContact Nothing Nothing IKPQOn False SMSubscribe
-    aliceId <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
     sqSecured <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
     liftIO $ sqSecured `shouldBe` False -- joining via contact address connection
-    ("", _, A.REQ invId PQSupportOn _ "bob's connInfo" _) <- get alice
+    ("", _, A.REQ invId PQSupportOn _ "bob's connInfo" _ _) <- get alice
     Left (A.CMD PROHIBITED _) <- tryError $ rejectContact alice NRMInteractive 1 invId (Just "no rejection without double ratchet")
     rejectContact alice NRMInteractive 1 invId Nothing
     liftIO $ noMessages bob "nothing delivered to bob"
@@ -1303,9 +1342,9 @@ testRejectContactRequestDR =
   withAgentClients2 $ \alice bob -> runRight_ $ do
     let userLinkData = UserContactLinkData UserContactData {direct = True, owners = [], relays = [], userData = UserLinkData "test user data", ratchetKeys = Nothing}
     (_addrConnId, CCLink connReq _) <- A.createConnection alice NRMInteractive 1 True True SCMContact (Just userLinkData) Nothing IKPQOn True SMSubscribe
-    aliceId <- A.prepareConnectionToJoin bob 1 True connReq PQSupportOn
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True connReq PQSupportOn
     void $ A.joinConnection bob NRMInteractive 1 aliceId True connReq "bob's connInfo" PQSupportOn SMSubscribe
-    ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
+    ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
     rejectContact alice NRMInteractive 1 invId (Just "not now")
     ("", _, A.RJCT "not now") <- get bob
     pure ()
@@ -1315,9 +1354,9 @@ testRejectContactRequestDRAsync =
   withAgentClients2 $ \alice bob -> runRight_ $ do
     let userLinkData = UserContactLinkData UserContactData {direct = True, owners = [], relays = [], userData = UserLinkData "test user data", ratchetKeys = Nothing}
     (_addrConnId, CCLink connReq _) <- A.createConnection alice NRMInteractive 1 True True SCMContact (Just userLinkData) Nothing IKPQOn True SMSubscribe
-    aliceId <- A.prepareConnectionToJoin bob 1 True connReq PQSupportOn
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True connReq PQSupportOn
     void $ A.joinConnection bob NRMInteractive 1 aliceId True connReq "bob's connInfo" PQSupportOn SMSubscribe
-    ("", _, A.REQ invId _ _ "bob's connInfo" _) <- get alice
+    ("", _, A.REQ invId _ _ "bob's connInfo" _ _) <- get alice
     rejectContactAsync alice "1" 1 invId (Just "not now")
     ("", _, A.RJCT "not now") <- get bob
     pure ()
@@ -1430,7 +1469,7 @@ testUpdateConnectionUserId =
     (connId, qInfo) <- createConnection alice 1 True SCMInvitation Nothing SMSubscribe
     newUserId <- createUser alice False [noAuthSrvCfg testSMPServer] [noAuthSrvCfg testXFTPServer]
     _ <- changeConnectionUser alice 1 connId newUserId
-    aliceId <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+    (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
     sqSecured' <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
     liftIO $ sqSecured' `shouldBe` True
     ("", _, A.CONF confId pqSup' _ "bob's connInfo") <- get alice
@@ -1589,7 +1628,7 @@ testInvitationErrors ps restart = do
   b <- getAgentB
   (bId, cReq) <- withServer1 ps $ runRight $ createConnection a 1 True SCMInvitation Nothing SMSubscribe
   ("", "", DOWN _ [_]) <- nGet a
-  aId <- runRight $ A.prepareConnectionToJoin b 1 True cReq PQSupportOn
+  (aId, _) <- runRight $ A.prepareConnectionToJoin b 1 True cReq PQSupportOn
   -- fails to secure the queue on testPort
   BROKER srv (NETWORK _) <- runLeft $ A.joinConnection b NRMInteractive 1 aId True cReq "bob's connInfo" PQSupportOn SMSubscribe
   (testPort `isSuffixOf` srv) `shouldBe` True
@@ -1659,7 +1698,7 @@ testContactErrors ps restart = do
   b <- getAgentB
   (contactId, cReq) <- withServer1 ps $ runRight $ createConnection a 1 True SCMContact Nothing SMSubscribe
   ("", "", DOWN _ [_]) <- nGet a
-  aId <- runRight $ A.prepareConnectionToJoin b 1 True cReq PQSupportOn
+  (aId, _) <- runRight $ A.prepareConnectionToJoin b 1 True cReq PQSupportOn
   -- fails to create queue on testPort2
   BROKER srv2 (NETWORK _) <- runLeft $ A.joinConnection b NRMInteractive 1 aId True cReq "bob's connInfo" PQSupportOn SMSubscribe
   (testPort2 `isSuffixOf` srv2) `shouldBe` True
@@ -1681,10 +1720,10 @@ testContactErrors ps restart = do
             Right r -> error $ "unexpected result " <> show r
             Left _ -> putStrLn "retrying send" >> threadDelay 200000 >> loopSend
     loopSend
-    ("", _, A.REQ invId PQSupportOn _ "bob's connInfo" _) <- get a
+    ("", _, A.REQ invId PQSupportOn _ "bob's connInfo" _ _) <- get a
     pure invId
   ("", "", DOWN _ [_]) <- nGet a
-  bId <- runRight $ A.prepareConnectionToAccept a 1 True invId PQSupportOn
+  (bId, _) <- runRight $ A.prepareConnectionToAccept a 1 True invId PQSupportOn
   withServer2 ps $ do
     ("", "", UP _ [_]) <- nGet b''
     let loopSecure = do
@@ -1764,7 +1803,7 @@ testInvitationShortLink viaProxy a b =
 
 testJoinConn_ :: Bool -> Bool -> AgentClient -> ConnId -> AgentClient -> ConnectionRequestUri c -> ExceptT AgentErrorType IO ()
 testJoinConn_ viaProxy sndSecure a bId b connReq = do
-  aId <- A.prepareConnectionToJoin b 1 True connReq PQSupportOn
+  (aId, _) <- A.prepareConnectionToJoin b 1 True connReq PQSupportOn
   sndSecure' <- A.joinConnection b NRMInteractive 1 aId True connReq "bob's connInfo" PQSupportOn SMSubscribe
   liftIO $ sndSecure' `shouldBe` sndSecure
   ("", _, CONF confId _ "bob's connInfo") <- get a
@@ -1774,14 +1813,6 @@ testJoinConn_ viaProxy sndSecure a bId b connReq = do
   get b ##> ("", aId, CON)
   exchangeGreetingsViaProxy viaProxy a bId b aId
 
-testInvitationShortLinkPrev :: HasCallStack => Bool -> Bool -> AgentClient -> AgentClient -> IO ()
-testInvitationShortLinkPrev viaProxy sndSecure a b = runRight_ $ do
-  let userData = UserLinkData "some user data"
-      newLinkData = UserInvLinkData userData
-  -- can't create short link with previous version
-  (bId, CCLink connReq Nothing) <- A.createConnection a NRMInteractive 1 True True SCMInvitation (Just newLinkData) Nothing CR.IKPQOn False SMSubscribe
-  testJoinConn_ viaProxy sndSecure a bId b connReq
-
 testInvitationShortLinkAsync :: HasCallStack => Bool -> AgentClient -> AgentClient -> IO ()
 testInvitationShortLinkAsync viaProxy a b = do
   let userData = UserLinkData "some user data"
@@ -1792,7 +1823,7 @@ testInvitationShortLinkAsync viaProxy a b = do
   connReq' `shouldBe` connReq
   linkUserData connData' `shouldBe` userData
   runRight $ do
-    aId <- A.prepareConnectionToJoin b 1 True connReq PQSupportOn
+    (aId, _) <- A.prepareConnectionToJoin b 1 True connReq PQSupportOn
     A.joinConnectionAsync b "123" False aId True connReq "bob's connInfo" PQSupportOn SMSubscribe
     get b =##> \case ("123", c, JOINED sndSecure) -> c == aId && sndSecure; _ -> False
     ("", _, CONF confId _ "bob's connInfo") <- get a
@@ -1834,7 +1865,7 @@ testContactShortLink viaProxy a b =
       (aId, sndSecure) <- joinConnection b 1 True connReq "bob's connInfo" SMSubscribe
       liftIO $ sndSecure `shouldBe` False
       ("", _, REQ invId _ "bob's connInfo") <- get a
-      bId <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
+      (bId, _) <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
       sndSecure' <- acceptContact a 1 bId True invId "alice's connInfo" PQSupportOn SMSubscribe
       liftIO $ sndSecure' `shouldBe` True
       ("", _, CONF confId  _ "alice's connInfo") <- get b
@@ -1887,7 +1918,7 @@ testAddContactShortLink viaProxy a b =
       (aId, sndSecure) <- joinConnection b 1 True connReq "bob's connInfo" SMSubscribe
       liftIO $ sndSecure `shouldBe` False
       ("", _, REQ invId _ "bob's connInfo") <- get a
-      bId <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
+      (bId, _) <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
       sndSecure' <- acceptContact a 1 bId True invId "alice's connInfo" PQSupportOn SMSubscribe
       liftIO $ sndSecure' `shouldBe` True
       ("", _, CONF confId  _ "alice's connInfo") <- get b
@@ -2046,7 +2077,7 @@ testPrepareCreateConnectionLink ps = withSmpServer ps $ withAgentClients2 $ \a b
     (bId, sndSecure) <- joinConnection b 1 True connReq' "bob's connInfo" SMSubscribe
     liftIO $ sndSecure `shouldBe` False
     ("", _, REQ invId _ "bob's connInfo") <- get a
-    aId <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
+    (aId, _) <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
     sndSecure' <- acceptContact a 1 aId True invId "alice's connInfo" PQSupportOn SMSubscribe
     liftIO $ sndSecure' `shouldBe` True
     ("", _, CONF confId _ "alice's connInfo") <- get b
@@ -2056,6 +2087,64 @@ testPrepareCreateConnectionLink ps = withSmpServer ps $ withAgentClients2 $ \a b
     get b ##> ("", bId, CON)
     exchangeGreetings a aId b bId
 
+testPrepareCreateInvitationLink :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testPrepareCreateInvitationLink ps = withSmpServer ps $ withAgentClients2 $ \a b -> do
+  let userData = UserLinkData "test user data"
+  runRight_ $ do
+    (preparedReq, preparedParams@PreparedLinkParams {plpLinkKey}) <- A.prepareInvitationLink a 1 CR.IKPQOn Nothing
+    (bId, CCLink connReq (Just shortLink@(CSLInvitation _ _ _ linkKey))) <- A.createInvitationForLink a NRMInteractive 1 True preparedReq preparedParams (UserInvLinkData userData) SMSubscribe
+    liftIO $ linkKey `shouldBe` plpLinkKey
+    (_, connData', connReq') <- getConnShortLink b 1 shortLink
+    liftIO $ connReq' `shouldBe` preparedReq
+    liftIO $ connReq `shouldNotBe` preparedReq
+    liftIO $ linkUserData connData' `shouldBe` userData
+    testJoinConn_ False True a bId b connReq
+
+testPrepareInvitationLinkTooLarge :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testPrepareInvitationLinkTooLarge ps = withSmpServer ps $ withAgent 1 agentCfg initAgentServers testDB $ \a -> do
+  runRight_ $ do
+    (preparedReq, preparedParams) <- A.prepareInvitationLink a 1 CR.IKPQOn Nothing
+    Left (A.CMD LARGE _) <- tryError $ A.createInvitationForLink a NRMInteractive 1 True preparedReq preparedParams (UserInvLinkData $ UserLinkData $ B.replicate 14000 'a') SMSubscribe
+    pure ()
+  withTransaction (store $ agentEnv a) getConnIds `shouldReturn` []
+
+testPrepareContactLinkTooLarge :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testPrepareContactLinkTooLarge ps = withSmpServer ps $ withAgent 1 agentCfg initAgentServers testDB $ \a -> do
+  let userCtData = UserContactData {direct = True, owners = [], relays = [], userData = UserLinkData $ B.replicate 14000 'a', ratchetKeys = Nothing}
+  g <- C.newRandom
+  rootKey <- atomically $ C.generateKeyPair g
+  linkEntId <- atomically $ C.randomBytes 32 g
+  runRight_ $ do
+    (ccLink, preparedParams) <- A.prepareConnectionLink a 1 rootKey linkEntId True Nothing CR.IKPQOn True Nothing
+    Left (A.CMD LARGE _) <- tryError $ A.createConnectionForLink a NRMInteractive 1 True ccLink preparedParams (UserContactLinkData userCtData) SMSubscribe
+    pure ()
+  withTransaction (store $ agentEnv a) getConnIds `shouldReturn` []
+
+testPrepareConnShortLink :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testPrepareConnShortLink ps = withSmpServer ps $ withAgentClients2 $ \a b -> do
+  let userCtData = UserContactData {direct = True, owners = [], relays = [], userData = UserLinkData "test user data", ratchetKeys = Nothing}
+  runRight_ $ do
+    (contactId, CCLink _ Nothing) <- A.createConnection a NRMInteractive 1 True True SCMContact Nothing Nothing CR.IKPQOn False SMSubscribe
+    shortLink <- A.prepareConnShortLink a contactId Nothing
+    shortLink' <- A.prepareConnShortLink a contactId Nothing
+    liftIO $ shortLink' `shouldBe` shortLink
+    shortLink'' <- A.setConnShortLink a NRMInteractive contactId SCMContact (UserContactLinkData userCtData) Nothing False Nothing
+    liftIO $ shortLink'' `shouldBe` shortLink
+    (_, ContactLinkData _ userCtData', connReq) <- getConnShortLink b 1 shortLink
+    liftIO $ userCtData' `shouldBe` userCtData
+    (aId, sndSecure) <- joinConnection b 1 True connReq "bob's connInfo" SMSubscribe
+    liftIO $ sndSecure `shouldBe` False
+    ("", _, REQ invId _ "bob's connInfo") <- get a
+    (bId, _) <- A.prepareConnectionToAccept a 1 True invId PQSupportOn
+    sndSecure' <- acceptContact a 1 bId True invId "alice's connInfo" PQSupportOn SMSubscribe
+    liftIO $ sndSecure' `shouldBe` True
+    ("", _, CONF confId _ "alice's connInfo") <- get b
+    allowConnection b aId confId "bob's connInfo"
+    get a ##> ("", bId, INFO "bob's connInfo")
+    get a ##> ("", bId, CON)
+    get b ##> ("", aId, CON)
+    exchangeGreetings a bId b aId
+
 connReqWithKeys :: BinaryConnectionRequestUri m -> Maybe AddressRatchetKeys -> ConnectionRequestUri m
 connReqWithKeys cr rk = case cr of
   BCRInvitationUri crData e2eParams -> CRInvitationUri crData e2eParams
@@ -2484,6 +2573,49 @@ testExpireManyMessagesQuota (t, msType) = withSmpServerConfigOn t cfg' testPort
   where
     cfg' = updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 1, maxJournalMsgCount = 2}
 
+testJoinFullContactAsync :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testJoinFullContactAsync (t, msType) = withSmpServerConfigOn t cfg' testPort $ \_ -> do
+  (contactId, qInfo) <- fillContactAddress
+  withAgent 2 agentCfg {commandQuotaRetryInterval = fastRetryInterval} initAgentServers testDB2 $ \bob -> do
+    aliceId <- runRight $ do
+      (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+      A.joinConnectionAsync bob "2" False aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
+      pure aliceId
+    threadDelay 500000
+    noMessages bob "joining should be retried while the address is full"
+    withAgent 1 agentCfg initAgentServers testDB $ \alice -> runRight_ $ do
+      subscribeConnection alice contactId
+      ("", _, A.REQ _ _ _ "carol's connInfo" _ _) <- get alice
+      get bob =##> \case ("2", c, JOINED False) -> c == aliceId; _ -> False
+      ("", _, A.REQ _ _ _ "bob's connInfo" _ _) <- get alice
+      pure ()
+  where
+    cfg' = updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 1, maxJournalMsgCount = 2}
+
+testJoinFullContactAsyncExpire :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testJoinFullContactAsyncExpire (t, msType) = withSmpServerConfigOn t cfg' testPort $ \_ -> do
+  (_, qInfo) <- fillContactAddress
+  withAgent 2 agentCfg {quotaExceededTimeout = 2, commandQuotaRetryInterval = fastRetryInterval} initAgentServers testDB2 $ \bob -> do
+    aliceId <- runRight $ do
+      (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+      A.joinConnectionAsync bob "2" False aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
+      pure aliceId
+    threadDelay 500000
+    noMessages bob "joining should be retried until quota exceeded timeout"
+    get bob =##> \case ("2", c, ERR (SMP _ QUOTA)) -> c == aliceId; _ -> False
+  where
+    cfg' = updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 1, maxJournalMsgCount = 2}
+
+fillContactAddress :: HasCallStack => IO (ConnId, ConnectionRequestUri 'CMContact)
+fillContactAddress = do
+  (contactId, qInfo) <- withAgent 1 agentCfg initAgentServers testDB $ \alice -> runRight $ do
+    (contactId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMContact Nothing Nothing IKPQOn False SMSubscribe
+    pure (contactId, qInfo)
+  withAgent 3 agentCfg initAgentServers testDB3 $ \carol -> runRight_ $ do
+    (aliceId, _) <- A.prepareConnectionToJoin carol 1 True qInfo PQSupportOn
+    void $ A.joinConnection carol NRMInteractive 1 aliceId True qInfo "carol's connInfo" PQSupportOn SMSubscribe
+  pure (contactId, qInfo)
+
 testDropMsgAfterRcvAttempts :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
 testDropMsgAfterRcvAttempts ps =
   withSmpServerStoreLogOn ps testPort $ \_ -> do
@@ -2698,6 +2830,117 @@ testRatchetSyncSimultaneous ps = do
   disposeAgentClient bob
   disposeAgentClient bob2
 
+testRatchetSyncReplayedKey :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testRatchetSyncReplayedKey ps = withAgentClients2 $ \alice bob -> do
+  (aliceId, bobId, bob2) <- withSmpServerStoreMsgLogOn ps testPort $ \_ ->
+    setupDesynchronizedRatchet alice bob
+  ("", "", DOWN _ _) <- nGet alice
+  ("", "", DOWN _ _) <- nGet bob2
+  _ <- runRight $ synchronizeRatchet bob2 aliceId PQSupportOn False
+  Right pks <- withTransaction (store $ agentEnv bob2) (`getRatchetX3dhKeys` aliceId)
+  Right (SomeConn _ (DuplexConnection _ _ (sq :| _))) <- withTransaction (store $ agentEnv bob2) (`getConn` aliceId)
+  withSmpServerStoreMsgLogOn ps testPort $ \_ -> do
+    concurrently_
+      (getInAnyOrder alice [ratchetSyncP' bobId RSAgreed, serverUpP])
+      (getInAnyOrder bob2 [ratchetSyncP' aliceId RSAgreed, serverUpP])
+    get alice =##> ratchetSyncP bobId RSOk
+    get bob2 =##> ratchetSyncP aliceId RSOk
+    withTransaction (store $ agentEnv alice) $ \db -> deleteRatchetKeyHashesExpired db 0 100
+    let keyMsg = AgentRatchetKey {agentVersion = currentSMPAgentVersion, e2eEncryption = CR.mkRcvE2ERatchetParams CR.currentE2EEncryptVersion pks, info = ""}
+    Right _ <- runReaderT (runExceptT $ sendAgentMessage bob2 sq SMP.noMsgFlags $ smpEncode keyMsg) (agentEnv bob2)
+    runRight_ $ exchangeGreetingsMsgIds alice bobId 10 bob2 aliceId 7
+  disposeAgentClient bob2
+
+testRatchetSyncRepeated :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testRatchetSyncRepeated ps = withAgentClients2 $ \alice bob -> do
+  (aliceId, bobId, bob2) <- startRatchetSyncOffline ps alice bob
+  ConnectionStats {ratchetSyncState = rss2} <- runRight $ synchronizeRatchet bob2 aliceId PQSupportOn True
+  rss2 `shouldBe` RSStarted
+
+  withSmpServerStoreMsgLogOn ps testPort $ \_ -> do
+    concurrently_
+      (getInAnyOrder alice [ratchetSyncP' bobId RSAgreed, serverUpP])
+      (getInAnyOrder bob2 [ratchetSyncP' aliceId RSAgreed, serverUpP])
+    runRight_ $ do
+      get alice =##> ratchetSyncP bobId RSAgreed
+      get alice =##> ratchetSyncP bobId RSOk
+      get bob2 =##> ratchetSyncP aliceId RSOk
+      msgId <- sendMessage alice bobId SMP.noMsgFlags "hello"
+      get alice ##> ("", bobId, SENT msgId)
+      get bob2 =##> \case ("", c, Msg "hello") -> c == aliceId; _ -> False
+      ackMessage bob2 aliceId 8 Nothing
+  map fst <$> processedRatchetKeyHashes bob2 `shouldReturn` [aliceId, aliceId]
+  disposeAgentClient bob2
+
+testRatchetSyncFailedKeyNotProcessed :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testRatchetSyncFailedKeyNotProcessed ps = withAgentClients2 $ \alice bob -> do
+  (aliceId, bobId, bob2) <- startRatchetSyncOffline ps alice bob
+  withTransaction (store $ agentEnv bob2) $ \db ->
+    DB.execute_ db "UPDATE ratchets SET x3dh_priv_key_1 = NULL"
+
+  withSmpServerStoreMsgLogOn ps testPort $ \_ ->
+    concurrently_
+      (getInAnyOrder alice [ratchetSyncP' bobId RSAgreed, serverUpP])
+      (getInAnyOrder bob2 [x3dhKeysNotFoundP aliceId, serverUpP])
+  map fst <$> processedRatchetKeyHashes alice `shouldReturn` [bobId]
+  processedRatchetKeyHashes bob2 `shouldReturn` []
+  disposeAgentClient bob2
+  where
+    x3dhKeysNotFoundP :: ConnId -> ATransmission -> Bool
+    x3dhKeysNotFoundP cId = \case
+      (_, cId', AEvt SAEConn (ERR (A.INTERNAL e))) -> cId' == cId && "SEX3dhKeysNotFound" `isPrefixOf` e
+      _ -> False
+
+testRatchetSyncFailedRecreationNoReply :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testRatchetSyncFailedRecreationNoReply ps = withAgentClients2 $ \alice bob -> do
+  (_, bobId, bob2) <- startRatchetSyncOffline ps alice bob
+  aliceSndMsgs <- sndMessages alice
+  withTransaction (store $ agentEnv alice) $ \db ->
+    DB.execute_ db "CREATE TRIGGER fail_ratchet_insert BEFORE INSERT ON ratchets BEGIN SELECT RAISE(ABORT, 'ratchet insert failed'); END"
+  withSmpServerStoreMsgLogOn ps testPort $ \_ ->
+    concurrently_
+      (getInAnyOrder alice [ratchetInsertFailedP bobId, serverUpP])
+      (getInAnyOrder bob2 [serverUpP])
+  processedRatchetKeyHashes alice `shouldReturn` []
+  sndMessages alice `shouldReturn` aliceSndMsgs
+  disposeAgentClient bob2
+  where
+    ratchetInsertFailedP :: ConnId -> ATransmission -> Bool
+    ratchetInsertFailedP cId = \case
+      (_, cId', AEvt SAEConn (ERR (A.INTERNAL e))) -> cId' == cId && "ratchet insert failed" `isInfixOf` e
+      _ -> False
+
+testRatchetSyncStartFailedNoKey :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testRatchetSyncStartFailedNoKey ps = withAgentClients2 $ \alice bob -> do
+  (aliceId, _, bob2) <- withSmpServerStoreMsgLogOn ps testPort $ \_ ->
+    setupDesynchronizedRatchet alice bob
+  bobSndMsgs <- sndMessages bob2
+  withTransaction (store $ agentEnv bob2) $ \db ->
+    DB.execute_ db "CREATE TRIGGER fail_ratchet_update BEFORE UPDATE ON ratchets BEGIN SELECT RAISE(ABORT, 'ratchet update failed'); END"
+  Left (A.INTERNAL e) <- runExceptT $ synchronizeRatchet bob2 aliceId PQSupportOff False
+  e `shouldContain` "ratchet update failed"
+  ConnectionStats {ratchetSyncState} <- runRight $ getConnectionServers bob2 aliceId
+  ratchetSyncState `shouldBe` RSRequired
+  withTransaction (store $ agentEnv bob2) (`DB.query_` "SELECT conn_id, pq_support FROM connections") `shouldReturn` [(aliceId, PQSupportOn)]
+  sndMessages bob2 `shouldReturn` bobSndMsgs
+  disposeAgentClient bob2
+
+startRatchetSyncOffline :: HasCallStack => (ASrvTransport, AStoreType) -> AgentClient -> AgentClient -> IO (ConnId, ConnId, AgentClient)
+startRatchetSyncOffline ps alice bob = do
+  (aliceId, bobId, bob2) <- withSmpServerStoreMsgLogOn ps testPort $ \_ ->
+    setupDesynchronizedRatchet alice bob
+  ("", "", DOWN _ _) <- nGet alice
+  ("", "", DOWN _ _) <- nGet bob2
+  ConnectionStats {ratchetSyncState} <- runRight $ synchronizeRatchet bob2 aliceId PQSupportOn False
+  ratchetSyncState `shouldBe` RSStarted
+  pure (aliceId, bobId, bob2)
+
+processedRatchetKeyHashes :: AgentClient -> IO [(ConnId, ByteString)]
+processedRatchetKeyHashes c = withTransaction (store $ agentEnv c) (`DB.query_` "SELECT conn_id, hash FROM processed_ratchet_key_hashes")
+
+sndMessages :: AgentClient -> IO [(ConnId, Int64)]
+sndMessages c = withTransaction (store $ agentEnv c) (`DB.query_` "SELECT conn_id, internal_id FROM snd_messages ORDER BY conn_id, internal_id")
+
 getMsg :: AgentClient -> ConnId -> ExceptT AgentErrorType IO a -> ExceptT AgentErrorType IO a
 getMsg c cId action = do
   liftIO $ noMessages c "nothing should be delivered before GET"
@@ -2749,7 +2992,7 @@ makeConnectionForUsers = makeConnectionForUsers_ PQSupportOn
 makeConnectionForUsers_ :: HasCallStack => PQSupport -> AgentClient -> UserId -> AgentClient -> UserId -> ExceptT AgentErrorType IO (ConnId, ConnId)
 makeConnectionForUsers_ pqSupport alice aliceUserId bob bobUserId = do
   (bobId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive aliceUserId True True SCMInvitation Nothing Nothing (IKLinkPQ pqSupport) False SMSubscribe
-  aliceId <- A.prepareConnectionToJoin bob bobUserId True qInfo pqSupport
+  (aliceId, _) <- A.prepareConnectionToJoin bob bobUserId True qInfo pqSupport
   sqSecured' <- A.joinConnection bob NRMInteractive bobUserId aliceId True qInfo "bob's connInfo" pqSupport SMSubscribe
   liftIO $ sqSecured' `shouldBe` True
   ("", _, A.CONF confId pqSup' _ "bob's connInfo") <- get alice
@@ -3031,7 +3274,7 @@ testAsyncCommands alice bob baseId =
     createConnectionAsync alice "1" bobId True SCMInvitation IKPQOn False SMSubscribe
     ("1", bobId', INV (ACR _ qInfo)) <- get alice
     liftIO $ bobId' `shouldBe` bobId
-    aliceId <- prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+    (aliceId, _) <- prepareConnectionToJoin bob 1 True qInfo PQSupportOn
     joinConnectionAsync bob "2" False aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
     ("2", aliceId', JOINED sqSecured') <- get bob
     liftIO $ do
@@ -3101,7 +3344,7 @@ testSetConnShortLinkAsync ps = withAgentClients2 $ \alice bob ->
     -- complete connection via contact address
     (aliceId, _) <- joinConnection bob 1 True qInfo "bob's connInfo" SMSubscribe
     ("", _, REQ invId _ "bob's connInfo") <- get alice
-    bobId <- A.prepareConnectionToAccept alice 1 True invId PQSupportOn
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId PQSupportOn
     _ <- acceptContact alice 1 bobId True invId "alice's connInfo" PQSupportOn SMSubscribe
     ("", _, CONF confId _ "alice's connInfo") <- get bob
     allowConnection bob aliceId confId "bob's connInfo"
@@ -3129,7 +3372,7 @@ testGetConnShortLinkAsync ps = withAgentClients2 $ \alice bob ->
     liftIO $ aliceId' `shouldBe` aliceId
     -- complete connection
     ("", _, REQ invId _ "bob's connInfo") <- get alice
-    bobId <- A.prepareConnectionToAccept alice 1 True invId PQSupportOn
+    (bobId, _) <- A.prepareConnectionToAccept alice 1 True invId PQSupportOn
     _ <- acceptContact alice 1 bobId True invId "alice's connInfo" PQSupportOn SMSubscribe
     ("", _, CONF confId _ "alice's connInfo") <- get bob
     allowConnection bob aliceId confId "bob's connInfo"
@@ -3159,7 +3402,7 @@ testAcceptContactAsync alice bob baseId =
     (aliceId, sqSecuredJoin) <- joinConnection bob 1 True qInfo "bob's connInfo" SMSubscribe
     liftIO $ sqSecuredJoin `shouldBe` False -- joining via contact address connection
     ("", _, REQ invId _ "bob's connInfo") <- get alice
-    bobId <- prepareConnectionToAccept alice 1 True invId PQSupportOn
+    (bobId, _) <- prepareConnectionToAccept alice 1 True invId PQSupportOn
     acceptContactAsync alice "1" bobId True invId "alice's connInfo" PQSupportOn SMSubscribe
     get alice =##> \case ("1", c, JOINED sqSecured') -> c == bobId && sqSecured' == True; _ -> False
     ("", _, CONF confId _ "alice's connInfo") <- get bob
@@ -3440,7 +3683,7 @@ testJoinConnectionAsyncReplyError ps@(t, ASType qsType _) = do
         createConnectionAsync a "1" bId True SCMInvitation IKPQOn False SMSubscribe
         ("1", bId', INV (ACR _ qInfo)) <- get a
         liftIO $ bId' `shouldBe` bId
-        aId <- prepareConnectionToJoin b 1 True qInfo PQSupportOn
+        (aId, _) <- prepareConnectionToJoin b 1 True qInfo PQSupportOn
         joinConnectionAsync b "2" False aId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
         liftIO $ threadDelay 500000
         ConnectionStats {rcvQueuesInfo = [], sndQueuesInfo = [SndQueueInfo {}]} <- getConnectionServers b aId
@@ -3580,7 +3823,7 @@ testFastSwitchDeadOldServer :: HasCallStack => (ASrvTransport, AStoreType) -> IO
 testFastSwitchDeadOldServer ps@(t, ASType qsType _) = do
   let bServers = initAgentServers {smp = userServers [testSMPServer2]}
   withSmpServerConfigOn t (cfgJ2QS qsType) testPort2 $ \_ ->
-    withAgent 1 agentCfg initAgentServers testDB $ \a ->
+    withAgent 1 aCfg initAgentServers testDB $ \a ->
       withAgent 2 agentCfg bServers testDB2 $ \b -> do
         (aId, bId) <- withSmpServerStoreLogOn ps testPort $ \_ -> runRight $ do
           (aId, bId) <- makeConnection a b
@@ -3596,7 +3839,10 @@ testFastSwitchDeadOldServer ps@(t, ASType qsType _) = do
           queuedReceived <- drainSwitchCompletedRcvMsg a bId "queued while down"
           liftIO $ queuedReceived `shouldBe` True
           drainSwitchCompleted b aId QDSnd
+          waitOneRcvQueue a bId
           exchangeGreetingsMsgId 7 a bId b aId
+  where
+    aCfg = agentCfg {initialCleanupDelay = 10000, cleanupInterval = 10000, deleteErrorCount = 3}
 
 -- drains switch and network events until the connection reports SPCompleted in the given direction,
 -- tolerating DOWN/UP and intermediate phases (the old server is stopped mid-rotation)
@@ -3616,6 +3862,69 @@ drainSwitchCompletedRcvMsg c connId body = go False
         (_, connId', AEvt SAEConn (Msg' mId _ body')) | connId' == connId && body' == body -> ackMessage c connId' mId Nothing >> go True
         _ -> go seen
 
+testFastSwitchDrainOldQueue :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testFastSwitchDrainOldQueue ps =
+  withSmpServers2 ps $ withAgentClients2 $ \a b -> runRight_ $ do
+    (aId, bId, aM1Id) <- endSwitchOnNewQueue a b
+    ackMessage a bId aM1Id Nothing
+    drainOldQueue a bId b aId
+
+testFastSwitchDrainOldQueueRestart :: HasCallStack => (ASrvTransport, AStoreType) -> IO ()
+testFastSwitchDrainOldQueueRestart ps =
+  withSmpServers2 ps $
+    withAgent 2 agentCfg initAgentServers testDB2 $ \b -> do
+      (aId, bId, _) <- withAgent 1 agentCfg initAgentServers testDB $ \a -> runRight $ endSwitchOnNewQueue a b
+      withAgent 1 agentCfg initAgentServers testDB $ \a -> runRight_ $ do
+        subscribeConnection a bId
+        drainOldQueue a bId b aId
+
+endSwitchOnNewQueue :: HasCallStack => AgentClient -> AgentClient -> ExceptT AgentErrorType IO (ConnId, ConnId, AgentMsgId)
+endSwitchOnNewQueue a b = do
+  (aId, bId) <- makeConnection a b
+  exchangeGreetings a bId b aId
+  liftIO $ setProtocolServers a 1 [noAuthSrvCfg testSMPServer2]
+  m1Id <- sendMessage b aId SMP.noMsgFlags "m1"
+  get b ##> ("", aId, SENT m1Id)
+  ("", c, Msg' aM1Id _ "m1") <- get a
+  liftIO $ c `shouldBe` bId
+  m2Id <- sendMessage b aId SMP.noMsgFlags "m2"
+  get b ##> ("", aId, SENT m2Id)
+  _ <- switchConnectionAsync a "" bId
+  fastSwitchComplete a bId b aId
+  stats <- getConnectionServers a bId
+  liftIO $ rcvSwchStatuses' stats `shouldMatchList` [Nothing, Just RSReceivedQEND]
+  pure (aId, bId, aM1Id)
+
+drainOldQueue :: HasCallStack => AgentClient -> ConnId -> AgentClient -> ConnId -> ExceptT AgentErrorType IO ()
+drainOldQueue a bId b aId = do
+  ("", c, MSG MsgMeta {recipient = (aM2Id, _)} _ "m2") <- get a
+  liftIO $ c `shouldBe` bId
+  ackMessage a bId aM2Id Nothing
+  waitOneRcvQueue a bId
+  bMsgId <- sendMessage b aId SMP.noMsgFlags "hello"
+  get b ##> ("", aId, SENT bMsgId)
+  ("", c', MSG MsgMeta {recipient = (aMsgId, _)} _ "hello") <- get a
+  liftIO $ c' `shouldBe` bId
+  ackMessage a bId aMsgId Nothing
+  aMsgId' <- sendMessage a bId SMP.noMsgFlags "hello too"
+  get a ##> ("", bId, SENT aMsgId')
+  ("", c'', MSG MsgMeta {recipient = (bMsgId', _)} _ "hello too") <- get b
+  liftIO $ c'' `shouldBe` aId
+  ackMessage b aId bMsgId' Nothing
+  liftIO $ noMessages a "nothing else should be delivered to alice"
+  liftIO $ noMessages b "nothing else should be delivered to bob"
+
+waitOneRcvQueue :: HasCallStack => AgentClient -> ConnId -> ExceptT AgentErrorType IO ()
+waitOneRcvQueue c connId = go (300 :: Int)
+  where
+    go n = do
+      stats <- getConnectionServers c connId
+      case rcvSwchStatuses' stats of
+        [Nothing] -> pure ()
+        statuses
+          | n > 0 -> liftIO (threadDelay 100000) >> go (n - 1)
+          | otherwise -> liftIO $ expectationFailure $ "old receive queue was not deleted: " <> show statuses
+
 phaseRcv :: AgentClient -> ByteString -> SwitchPhase -> [Maybe RcvSwitchStatus] -> ExceptT AgentErrorType IO ()
 phaseRcv c connId p swchStatuses = phase c connId QDRcv p (\stats -> rcvSwchStatuses' stats `shouldMatchList` swchStatuses)
 
@@ -4023,13 +4332,23 @@ testServerInformation =
       serverCountry = Nothing
     }
 
-testRatchetAdHash :: HasCallStack => IO ()
-testRatchetAdHash =
+testConnectionVerifyCodes :: HasCallStack => IO ()
+testConnectionVerifyCodes =
   withAgentClients2 $ \a b -> runRight_ $ do
     (aId, bId) <- makeConnection a b
-    ad1 <- getConnectionRatchetAdHash a bId
-    ad2 <- getConnectionRatchetAdHash b aId
-    liftIO $ ad1 `shouldBe` ad2
+    codes1 <- getConnectionVerifyCodes a bId
+    codes2 <- getConnectionVerifyCodes b aId
+    liftIO $ do
+      codes1 `shouldBe` codes2
+      codePQ codes1 `shouldNotBe` Nothing
+    liftIO $ withTransaction (store $ agentEnv a) $ \db ->
+      DB.execute_ db "UPDATE ratchets SET rc_verify_code_ad = NULL, rc_verify_code_pq = NULL"
+    codes1' <- getConnectionVerifyCodes a bId
+    liftIO $ codes1' `shouldBe` codes1
+    liftIO $ withTransaction (store $ agentEnv a) $ \db ->
+      DB.execute_ db "UPDATE ratchets SET ratchet_state = NULL"
+    codes1'' <- getConnectionVerifyCodes a bId
+    liftIO $ codes1'' `shouldBe` codes1
 
 testDeliveryReceipts :: HasCallStack => IO ()
 testDeliveryReceipts =
@@ -4618,8 +4937,7 @@ testServerMultipleIdentities =
         testE2ERatchetParams12
 
 testWaitForUserNetwork :: IO ()
-testWaitForUserNetwork = do
-  a <- getSMPAgentClient' 1 aCfg initAgentServers testDB
+testWaitForUserNetwork = withAgent 1 aCfg initAgentServers testDB $ \a -> do
   noNetworkDelay a
   setUserNetworkInfo a $ UserNetworkInfo UNNone False
   networkDelay a 100000
@@ -4636,8 +4954,7 @@ testWaitForUserNetwork = do
     aCfg = agentCfg {userNetworkInterval = 100000, userOfflineDelay = 0}
 
 testDoNotResetOnlineToOffline :: IO ()
-testDoNotResetOnlineToOffline = do
-  a <- getSMPAgentClient' 1 aCfg initAgentServers testDB
+testDoNotResetOnlineToOffline = withAgent 1 aCfg initAgentServers testDB $ \a -> do
   noNetworkDelay a
   setUserNetworkInfo a $ UserNetworkInfo UNWifi False
   networkDelay a 100000
@@ -4658,8 +4975,7 @@ testDoNotResetOnlineToOffline = do
     aCfg = agentCfg {userNetworkInterval = 100000, userOfflineDelay = 0.1}
 
 testResumeMultipleThreads :: IO ()
-testResumeMultipleThreads = do
-  a <- getSMPAgentClient' 1 aCfg initAgentServers testDB
+testResumeMultipleThreads = withAgent 1 aCfg initAgentServers testDB $ \a -> do
   noNetworkDelay a
   setUserNetworkInfo a $ UserNetworkInfo UNNone False
   vs <-
@@ -4705,7 +5021,7 @@ testServerQueueInfo = do
     Just srvMsgId <- checkMsgQ bob aliceId 1
     get bob =##> \case
       ("", c, MSG MsgMeta {integrity = MsgOk, broker = (smId, _), recipient = (mId, _), pqEncryption = PQEncOn} _ "hello") ->
-        c == aliceId && decodeLatin1 (B64.encode smId) == srvMsgId && mId == msgId
+        c == aliceId && decodeLatin1 (B64.encode $ SMP.unMsgId smId) == srvMsgId && mId == msgId
       _ -> False
     ackMessage bob aliceId msgId Nothing
     liftIO $ threadDelay 200000
@@ -4736,9 +5052,9 @@ testServerQueueInfo = do
     get bob =##> \case ("", c, Msg' mId PQEncOn "hello 4") -> c == aliceId && mId == msgId4; _ -> False
     ackMessage bob aliceId msgId4 Nothing
     liftIO $ threadDelay 200000
-    Just _ <- checkMsgQ bob aliceId 1 -- the one that did not fit now accepted
     get alice ##> ("", bobId, QCONT)
     get alice ##> ("", bobId, SENT msgId5)
+    Just _ <- checkMsgQ bob aliceId 1 -- the one that did not fit now accepted
     liftIO $ threadDelay 200000
     Just _srvMsgId <- checkQ bob aliceId True (Just QNoSub) 1 (Just MTMessage)
     get bob =##> \case ("", c, Msg' mId PQEncOn "hello: quota exceeded") -> c == aliceId && mId == msgId5 + 1; _ -> False

tests/AgentTests/NotificationTests.hs

@@ -230,7 +230,7 @@ testNotificationToken apns = do
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
       getMockNotification apns tkn
     verification <- ntfData .-> "verification"
-    nonce <- C.cbNonce <$> ntfData .-> "nonce"
+    nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
     verifyNtfToken a tkn nonce verification
     NTActive <- checkNtfToken a tkn
     deleteNtfToken a tkn
@@ -254,7 +254,7 @@ testNtfTokenRepeatRegistration apns = do
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
       getMockNotification apns tkn
     verification <- ntfData .-> "verification"
-    nonce <- C.cbNonce <$> ntfData .-> "nonce"
+    nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
     NTRegistered <- registerNtfToken a tkn NMPeriodic
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData'}} <-
       getMockNotification apns tkn
@@ -273,14 +273,14 @@ testNtfTokenSecondRegistration apns =
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
       getMockNotification apns tkn
     verification <- ntfData .-> "verification"
-    nonce <- C.cbNonce <$> ntfData .-> "nonce"
+    nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
     verifyNtfToken a tkn nonce verification
 
     NTRegistered <- registerNtfToken a' tkn NMPeriodic
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData'}} <-
       getMockNotification apns tkn
     verification' <- ntfData' .-> "verification"
-    nonce' <- C.cbNonce <$> ntfData' .-> "nonce"
+    nonce' <- C.unsafeCbNonce <$> ntfData' .-> "nonce"
 
     -- at this point the first token is still active
     NTActive <- checkNtfToken a tkn
@@ -311,7 +311,7 @@ testNtfTokenServerRestart t apns = do
     -- so that repeat verification happens without restarting the clients, when notification arrives
     withNtfServer t $ runRight_ $ do
       verification <- ntfData .-> "verification"
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       verifyNtfToken a' tkn nonce verification
       NTActive <- checkNtfToken a' tkn
       pure ()
@@ -327,7 +327,7 @@ testNtfTokenServerRestartReverify t apns = do
       pure ntfData
     runRight_ $ do
       verification <- ntfData .-> "verification"
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       Left (BROKER _ (NETWORK _)) <- tryE $ verifyNtfToken a tkn nonce verification
       pure ()
   threadDelay 1500000
@@ -348,7 +348,7 @@ testNtfTokenServerRestartReverifyTimeout t apns = do
       APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
         getMockNotification apns tkn
       verification <- ntfData .-> "verification"
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       verifyNtfToken a tkn nonce verification
       pure (nonce, verification)
     -- this emulates the situation when server verified token but the client did not receive the response
@@ -393,7 +393,7 @@ testNtfTokenServerRestartReregister t apns = do
       APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
         getMockNotification apns tkn
       verification <- ntfData .-> "verification"
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       verifyNtfToken a' tkn nonce verification
       NTActive <- checkNtfToken a' tkn
       pure ()
@@ -428,7 +428,7 @@ testNtfTokenServerRestartReregisterTimeout t apns = do
       APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
         getMockNotification apns tkn
       verification <- ntfData .-> "verification"
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       verifyNtfToken a' tkn nonce verification
       NTActive <- checkNtfToken a' tkn
       pure ()
@@ -450,7 +450,7 @@ testNtfTokenMultipleServers t apns = do
         APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
           getMockNotification apns tkn
         verification <- ntfData .-> "verification"
-        nonce <- C.cbNonce <$> ntfData .-> "nonce"
+        nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
         verifyNtfToken a tkn nonce verification
         NTActive <- checkNtfToken a tkn
         -- shut down the "other" server
@@ -566,7 +566,7 @@ testNotificationSubscriptionExistingConnection apns baseId alice@AgentClient {ag
     APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData}} <-
       getMockNotification apns tkn
     verification <- ntfData .-> "verification"
-    vNonce <- C.cbNonce <$> ntfData .-> "nonce"
+    vNonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
     verifyNtfToken alice tkn vNonce verification
     NTActive <- checkNtfToken alice tkn
     -- send message
@@ -657,7 +657,7 @@ registerTestToken a token mode apns = do
   Just APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData'}} <-
     timeout 1000000 $ getMockNotification apns tkn
   verification' <- ntfData' .-> "verification"
-  nonce' <- C.cbNonce <$> ntfData' .-> "nonce"
+  nonce' <- C.unsafeCbNonce <$> ntfData' .-> "nonce"
   verifyNtfToken a tkn nonce' verification'
   NTActive <- checkNtfToken a tkn
   pure tkn
@@ -1035,7 +1035,7 @@ messageNotification apns tkn = do
   500000 `timeout` getMockNotification apns tkn >>= \case
     Nothing -> error "no notification"
     Just APNSMockRequest {notification = APNSNotification {aps = APNSMutableContent {}, notificationData = Just ntfData}} -> do
-      nonce <- C.cbNonce <$> ntfData .-> "nonce"
+      nonce <- C.unsafeCbNonce <$> ntfData .-> "nonce"
       message <- ntfData .-> "message"
       pure (nonce, message)
     _ -> error "bad notification"

tests/AgentTests/SQLiteTests.hs

@@ -20,12 +20,13 @@ import Control.Concurrent.Async (concurrently_)
 import Control.Concurrent.MVar
 import Control.Concurrent.STM
 import Control.Exception (SomeException)
-import Control.Monad (replicateM_)
+import Control.Monad (forM_, replicateM_)
 import Control.Monad.Trans.Except
 import Crypto.Random (ChaChaDRG)
 import Data.ByteArray (ScrubbedBytes)
 import Data.ByteString.Char8 (ByteString)
 import Data.List (isInfixOf)
+import qualified Data.Map.Strict as M
 import qualified Data.Text as T
 import Data.Text.Encoding (encodeUtf8)
 import Data.Time
@@ -51,6 +52,7 @@ import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Crypto.File (CryptoFile (..))
 import Simplex.Messaging.Crypto.Ratchet (pattern IKPQOn)
 import qualified Simplex.Messaging.Crypto.Ratchet as CR
+import Simplex.Messaging.Encoding (Encoding (..))
 import Simplex.Messaging.Encoding.String (StrEncoding (..))
 import Simplex.Messaging.Protocol (EntityId (..), QueueMode (..), SubscriptionMode (..), pattern VersionSMPC)
 import qualified Simplex.Messaging.Protocol as SMP
@@ -135,6 +137,8 @@ storeTests = do
           testCreateRcvMsg
           testCreateSndMsg
           testCreateRcvAndSndMsgs
+      describe "deleteRatchetKeyHashesExpired" testDeleteRatchetKeyHashesExpired
+      it "should keep only the newest skipped message keys" testGetSkippedMsgKeys
       describe "Work items" $ do
         it "should getPendingQueueMsg" testGetPendingQueueMsg
         it "should getPendingServerCommand" testGetPendingServerCommand
@@ -509,7 +513,7 @@ hw = encodeUtf8 "Hello world!"
 ts :: UTCTime
 ts = UTCTime (fromGregorian 2021 02 24) (secondsToDiffTime 0)
 
-mkRcvMsgData :: InternalId -> InternalRcvId -> ExternalSndId -> BrokerId -> MsgHash -> RcvMsgData
+mkRcvMsgData :: InternalId -> InternalRcvId -> ExternalSndId -> ByteString -> MsgHash -> RcvMsgData
 mkRcvMsgData internalId internalRcvId externalSndId brokerId internalHash =
   RcvMsgData
     { internalRcvId,
@@ -518,7 +522,7 @@ mkRcvMsgData internalId internalRcvId externalSndId brokerId internalHash =
           { integrity = MsgOk,
             recipient = (unId internalId, ts),
             sndMsgId = externalSndId,
-            broker = (brokerId, ts),
+            broker = (SMP.unsafeMsgId $ mconcat (replicate 24 brokerId), ts),
             pqEncryption = CR.PQEncOn
           },
       msgType = AM_A_MSG_,
@@ -599,6 +603,42 @@ testCreateRcvAndSndMsgs =
       testCreateSndMsg_ db "snd_hash_1" connId sq $ mkSndMsgData (InternalId 5) (InternalSndId 2) "snd_hash_2"
       testCreateSndMsg_ db "snd_hash_2" connId sq $ mkSndMsgData (InternalId 6) (InternalSndId 3) "snd_hash_3"
 
+testDeleteRatchetKeyHashesExpired :: SpecWith DBStore
+testDeleteRatchetKeyHashesExpired =
+  it "should delete expired ratchet key hashes except the newest in each connection" . withStoreTransaction $ \db -> do
+    g <- C.newRandom
+    Right connId <- createNewConn db g cData1 {connId = ""} SCMInvitation
+    Right connId' <- createNewConn db g cData1 {connId = ""} SCMContact
+    let hashes = ["h1", "h2", "h3", "h4", "h5", "h6"]
+    forM_ hashes $ addProcessedRatchetKeyHash db connId
+    forM_ (take 4 hashes) $ addProcessedRatchetKeyHash db connId'
+    deleteRatchetKeyHashesExpired db 86400 4
+    mapM (checkRatchetKeyHashExists db connId) hashes `shouldReturn` replicate 6 True
+    deleteRatchetKeyHashesExpired db 0 4
+    mapM (checkRatchetKeyHashExists db connId) hashes `shouldReturn` [False, False, True, True, True, True]
+    mapM (checkRatchetKeyHashExists db connId') (take 4 hashes) `shouldReturn` replicate 4 True
+
+testGetSkippedMsgKeys :: DBStore -> Expectation
+testGetSkippedMsgKeys st = do
+  g <- C.newRandom
+  withTransaction st $ \db -> do
+    Right connId <- createNewConn db g cData1 {connId = ""} SCMInvitation
+    Right connId' <- createNewConn db g cData1 {connId = ""} SCMInvitation
+    createSkippedKeys db connId'
+    createSkippedKeys db connId
+    M.map M.keys <$> getSkippedMsgKeys db connId 4
+      `shouldReturn` M.singleton (C.Key "header_key") [1, 2, 3, 4]
+    getMsgNs db connId `shouldReturn` [1 .. 4]
+    getMsgNs db connId' `shouldReturn` [1 .. 10]
+  where
+    createSkippedKeys :: DB.Connection -> ConnId -> IO ()
+    createSkippedKeys db connId = do
+      DB.execute db "INSERT INTO ratchets (conn_id) VALUES (?)" (Only connId)
+      forM_ ([10, 9 .. 1] :: [Int]) $ \msgN ->
+        DB.execute db "INSERT INTO skipped_messages (conn_id, header_key, msg_n, msg_key) VALUES (?, ?, ?, ?)" (connId, "header_key" :: ByteString, msgN, smpEncode ("key" :: ByteString, "iv" :: ByteString))
+    getMsgNs :: DB.Connection -> ConnId -> IO [Int]
+    getMsgNs db connId = map fromOnly <$> DB.query db "SELECT msg_n FROM skipped_messages WHERE conn_id = ? ORDER BY msg_n" (Only connId)
+
 testCloseReopenStore :: IO ()
 testCloseReopenStore = do
   st <- createStore'
@@ -648,7 +688,7 @@ hasMigrations :: DBStore -> Expectation
 hasMigrations st = getMigrations st `shouldReturn` True
 
 errorGettingMigrations :: DBStore -> Expectation
-errorGettingMigrations st = getMigrations st `shouldThrow` \(e :: SomeException) -> "ErrorMisuse" `isInfixOf` show e
+errorGettingMigrations st = getMigrations st `shouldThrow` \(e :: SomeException) -> "withTransaction: database closed" `isInfixOf` show e
 
 testGetPendingQueueMsg :: DBStore -> Expectation
 testGetPendingQueueMsg st = do
@@ -743,9 +783,11 @@ testGetNextRcvChunkToDownload st = do
   withTransaction st $ \db -> do
     Right Nothing <- getNextRcvChunkToDownload db xftpServer1 86400
 
-    Right _ <- createRcvFile db g 1 rcvFileDescr1 "filepath" "filepath" (CryptoFile "filepath" Nothing) True
+    Right fId1 <- createRcvFile db g 1 rcvFileDescr1 "filepath" "filepath" (CryptoFile "filepath" Nothing) True
+    Right _ <- startPreparedRcvFile db fId1
     DB.execute_ db "UPDATE rcv_file_chunk_replicas SET replica_key = cast('bad' as blob) WHERE rcv_file_chunk_replica_id = 1"
     Right fId2 <- createRcvFile db g 1 rcvFileDescr1 "filepath" "filepath" (CryptoFile "filepath" Nothing) True
+    Right _ <- startPreparedRcvFile db fId2
 
     Left e <- getNextRcvChunkToDownload db xftpServer1 86400
     show e `shouldContain` "ConversionFailed"
@@ -783,6 +825,7 @@ testGetNextSndFileToPrepare st = do
     -- Right _ <- createSndFile db g 1 (CryptoFile "filepath" Nothing) 1 "filepath" testFileSbKey testFileCbNonce Nothing
     -- DB.execute_ db "UPDATE snd_files SET status = 'new', num_recipients = 'bad' WHERE snd_file_id = 1"
     Right fId2 <- createSndFile db g 1 (CryptoFile "filepath" Nothing) 1 "filepath" testFileSbKey testFileCbNonce Nothing Nothing
+    Right _ <- startPreparedSndFile db fId2
     DB.execute_ db "UPDATE snd_files SET status = 'new' WHERE snd_file_id = 2"
 
     -- Left e <- getNextSndFileToPrepare db 86400

tests/AgentTests/SchemaDump.hs

@@ -48,6 +48,8 @@ schemaDumpTest = do
   it "verify strict tables" testVerifyStrict
   it "should NOT create user record for new database" testUsersMigrationNew
   it "should create user record for old database" testUsersMigrationOld
+  it "should remove duplicate ratchet key hashes before adding unique index" testRatchetKeyHashesUniqueMigration
+  it "should clear received_qend switch status on downgrade" testRcvSwitchStatusDowngrade
 
 testVerifySchemaDump :: IO ()
 testVerifySchemaDump = do
@@ -114,12 +116,45 @@ testUsersMigrationOld = do
     `shouldReturn` ([Only (1 :: Int)])
   closeDBStore st'
 
+testRatchetKeyHashesUniqueMigration :: IO ()
+testRatchetKeyHashesUniqueMigration = do
+  let beforeUnique = takeWhile (("m20260929_ratchet_indexes" /=) . name) appMigrations
+  Right st <- createDBStore (DBOpts testDB [] "" False True TQOff) beforeUnique (MigrationConfig MCError Nothing)
+  withTransaction' st $ \db -> do
+    SQL.execute_ db "INSERT INTO users (user_id) VALUES (1)"
+    SQL.execute_ db "INSERT INTO connections (conn_id, conn_mode, user_id) VALUES (x'01', 'INV', 1), (x'02', 'INV', 1)"
+    SQL.execute_ db "INSERT INTO processed_ratchet_key_hashes (conn_id, hash) VALUES (x'01', x'aa'), (x'01', x'aa'), (x'01', x'bb'), (x'02', x'aa'), (x'01', x'aa')"
+  closeDBStore st
+  Right st' <- createDBStore (DBOpts testDB [] "" False True TQOff) appMigrations (MigrationConfig MCYesUp Nothing)
+  withTransaction' st' (`SQL.query_` "SELECT processed_ratchet_key_hash_id FROM processed_ratchet_key_hashes ORDER BY processed_ratchet_key_hash_id")
+    `shouldReturn` [Only (1 :: Int), Only 3, Only 4]
+  closeDBStore st'
+
+testRcvSwitchStatusDowngrade :: IO ()
+testRcvSwitchStatusDowngrade = do
+  Right st <- createDBStore (DBOpts testDB [] "" False True TQOff) appMigrations (MigrationConfig MCError Nothing)
+  withTransaction' st $ \db -> do
+    SQL.execute_ db "INSERT INTO users (user_id) VALUES (1)"
+    SQL.execute_ db "INSERT INTO connections (conn_id, conn_mode, user_id) VALUES (x'01', 'INV', 1)"
+    SQL.execute_ db "INSERT INTO servers (host, port, key_hash) VALUES ('localhost', '5001', x'00')"
+    SQL.execute_ db "INSERT INTO rcv_queues (host, port, rcv_id, conn_id, rcv_private_key, rcv_dh_secret, e2e_priv_key, snd_id, status, rcv_queue_id, rcv_primary, switch_status) VALUES ('localhost', '5001', x'01', x'01', x'00', x'00', x'00', x'01', 'active', 1, 0, 'received_qend'), ('localhost', '5001', x'02', x'01', x'00', x'00', x'00', x'02', 'active', 2, 1, 'sending_qadd')"
+  closeDBStore st
+  let beforeRcvSwitchStatus = takeWhile (("m20261008_rcv_switch_status" /=) . name) appMigrations
+  Right st' <- createDBStore (DBOpts testDB [] "" False True TQOff) beforeRcvSwitchStatus (MigrationConfig MCYesUpDown Nothing)
+  withTransaction' st' (`SQL.query_` "SELECT switch_status FROM rcv_queues ORDER BY rcv_queue_id")
+    `shouldReturn` [Only (Nothing :: Maybe Text), Only (Just "sending_qadd")]
+  closeDBStore st'
+
 skipComparisonForDownMigrations :: [String]
 skipComparisonForDownMigrations =
   [ -- on down migration idx_messages_internal_snd_id_ts index moves down to the end of the file
     "m20230814_indexes",
     -- snd_secure and last_broker_ts columns swap order on down migration
-    "m20250322_short_links"
+    "m20250322_short_links",
+    -- on down migration idx_processed_ratchet_key_hashes_hash index moves down to the end of the file
+    "m20260929_ratchet_indexes",
+    -- sqlite_sequence moves after commands table after VACUUM
+    "m20261005_commands_autoincrement"
   ]
 
 getSchema :: FilePath -> FilePath -> IO String

tests/AgentTests/ServerChoice.hs

@@ -71,9 +71,7 @@ initServers =
     }
 
 testChooseDifferentOperator :: IO ()
-testChooseDifferentOperator = do
-  c <- getSMPAgentClient' 1 agentCfg initServers testDB
-  runRight_ $ do
+testChooseDifferentOperator = withAgent 1 agentCfg initServers testDB $ \c -> runRight_ $ do
     -- chooses the only operator with storage role
     srv1 <- withAgentEnv c $ getNextServer c 1 storageSrvs []
     liftIO $ srv1 == testOp1Srv1 || srv1 == testOp1Srv2 `shouldBe` True

tests/CoreTests/BatchingTests.hs

@@ -14,11 +14,13 @@ import Control.Monad
 import Crypto.Random (ChaChaDRG)
 import qualified Data.ByteString as B
 import Data.ByteString.Char8 (ByteString)
+import Data.ByteString.Unsafe (unsafeUseAsCString)
 import qualified Data.List.NonEmpty as L
 import Data.Time.Clock.System (SystemTime, getSystemTime)
 import qualified Data.X509 as X
 import qualified Data.X509.CertificateStore as XS
 import qualified Data.X509.File as XF
+import Foreign.Ptr (plusPtr)
 import Simplex.Messaging.Client
 import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Encoding
@@ -40,6 +42,8 @@ batchingTests = do
     it "should batch subscription responses with message" testBatchSubResponses
     it "should break on message that does not fit" testClientBatchWithMessage
     it "should break on large message" testClientBatchWithLargeMessage
+  describe "tDecodeServer" $
+    it "should copy IDs out of received block" testDecodeServerCopiesIds
 
 testBatchSubscriptions :: IO ()
 testBatchSubscriptions = do
@@ -166,6 +170,26 @@ testClientBatchWithLargeMessage = do
   (length rs1', length rs2') `shouldBe` (75, 135)
   all lenOk [s1', s2'] `shouldBe` True
 
+testDecodeServerCopiesIds :: IO ()
+testDecodeServerCopiesIds = do
+  sessId <- atomically . C.randomBytes 32 =<< C.newRandom
+  subs <- replicateM 2 $ randomSUB sessId
+  let thParams = testTHandleParams sessId
+  [TBTransmissions s 2 _] <- pure $ batchTransmissions thParams $ L.fromList subs
+  forM_ (tParse thParams s) $ \t -> do
+    let (corrId, entId, _) = tDecodeClient @SMPVersion @ErrorType @Cmd thParams t
+    Right (_, _, (corrId', entId', _)) <- pure $ tDecodeServer @SMPVersion @ErrorType @Cmd (testTHandleParams sessId) t
+    (corrId', entId') `shouldBe` (corrId, entId)
+    sharesBuffer s (bs corrId) `shouldReturn` True
+    sharesBuffer s (unEntityId entId) `shouldReturn` True
+    sharesBuffer s (bs corrId') `shouldReturn` False
+    sharesBuffer s (unEntityId entId') `shouldReturn` False
+
+sharesBuffer :: ByteString -> ByteString -> IO Bool
+sharesBuffer block s =
+  unsafeUseAsCString block $ \blockPtr -> unsafeUseAsCString s $ \ptr ->
+    pure $ ptr >= blockPtr && ptr < blockPtr `plusPtr` B.length block
+
 testClientStub :: IO (ProtocolClient SMPVersion ErrorType BrokerMsg)
 testClientStub = do
   g <- C.newRandom
@@ -183,7 +207,7 @@ randomSUB sessId = do
   thAuth_ <- testTHandleAuth g rKey
   let thParams = testTHandleParams sessId
       TransmissionForAuth {tForAuth, tToSend} = encodeTransmissionForAuth thParams (CorrId corrId, EntityId rId, Cmd SRecipient SUB)
-  pure $ (,tToSend) <$> authTransmission thAuth_ True (Just rpKey) nonce tForAuth
+  pure $ (,tToSend) <$> authTransmission thAuth_ True (Just rpKey) (Just nonce) tForAuth
 
 randomSUBCmd :: ProtocolClient SMPVersion ErrorType BrokerMsg -> IO (PCTransmission ErrorType BrokerMsg)
 randomSUBCmd c = do
@@ -202,7 +226,7 @@ randomNMSGCmd :: SystemTime -> IO (Transmission BrokerMsg)
 randomNMSGCmd ts = do
   g <- C.newRandom
   nId <- atomically $ C.randomBytes 24 g
-  msgId <- atomically $ C.randomBytes 24 g
+  msgId <- atomically $ randomMsgId 24 g
   (k, pk) <- atomically $ C.generateKeyPair g
   nonce <- atomically $ C.randomCbNonce g
   let msgMeta = NMsgMeta {msgId, msgTs = ts}
@@ -221,7 +245,7 @@ randomMSG = do
   g <- C.newRandom
   corrId <- atomically $ C.randomBytes 24 g
   rId <- atomically $ C.randomBytes 24 g
-  msgId <- atomically $ C.randomBytes 24 g
+  msgId <- atomically $ randomMsgId 24 g
   msg <- atomically $ C.randomBytes maxMessageLength g
   pure (CorrId corrId, EntityId rId, MSG RcvMessage {msgId, msgBody = EncRcvMsgBody msg})
 
@@ -235,9 +259,9 @@ randomSEND sessId len = do
   msg <- atomically $ C.randomBytes len g
   let thParams = testTHandleParams sessId
       TransmissionForAuth {tForAuth, tToSend} = encodeTransmissionForAuth thParams (CorrId corrId, EntityId sId, Cmd SSender $ SEND noMsgFlags msg)
-  pure $ (,tToSend) <$> authTransmission thAuth_ False (Just spKey) nonce tForAuth
+  pure $ (,tToSend) <$> authTransmission thAuth_ False (Just spKey) (Just nonce) tForAuth
 
-testTHandleParams :: ByteString -> THandleParams SMPVersion 'TClient
+testTHandleParams :: ByteString -> THandleParams SMPVersion p
 testTHandleParams sessionId =
   THandleParams
     { sessionId,

tests/CoreTests/CryptoTests.hs

@@ -6,8 +6,9 @@
 
 module CoreTests.CryptoTests (cryptoTests) where
 
+import Control.Concurrent (forkIO, newEmptyMVar, putMVar, takeMVar)
 import Control.Concurrent.STM
-import Control.Exception (evaluate)
+import Control.Exception (bracket, evaluate)
 import Control.Monad.Except
 import qualified Data.Aeson as J
 import qualified Data.ByteString.Char8 as B
@@ -22,9 +23,12 @@ import Data.Time.Clock (UTCTime (..))
 import qualified Data.Text.Lazy as LT
 import qualified Data.Text.Lazy.Encoding as LE
 import Data.Type.Equality
+import Data.Word (Word8)
 import qualified Data.X509 as X
 import qualified Data.X509.CertificateStore as XS
 import qualified Data.X509.Validation as XV
+import Foreign (FunPtr, allocaBytes, fillBytes, freeHaskellFunPtr, nullPtr)
+import Foreign.C.Types (CInt, CSize (..))
 import qualified SMPClient
 import qualified Simplex.Messaging.Crypto as C
 import qualified Simplex.Messaging.Crypto.Lazy as LC
@@ -32,9 +36,12 @@ import Simplex.Messaging.Crypto.BBS
 import Simplex.Messaging.Crypto.Entitlement
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings
 import Simplex.Messaging.Crypto.SNTRUP761.Bindings.Defines
+import Simplex.Messaging.Crypto.SNTRUP761.Bindings.FFI (c_sntrup761_keypair)
+import Simplex.Messaging.Crypto.SNTRUP761.Bindings.RNG (RNGFunc)
 import Simplex.Messaging.Encoding (Large (..), smpDecode, smpEncode)
 import Simplex.Messaging.Encoding.String (strDecode, strEncode)
 import Simplex.Messaging.Transport.Client
+import System.Timeout (timeout)
 import Test.Hspec hiding (fit, it)
 import Test.Hspec.QuickCheck (modifyMaxSuccess)
 import Test.QuickCheck hiding (Large)
@@ -113,6 +120,7 @@ cryptoTests = do
   describe "sntrup761" $ do
     it "should enc/dec key" testSNTRUP761
     it "should reject malformed KEM encodings" testSNTRUP761RejectsMalformedEncodings
+    it "should fail key generation with degenerate RNG" testSNTRUP761KeypairDegenerateRNG
   describe "BBS+" $ do
     it "should sign and verify" testBBSSignVerify
     it "should derive public key from secret key" testBBSPublicKeyDerivation
@@ -298,6 +306,26 @@ testSNTRUP761 = do
   KEMSharedKey k' <- sntrup761Dec c sk
   k' `shouldBe` k
 
+foreign import ccall "wrapper"
+  mkRNGFunc :: RNGFunc -> IO (FunPtr RNGFunc)
+
+testSNTRUP761KeypairDegenerateRNG :: IO ()
+testSNTRUP761KeypairDegenerateRNG = do
+  -- constant byte 0 draws invertible g = -(1 + x + ... + x^760), byte 0x20 draws g = 0
+  keypairWithConstantRNG 0 `shouldReturn` Just 0
+  keypairWithConstantRNG 0x20 `shouldReturn` Just (-1)
+  where
+    keypairWithConstantRNG :: Word8 -> IO (Maybe CInt)
+    keypairWithConstantRNG b = do
+      result <- newEmptyMVar
+      -- timeout cannot interrupt a foreign call, so the call runs in another thread
+      _ <- forkIO $
+        bracket (mkRNGFunc $ \_ sz buf -> fillBytes buf b (fromIntegral sz)) freeHaskellFunPtr $ \rng ->
+          allocaBytes c_SNTRUP761_PUBLICKEY_SIZE $ \pkPtr ->
+            allocaBytes c_SNTRUP761_SECRETKEY_SIZE $ \skPtr ->
+              c_sntrup761_keypair pkPtr skPtr nullPtr rng >>= putMVar result
+      timeout 10000000 $ takeMVar result
+
 testSNTRUP761RejectsMalformedEncodings :: IO ()
 testSNTRUP761RejectsMalformedEncodings = do
   smpDecode @KEMPublicKey (smpEncode $ Large shortPublicKey) `shouldSatisfy` isLeft

tests/CoreTests/MsgStoreTests.hs

@@ -34,7 +34,7 @@ import Data.Time.Clock.System (SystemTime (..), getSystemTime)
 import SMPClient (testStoreLogFile, testStoreMsgsDir, testStoreMsgsDir2, testStoreMsgsFile, testStoreMsgsFile2)
 import Simplex.Messaging.Crypto (pattern MaxLenBS)
 import qualified Simplex.Messaging.Crypto as C
-import Simplex.Messaging.Protocol (EntityId (..), ErrorType, LinkId, Message (..), QueueLinkData, RecipientId, SParty (..), noMsgFlags)
+import Simplex.Messaging.Protocol (EncDataBytes (..), EntityId (..), ErrorType (..), LinkId, Message (..), QueueLinkData, RecipientId, SParty (..), noMsgFlags, randomMsgId)
 import Simplex.Messaging.Server (exportMessages, importMessages, printMessageStats)
 import Simplex.Messaging.Server.Env.STM (MsgStore (..), journalMsgStoreDepth, readWriteQueueStore)
 import Simplex.Messaging.Server.Expiration (ExpirationConfig (..), expireBeforeEpoch)
@@ -43,6 +43,7 @@ import Simplex.Messaging.Server.MsgStore.STM
 import Simplex.Messaging.Server.MsgStore.Types
 import Simplex.Messaging.Server.QueueStore
 import Simplex.Messaging.Server.QueueStore.QueueInfo
+import Simplex.Messaging.Server.QueueStore.Types
 import Simplex.Messaging.Server.StoreLog (closeStoreLog, logCreateQueue)
 import System.Directory (copyFile, createDirectoryIfMissing, listDirectory, removeFile, renameFile)
 import System.FilePath ((</>))
@@ -57,7 +58,6 @@ import Simplex.Messaging.Agent.Store.Postgres.Common
 import Simplex.Messaging.Agent.Store.Shared (MigrationConfirmation (..))
 import Simplex.Messaging.Server.MsgStore.Postgres
 import Simplex.Messaging.Server.QueueStore.Postgres
-import Simplex.Messaging.Server.QueueStore.Types
 import SMPClient (postgressBracket, testServerDBConnectInfo, testStoreDBOpts)
 #endif
 
@@ -79,6 +79,7 @@ msgStoreTests = do
         someMsgStoreTests
         it "should correctly update message counts and canWrite flag" testUpdateMessageCounts
         it "tryDelPeekMsg (ACK not from NSE) should reset message counts when queue is empty" testResetMessageCounts
+        it "should expire messages across commit batches" testExpireMessagesInBatches
 #endif
   describe "Journal message store: queue state backup expiration" $ do
     it "should remove old queue state backups" testRemoveQueueStateBackups
@@ -100,6 +101,7 @@ msgStoreTests = do
       it "should get queue and store/read messages" testGetQueue
       it "should write/ack messages" testWriteAckMessages
       it "should not fail on EOF when changing read journal" testChangeReadJournal
+      it "should not add link data to secured messaging queue" testLinkDataSecuredQueue
 
 -- TODO constrain to STM stores?
 withMsgStore :: MsgStoreClass s => MsgStoreConfig s -> (s -> IO ()) -> IO ()
@@ -145,7 +147,7 @@ mkMessage :: MonadIO m => ByteString -> m Message
 mkMessage body = liftIO $ do
   g <- C.newRandom
   msgTs <- getSystemTime
-  msgId <- atomically $ C.randomBytes 24 g
+  msgId <- atomically $ randomMsgId 24 g
   pure Message {msgId, msgTs, msgFlags = noMsgFlags, msgBody = C.unsafeMaxLenBS body}
 
 pattern Msg :: ByteString -> Maybe Message
@@ -266,6 +268,33 @@ testChangeReadJournal ms = do
     (Msg "message 5", Nothing) <- tryDelPeekMsg ms q mId5
     void $ ExceptT $ deleteQueue ms q
 
+testLinkDataSecuredQueue :: MsgStoreClass s => s -> IO ()
+testLinkDataSecuredQueue ms = do
+  g <- C.newRandom
+  (sKey, _) <- atomically $ C.generateAuthKeyPair C.SEd25519 g
+  let st = queueStore ms
+      ld = (EncDataBytes "fixed data", EncDataBytes "user data")
+      rndId = atomically $ EntityId <$> C.randomBytes 24 g
+  (rId, qr) <- testNewQueueRec g QMMessaging
+  (cId, cqr) <- testNewQueueRec g QMContact
+  lnkId <- rndId
+  cLnkId <- rndId
+  runRight_ $ do
+    q <- ExceptT $ addQueue ms rId qr
+    -- the handle is read before SKEY, as in a command that raced with it
+    staleQ <- ExceptT $ getQueue ms SRecipient rId
+    ExceptT $ secureQueue st q sKey
+    liftIO $ addQueueLinkData st staleQ lnkId ld `shouldReturn` Left AUTH
+    freshQ <- ExceptT $ getQueue ms SRecipient rId
+    liftIO $ getQueueLinkData st freshQ lnkId `shouldReturn` Left AUTH
+    cq <- ExceptT $ addQueue ms cId cqr
+    ExceptT $ secureQueue st cq sKey
+    ExceptT $ addQueueLinkData st cq cLnkId ld
+    ld' <- ExceptT $ getQueueLinkData st cq cLnkId
+    liftIO $ ld' `shouldBe` ld
+    void $ ExceptT $ deleteQueue ms q
+    void $ ExceptT $ deleteQueue ms cq
+
 testExportImportStore :: JournalMsgStore 'QSMemory -> IO ()
 testExportImportStore ms = do
   g <- C.newRandom
@@ -327,35 +356,34 @@ testUpdateMessageCounts ms = do
     q <- ExceptT $ addQueue ms rId qr
     let write s = writeMsg ms q True =<< mkMessage s
         hasSize = checkQueueSize ms
-    q `hasSize` (0, True, False)
+    q `hasSize` (0, True)
     Just (Message {msgId = mId1}, True) <- write "message 1"
-    q `hasSize` (1, True, True)
+    q `hasSize` (1, True)
     Just (Message {msgId = mId2}, False) <- write "message 2"
-    q `hasSize` (2, True, True)
+    q `hasSize` (2, True)
     Just (Message {msgId = mId3}, False) <- write "message 3"
-    q `hasSize` (3, True, True)
+    q `hasSize` (3, True)
     Nothing <- write "message 4"
-    q `hasSize` (4, False, True)
+    q `hasSize` (4, False)
     Msg "message 1" <- tryPeekMsg ms q
-    q `hasSize` (4, False, True)
+    q `hasSize` (4, False)
     Msg "message 1" <- tryDelMsg ms q mId1
-    q `hasSize` (3, False, True)
+    q `hasSize` (3, False)
     Msg "message 2" <- tryPeekMsg ms q
     (Msg "message 2", Msg "message 3") <- tryDelPeekMsg ms q mId2
-    q `hasSize` (2, False, True)
+    q `hasSize` (2, False)
     (Msg "message 3", Just MessageQuota {msgId = mId4}) <- tryDelPeekMsg ms q mId3
-    q `hasSize` (1, False, True)
+    q `hasSize` (1, False)
     (Just MessageQuota {}, Nothing) <- tryDelPeekMsg ms q mId4
-    q `hasSize` (0, True, False)
+    q `hasSize` (0, True)
 
-checkQueueSize :: PostgresMsgStore -> PostgresQueue -> (Int64, Bool, Bool) -> ExceptT ErrorType IO ()
-checkQueueSize ms q (size, canWrt, expire) = liftIO $ do
-  [(size', canWrt', expire')] <-
+checkQueueSize :: PostgresMsgStore -> PostgresQueue -> (Int64, Bool) -> ExceptT ErrorType IO ()
+checkQueueSize ms q (size, canWrt) = liftIO $ do
+  [(size', canWrt')] <-
     withTransaction (dbStore $ queueStore ms) $ \db ->
-      DB.query db "SELECT msg_queue_size, msg_can_write, msg_queue_expire FROM msg_queues WHERE recipient_id = ?" (Only (recipientId q))
+      DB.query db "SELECT msg_queue_size, msg_can_write FROM msg_queues WHERE recipient_id = ?" (Only (recipientId q))
   size' `shouldBe` size
   canWrt' `shouldBe` canWrt
-  expire' `shouldBe` expire
 
 testResetMessageCounts :: PostgresMsgStore -> IO ()
 testResetMessageCounts ms = do
@@ -369,25 +397,63 @@ testResetMessageCounts ms = do
     Just (Message {msgId = mId2}, False) <- write "message 2"
     Just (Message {msgId = mId3}, False) <- write "message 3"
     Nothing <- write "message 4"
-    q `hasSize` (4, False, True)
+    q `hasSize` (4, False)
     liftIO $ setIncorrectSize q (10, True)
     Nothing <- write "message 5"
-    q `hasSize` (11, False, True)
+    q `hasSize` (11, False)
     (Msg "message 1", Msg "message 2") <- tryDelPeekMsg ms q mId1
-    q `hasSize` (10, False, True)
+    q `hasSize` (10, False)
     (Msg "message 2", Msg "message 3") <- tryDelPeekMsg ms q mId2
-    q `hasSize` (9, False, True)
+    q `hasSize` (9, False)
     (Msg "message 3", Just MessageQuota {msgId = mId4}) <- tryDelPeekMsg ms q mId3
-    q `hasSize` (8, False, True)
+    q `hasSize` (8, False)
     (Just MessageQuota {}, Just MessageQuota {msgId = mId5}) <- tryDelPeekMsg ms q mId4
-    q `hasSize` (7, False, True)
+    q `hasSize` (7, False)
     (Just MessageQuota {}, Nothing) <- tryDelPeekMsg ms q mId5
-    q `hasSize` (0, True, False) -- reset
+    q `hasSize` (0, True) -- reset
   where
     setIncorrectSize :: PostgresQueue -> (Int64, Bool) -> IO ()
     setIncorrectSize q (size, canWrt) =
       void $ withTransaction (dbStore $ queueStore ms) $ \db ->
         DB.execute db "UPDATE msg_queues SET msg_queue_size = ?, msg_can_write = ? WHERE recipient_id = ?" (size, canWrt, recipientId q)
+
+testExpireMessagesInBatches :: PostgresMsgStore -> IO ()
+testExpireMessagesInBatches ms = do
+  g <- C.newRandom
+  emptiedQs <- replicateM emptiedCount $ newQueue g
+  partialQs <- replicateM partialCount $ newQueue g
+  overQuotaQs <- replicateM overQuotaCount $ newQueue g
+  quotaMsgs <- runRight $ do
+    forM_ (emptiedQs <> partialQs) $ \q -> void $ write q "old 1"
+    forM_ emptiedQs $ \q -> void $ write q "old 2"
+    mapM fillPastQuota overQuotaQs
+  -- msg_ts has second granularity, so the recent messages need a new second to be kept
+  threadDelay 1100000
+  boundary <- systemSeconds <$> getSystemTime
+  runRight_ $ forM_ partialQs $ \q -> void $ write q "recent"
+
+  MessageStats {expiredMsgsCount, storedMsgsCount, storedQueues} <- expireOldMessages False ms boundary 0
+  expiredMsgsCount `shouldBe` (emptiedCount * 2 + partialCount + sum quotaMsgs)
+  storedMsgsCount `shouldBe` (partialCount + overQuotaCount) -- recent messages and quota markers
+  storedQueues `shouldBe` (emptiedCount + partialCount + overQuotaCount)
+  runRight_ $ do
+    forM_ emptiedQs $ \q -> checkQueueSize ms q (0, True)
+    forM_ partialQs $ \q -> checkQueueSize ms q (1, True)
+    -- the quota marker is never expired, and the queue stays blocked until it is acked
+    forM_ overQuotaQs $ \q -> checkQueueSize ms q (1, False)
+  where
+    -- expire_old_messages pages through expired messages and commits per page, so these
+    -- counts put a page boundary inside each group of queues.
+    emptiedCount = 120 :: Int
+    partialCount = 40 :: Int
+    overQuotaCount = 10 :: Int
+    newQueue g = do
+      (rId, qr) <- testNewQueueRec g QMMessaging
+      runRight $ ExceptT $ addQueue ms rId qr
+    write q s = writeMsg ms q True =<< mkMessage s
+    fillPastQuota q = go 0
+      where
+        go n = write q "fill" >>= maybe (pure n) (const $ go (n + 1))
 #endif
 
 testQueueState :: JournalMsgStore s -> IO ()

tests/CoreTests/RetryIntervalTests.hs

@@ -14,8 +14,11 @@ import Util
 retryIntervalTests :: Spec
 retryIntervalTests = do
   describe "Retry interval with 2 modes and lock" $ do
-    testRetryIntervalSameMode
-    testRetryIntervalSwitchMode
+    testRetryIntervalSameMode withRetryNewLock2
+    testRetryIntervalSwitchMode withRetryNewLock2
+  describe "Retry interval with 2 modes" $ do
+    testRetryIntervalSameMode withRetryInterval2
+    testRetryIntervalSwitchMode withRetryInterval2
   describe "Foreground retry interval" $ do
     testRetryForeground
     testRetryToBackground
@@ -41,14 +44,18 @@ testFastRI =
       maxInterval = 40000
     }
 
-testRetryIntervalSameMode :: Spec
-testRetryIntervalSameMode =
+type WithRetry2 = RetryInterval2 -> (RI2State -> (RetryIntervalMode -> IO ()) -> IO ()) -> IO ()
+
+withRetryNewLock2 :: WithRetry2
+withRetryNewLock2 ri action = newEmptyTMVarIO >>= \lock -> withRetryLock2 ri lock action
+
+testRetryIntervalSameMode :: WithRetry2 -> Spec
+testRetryIntervalSameMode withRetry =
   it "should increase elapased time and interval when the mode stays the same" $ do
-    lock <- newEmptyTMVarIO
     intervals <- newTVarIO []
     reportedIntervals <- newTVarIO []
     ts <- newTVarIO =<< getCurrentTime
-    withRetryLock2 testRI lock $ \(RI2State slow fast) loop -> do
+    withRetry testRI $ \(RI2State slow fast) loop -> do
       ints <- addInterval intervals ts
       atomically $ modifyTVar' reportedIntervals ((slow, fast) :)
       when (length ints < 9) $ loop RIFast
@@ -65,14 +72,13 @@ testRetryIntervalSameMode =
                        (20000, 40000)
                      ]
 
-testRetryIntervalSwitchMode :: Spec
-testRetryIntervalSwitchMode =
+testRetryIntervalSwitchMode :: WithRetry2 -> Spec
+testRetryIntervalSwitchMode withRetry =
   it "should increase elapased time and interval when the mode switches" $ do
-    lock <- newEmptyTMVarIO
     intervals <- newTVarIO []
     reportedIntervals <- newTVarIO []
     ts <- newTVarIO =<< getCurrentTime
-    withRetryLock2 testRI lock $ \(RI2State slow fast) loop -> do
+    withRetry testRI $ \(RI2State slow fast) loop -> do
       ints <- addInterval intervals ts
       atomically $ modifyTVar' reportedIntervals ((slow, fast) :)
       when (length ints < 11) $ loop $ if length ints <= 5 then RIFast else RISlow

tests/CoreTests/SOCKSSettings.hs

@@ -2,15 +2,18 @@
 {-# LANGUAGE NamedFieldPuns #-}
 {-# LANGUAGE OverloadedLists #-}
 {-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE PatternSynonyms #-}
 {-# LANGUAGE TypeApplications #-}
 {-# OPTIONS_GHC -fno-warn-ambiguous-fields #-}
 
 module CoreTests.SOCKSSettings where
 
 import Network.Socket (SockAddr (..), tupleToHostAddress)
+import Simplex.Messaging.Agent.Client (ipAddressProtected)
 import Simplex.Messaging.Client
+import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Encoding.String
-import Simplex.Messaging.Protocol (ErrorType)
+import Simplex.Messaging.Protocol (ErrorType, pattern SMPServer)
 import Simplex.Messaging.Transport.Client
 import Test.Hspec hiding (fit, it)
 import Util
@@ -19,6 +22,7 @@ socksSettingsTests :: Spec
 socksSettingsTests = do
   describe "hostMode and requiredHostMode settings" testHostMode
   describe "socksMode setting, independent of hostMode setting" testSocksMode
+  describe "ipAddressProtected, consistent with chosen host and socksMode" testIPAddressProtected
   describe "socks proxy address encoding" testSocksProxyEncoding
 
 testPublicHost :: TransportHost
@@ -94,6 +98,19 @@ testSocksMode = do
       let TransportClientConfig {socksProxy} = transportClientConfig cfg NRMInteractive host False Nothing
        in socksProxy
 
+testIPAddressProtected :: Spec
+testIPAddressProtected = do
+  it "should be protected if SOCKS proxy is used for the chosen host" $ do
+    protected SMAlways HMOnionViaSocks [testPublicHost] `shouldBe` True
+    protected SMOnion HMOnionViaSocks [testPublicHost, testOnionHost] `shouldBe` True
+    protected SMOnion HMPublic [testOnionHost] `shouldBe` True
+  it "should not be protected if SOCKS proxy is not used for the chosen host" $ do
+    protected SMOnion HMOnionViaSocks [testPublicHost] `shouldBe` False
+    protected SMOnion HMPublic [testPublicHost, testOnionHost] `shouldBe` False
+  where
+    protected socksMode hostMode hosts =
+      ipAddressProtected defaultNetworkConfig {socksProxy = Just defaultSocksProxyWithAuth, socksMode, hostMode} (SMPServer hosts "" (C.KeyHash ""))
+
 testSocksProxyEncoding :: Spec
 testSocksProxyEncoding = do
   it "should decode SOCKS proxy with isolate-by-auth mode" $ do

tests/CoreTests/StoreLogTests.hs

@@ -22,7 +22,7 @@ import qualified Data.X509.Validation as XV
 import SMPClient
 import Simplex.FileTransfer.Protocol (FileInfo (..))
 import Simplex.FileTransfer.Server.Store (FileRec (..), FileRecipient (..), FileStoreClass (..), RoundedFileTime, STMFileStore (..))
-import Simplex.FileTransfer.Server.StoreLog (FileStoreLogRecord (..), readWriteFileStore)
+import Simplex.FileTransfer.Server.StoreLog (FileStoreLogRecord (..), readWriteFileStore, serverFileName)
 import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Encoding.String
 import Simplex.Messaging.Protocol
@@ -37,6 +37,7 @@ import Simplex.Messaging.Server.StoreLog
 import Simplex.Messaging.SystemTime
 import Simplex.Messaging.Transport (SMPServiceRole (..))
 import Simplex.Messaging.Transport.Credentials (genCredentials)
+import System.FilePath ((</>))
 import Test.Hspec hiding (fit, it)
 import Util
 
@@ -269,6 +270,18 @@ fileStoreLogTests = do
           saved = [AddFile sId file createdAt Nothing EntityActive, BlockFile sId blockedWithNotice],
           compacted = [AddFile sId file createdAt Nothing (EntityBlocked blockedWithNotice)],
           state = M.fromList [(sId, (file, createdAt, Nothing, EntityBlocked blockedWithNotice))]
+        },
+      SLTC
+        { name = "put file with wrong file name",
+          saved = [AddFile sId file createdAt (Just expiresAt) EntityActive, PutFile sId "/etc/passwd"],
+          compacted = [AddFile sId file createdAt (Just expiresAt) EntityActive],
+          state = M.fromList [(sId, (file, createdAt, Just expiresAt, EntityActive))]
+        },
+      SLTC
+        { name = "put file in another directory",
+          saved = [AddFile sId file createdAt (Just expiresAt) EntityActive, PutFile sId ("/srv/xftp" </> serverFileName sId)],
+          compacted = [AddFile sId file createdAt (Just expiresAt) EntityActive, PutFile sId ("/srv/xftp" </> serverFileName sId)],
+          state = M.fromList [(sId, (file, createdAt, Just expiresAt, EntityActive))]
         }
     ]
 

tests/CoreTests/XFTPStoreTests.hs

@@ -9,7 +9,7 @@ import Data.Word (Word32)
 import Simplex.FileTransfer.Protocol (FileInfo (..), SFileParty (..))
 import Simplex.FileTransfer.Server.Store
 import Simplex.FileTransfer.Server.Store.Postgres (PostgresFileStore, importFileStore, exportFileStore)
-import Simplex.FileTransfer.Server.StoreLog (closeStoreLog, readWriteFileStore, writeFileStore)
+import Simplex.FileTransfer.Server.StoreLog (closeStoreLog, readWriteFileStore, serverFileName, writeFileStore)
 import Simplex.FileTransfer.Transport (XFTPErrorType (..))
 import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Protocol (BlockingInfo (..), BlockingReason (..), EntityId (..))
@@ -17,6 +17,7 @@ import Simplex.Messaging.Server.QueueStore (ServerEntityStatus (..))
 import Simplex.Messaging.Server.StoreLog (openWriteStoreLog)
 import Simplex.Messaging.SystemTime (RoundedSystemTime (..))
 import System.Directory (doesFileExist, removeFile)
+import System.FilePath ((</>))
 import Test.Hspec hiding (fit, it)
 import UnliftIO.STM
 import Util
@@ -265,7 +266,7 @@ testMigrationRoundTrip = do
       sId2 = EntityId "migration_file_2"
       rId1 = EntityId "migration_rcp_1_"
   addFile stmStore sId1 fileInfo1 testCreatedAt (Just testExpiresAt) EntityActive `shouldReturn` Right ()
-  void $ setFilePath stmStore sId1 "/tmp/file1"
+  void $ setFilePath stmStore sId1 ("/tmp" </> serverFileName sId1)
   addRecipient stmStore sId1 (FileRecipient rId1 rcpKey1) `shouldReturn` Right ()
   let testBlockInfo = BlockingInfo {reason = BRSpam, notice = Nothing}
   addFile stmStore sId2 fileInfo2 testCreatedAt Nothing (EntityBlocked testBlockInfo) `shouldReturn` Right ()
@@ -289,7 +290,7 @@ testMigrationRoundTrip = do
   case result1 of
     Right (FileRec {fileInfo = fi, filePath, expiresAt, fileStatus}, _) -> do
       size fi `shouldBe` 128000
-      readTVarIO filePath `shouldReturn` Just "/tmp/file1"
+      readTVarIO filePath `shouldReturn` Just ("/tmp" </> serverFileName sId1)
       expiresAt `shouldBe` Just testExpiresAt
       readTVarIO fileStatus `shouldReturn` EntityActive
     Left e -> expectationFailure $ "getFile sId1 failed: " <> show e

tests/NtfServerTests.hs

@@ -136,7 +136,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           let dhSecret = C.dh' ntfDh dhPriv
               decryptCode nd =
                 let Right verification = nd .-> "verification"
-                    Right nonce = C.cbNonce <$> nd .-> "nonce"
+                    Right nonce = C.cbNonce =<< nd .-> "nonce"
                     Right pt = C.cbDecrypt dhSecret nonce verification
                  in NtfRegCode pt
           let code = decryptCode ntfData
@@ -160,7 +160,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           -- receive notification
           APNSMockRequest {notification} <- getMockNotification apns tkn
           let APNSNotification {aps = APNSMutableContent {}, notificationData = Just ntfData'} = notification
-              Right nonce' = C.cbNonce <$> ntfData' .-> "nonce"
+              Right nonce' = C.cbNonce =<< ntfData' .-> "nonce"
               Right message = ntfData' .-> "message"
               Right ntfDataDecrypted = C.cbDecrypt dhSecret nonce' message
               Right pnMsgs1 = parse pnMessagesP (AP.INTERNAL "error parsing PNMessageData") ntfDataDecrypted
@@ -171,7 +171,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           notifierId `shouldBe` nId
           -- receive message
           Resp "" _ (MSG RcvMessage {msgId = mId1, msgBody = EncRcvMsgBody body}) <- tGet1 rh
-          Right ClientRcvMsgBody {msgTs = mTs, msgBody} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret (C.cbNonce mId1) body)
+          Right ClientRcvMsgBody {msgTs = mTs, msgBody} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret mId1 body)
           mId1 `shouldBe` msgId
           mTs `shouldBe` msgTs
           (msgBody, "hello") #== "delivered from queue"
@@ -183,7 +183,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           APNSMockRequest {notification = APNSNotification {aps = APNSBackground _, notificationData = Just ntfData2}} <-
             getMockNotification apns tkn'
           let Right verification2 = ntfData2 .-> "verification"
-              Right nonce2 = C.cbNonce <$> ntfData2 .-> "nonce"
+              Right nonce2 = C.cbNonce =<< ntfData2 .-> "nonce"
               Right code2 = NtfRegCode <$> C.cbDecrypt dhSecret nonce2 verification2
           RespNtf "8" _ NROk <- signSendRecvNtf nh tknKey ("8", tId, TVFY code2)
           RespNtf "8a" _ (NRTkn NTActive) <- signSendRecvNtf nh tknKey ("8a", tId, TCHK)
@@ -191,7 +191,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           Resp "9" _ OK <- signSendRecv sh sKey ("9", sId, _SEND' "hello 2")
           APNSMockRequest {notification = notification3} <- getMockNotification apns tkn'
           let APNSNotification {aps = APNSMutableContent {}, notificationData = Just ntfData3} = notification3
-              Right nonce3 = C.cbNonce <$> ntfData3 .-> "nonce"
+              Right nonce3 = C.cbNonce =<< ntfData3 .-> "nonce"
               Right message3 = ntfData3 .-> "message"
               Right ntfDataDecrypted3 = C.cbDecrypt dhSecret nonce3 message3
               Right pnMsgs2 = parse pnMessagesP (AP.INTERNAL "error parsing PNMessageData") ntfDataDecrypted3
@@ -199,7 +199,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           smpServer3 `shouldBe` srv
           notifierId3 `shouldBe` nId
           Resp "" _ (MSG RcvMessage {msgId = mId2, msgBody = EncRcvMsgBody body2}) <- tGet1 rh
-          Right ClientRcvMsgBody {msgBody = "hello 2"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret (C.cbNonce mId2) body2)
+          Right ClientRcvMsgBody {msgBody = "hello 2"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret mId2 body2)
           Resp "10" _ OK <- signSendRecv rh rKey ("10", rId, ACK mId2)
 
           q2 <- createQueue rh sPub nPub
@@ -218,7 +218,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           Resp "14" _ OK <- signSendRecv sh sKey ("14", sId, _SEND' "hello 3")
           APNSMockRequest {notification = notification4} <- getMockNotification apns tkn'
           let APNSNotification {aps = APNSMutableContent {}, notificationData = Just ntfData4} = notification4
-              Right nonce4 = C.cbNonce <$> ntfData4 .-> "nonce"
+              Right nonce4 = C.cbNonce =<< ntfData4 .-> "nonce"
               Right message4 = ntfData4 .-> "message"
               Right ntfDataDecrypted4 = C.cbDecrypt dhSecret nonce4 message4
               Right pnMsgs4 = parse pnMessagesP (AP.INTERNAL "error parsing PNMessageData") ntfDataDecrypted4
@@ -226,14 +226,14 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           smpServer4 `shouldBe` srv
           notifierId4 `shouldBe` nId
           Resp "" _ (MSG RcvMessage {msgId = mId3, msgBody = EncRcvMsgBody body3}) <- tGet1 rh
-          Right ClientRcvMsgBody {msgBody = "hello 3"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret (C.cbNonce mId3) body3)
+          Right ClientRcvMsgBody {msgBody = "hello 3"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret mId3 body3)
           Resp "15" _ OK <- signSendRecv rh rKey ("15", rId, ACK mId3)
 
           -- deliver to queue with ntf sub created while SMP was offline
           Resp "16" _ OK <- signSendRecv sh sKey ("16", sId', _SEND' "hello 4")
           APNSMockRequest {notification = notification5} <- getMockNotification apns tkn'
           let APNSNotification {aps = APNSMutableContent {}, notificationData = Just ntfData5} = notification5
-              Right nonce5 = C.cbNonce <$> ntfData5 .-> "nonce"
+              Right nonce5 = C.cbNonce =<< ntfData5 .-> "nonce"
               Right message5 = ntfData5 .-> "message"
               Right ntfDataDecrypted5 = C.cbDecrypt dhSecret nonce5 message5
               Right pnMsgs5 = parse pnMessagesP (AP.INTERNAL "error parsing PNMessageData") ntfDataDecrypted5
@@ -241,7 +241,7 @@ testNotificationSubscription (ATransport t, msType) createQueue =
           smpServer5 `shouldBe` srv
           notifierId5 `shouldBe` nId'
           Resp "" _ (MSG RcvMessage {msgId = mId4, msgBody = EncRcvMsgBody body4}) <- tGet1 rh
-          Right ClientRcvMsgBody {msgBody = "hello 4"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret' (C.cbNonce mId4) body4)
+          Right ClientRcvMsgBody {msgBody = "hello 4"} <- pure $ parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt rcvDhSecret' mId4 body4)
           Resp "17" _ OK <- signSendRecv rh rKey' ("17", rId', ACK mId4)
           pure ()
 
@@ -307,7 +307,7 @@ registerToken nh apns token = do
   let dhSecret = C.dh' ntfDh dhPriv
       decryptCode nd =
         let Right verification = nd .-> "verification"
-            Right nonce = C.cbNonce <$> nd .-> "nonce"
+            Right nonce = C.cbNonce =<< nd .-> "nonce"
             Right pt = C.cbDecrypt dhSecret nonce verification
          in NtfRegCode pt
   let code = decryptCode ntfData

tests/PostgresSchemaDump.hs

@@ -15,6 +15,7 @@ import qualified Simplex.Messaging.Agent.Store.Postgres.Migrations as Migrations
 import Simplex.Messaging.Agent.Store.Shared (Migration (..), MigrationConfig (..), MigrationConfirmation (..), MigrationsToRun (..), toDownMigration)
 import Simplex.Messaging.Util (ifM, whenM)
 import System.Directory (doesFileExist, removeFile)
+import System.Info (os)
 import System.Process (readCreateProcess, shell)
 import Test.Hspec hiding (fit, it)
 import Util
@@ -57,14 +58,13 @@ postgresSchemaDumpTest migrations skipComparisonForDownMigrations testDBOpts@DBO
 
     getSchema :: FilePath -> IO String
     getSchema schemaPath = do
-      ci <- envCI
       let cmd =
             ("pg_dump " <> B.unpack connstr <> " --schema " <> B.unpack testDBSchema)
               <> " --schema-only --no-owner --no-privileges --no-acl --no-subscriptions --no-tablespaces > "
               <> schemaPath
       void $ readCreateProcess (shell cmd) ""
       threadDelay 20000
-      let sed = (if ci then "sed -i" else "sed -i ''")
+      let sed = (if os == "linux" then "sed -i" else "sed -i ''")
       void $ readCreateProcess (shell $ sed <> " '/^--/d; /^\\\\restrict/d; /^\\\\unrestrict/d' " <> schemaPath) ""
       sch <- readFile schemaPath
       sch `deepseq` pure sch

tests/RemoteControl.hs

@@ -9,13 +9,15 @@ module RemoteControl where
 
 import AgentTests.FunctionalAPITests (runRight)
 import Control.Logger.Simple
+import Control.Monad (void)
+import Control.Monad.Trans.Except (runExceptT)
 import Crypto.Random (ChaChaDRG)
 import qualified Data.Aeson as J
 import qualified Data.ByteString.Char8 as B
 import qualified Data.ByteString.Lazy.Char8 as LB
 import Data.List (stripPrefix)
 import Data.List.NonEmpty (NonEmpty (..))
-import Data.Time.Clock.System (SystemTime (..))
+import Data.Time.Clock.System (SystemTime (..), getSystemTime)
 import qualified Simplex.Messaging.Crypto as C
 import Simplex.Messaging.Encoding.String (StrEncoding (..))
 import Simplex.Messaging.Transport (TSbChainKeys (..))
@@ -24,8 +26,10 @@ import qualified Simplex.RemoteControl.Client as HC (RCHostClient (action))
 import qualified Simplex.RemoteControl.Client as RC
 import Simplex.RemoteControl.Discovery (mkLastLocalHost, preferAddress)
 import Simplex.RemoteControl.Invitation
-  ( RCInvitation (..),
+  ( RCEncInvitation (..),
+    RCInvitation (..),
     RCSignedInvitation,
+    signInvitation,
     verifySignedInvitation,
   )
 import Simplex.RemoteControl.Types
@@ -45,6 +49,7 @@ remoteControlTests = do
     it "should connect to existing pairing" testExistingPairing
   describe "Multicast discovery" $ do
     it "should find paired host and connect" testMulticast
+    it "should accept announcement only within timestamp window" testAnnouncementTimestamp
 
 testPreferAddress :: Spec
 testPreferAddress = do
@@ -244,6 +249,26 @@ testMulticast = do
         Nothing -> fail "timeout"
         Just _ -> pure ()
 
+testAnnouncementTimestamp :: IO ()
+testAnnouncementTimestamp = do
+  drg <- C.newRandom
+  RCHostPairing {caKey, caCert, idPrivKey} <- RC.newRCHostPairing drg
+  (hostDhPubKey, dhPrivKey) <- atomically $ C.generateKeyPair @'C.X25519 drg
+  (skey, sessPrivKey) <- atomically $ C.generateKeyPair @'C.Ed25519 drg
+  (dh, ctrlDhPrivKey) <- atomically $ C.generateKeyPair @'C.X25519 drg
+  nonce <- atomically $ C.randomCbNonce drg
+  now <- systemSeconds <$> getSystemTime
+  let pairing = RCCtrlPairing {caKey, caCert, ctrlFingerprint = C.KeyHash "test-ca", idPubKey = C.publicKey idPrivKey, dhPrivKey, prevDhPrivKey = Nothing}
+      announce offset = do
+        let inv = RCInvitation {ca = C.KeyHash "test-ca", host = "127.0.0.1", port = 5223, v = supportedRCPVRange, app = J.String "app", ts = MkSystemTime (now + offset) 0, skey, idkey = C.publicKey idPrivKey, dh}
+        encInvitation <- either (fail . show) pure $ C.cbEncrypt (C.dh' hostDhPubKey ctrlDhPrivKey) nonce (strEncode $ signInvitation sessPrivKey idPrivKey inv) 900
+        runExceptT . void $ RC.findRCCtrlPairing (pairing :| []) RCEncInvitation {dhPubKey = dh, nonce, encInvitation}
+  announce 0 `shouldReturn` Right ()
+  announce (-3600) `shouldReturn` Right ()
+  announce 3500 `shouldReturn` Right ()
+  announce (-3700) `shouldReturn` Left RCEInvitation
+  announce 3700 `shouldReturn` Left RCEInvitation
+
 runCtrl :: TVar ChaChaDRG -> Bool -> RCHostPairing -> MVar RCSignedInvitation -> IO (Async RCHostPairing)
 runCtrl drg multicast hp invVar = async . runRight $ do
   (_found, inv, hc, r) <- RC.connectRCHost drg hp (J.String "app") multicast Nothing Nothing

tests/SMPClient.hs

@@ -263,6 +263,7 @@ cfgMS msType = withStoreCfg (testServerStoreConfig msType) $ \serverStoreCfg ->
       prometheusMetricsFile = testPrometheusMetricsFile,
       pendingENDInterval = 500000,
       ntfDeliveryInterval = 200000,
+      authErrorDelay = 20000,
       smpCredentials =
         ServerCredentials
           { caCertificateFile = Just "tests/fixtures/ca.crt",
@@ -316,6 +317,9 @@ prevRange vr = vr {maxVersion = max (minVersion vr) (prevVersion $ maxVersion vr
 prevVersion :: Version v -> Version v
 prevVersion (Version v) = Version (v - 1)
 
+nextVersion :: Version v -> Version v
+nextVersion (Version v) = Version (v + 1)
+
 proxyCfg :: AServerConfig
 proxyCfg = proxyCfgMS (ASType SQSMemory SMSJournal)
 
@@ -346,6 +350,16 @@ proxyCfgShortTimeout =
         nt = NetworkTimeout {backgroundTimeout = 4_000000, interactiveTimeout = 4_000000}
      in cfg' {smpAgentCfg = aCfg {smpCfg = cCfg {networkConfig = (networkConfig cCfg) {tcpConnectTimeout = nt}}}}
 
+proxyCfgVPrev :: AStoreType -> AServerConfig
+proxyCfgVPrev msType =
+  updateCfg (proxyCfgMS msType) $ \cfg' ->
+    let aCfg = smpAgentCfg cfg'
+        cCfg = smpCfg aCfg
+     in cfg'
+          { smpServerVRange = prevRange $ smpServerVRange cfg',
+            smpAgentCfg = aCfg {smpCfg = cCfg {serverVRange = prevRange $ serverVRange cCfg}}
+          }
+
 withSmpServerStoreMsgLogOn :: HasCallStack => (ASrvTransport, AStoreType) -> ServiceName -> (HasCallStack => ThreadId -> IO a) -> IO a
 withSmpServerStoreMsgLogOn (t, msType) =
   withSmpServerConfigOn t $ updateCfg (cfgMS msType) $ \cfg' -> cfg' {storeNtfsFile = Just testStoreNtfsFile, serverStatsBackupFile = Just testServerStatsBackupFile}

tests/SMPProxyTests.hs

@@ -26,7 +26,7 @@ import qualified Data.List.NonEmpty as L
 import Data.Time.Clock (getCurrentTime)
 import SMPAgentClient
 import SMPClient
-import ServerTests (decryptMsgV3, sendRecv)
+import ServerTests (decryptMsgV3, responseTime, sendRecv)
 import Simplex.Messaging.Agent hiding (createConnection, joinConnection, sendMessage)
 import qualified Simplex.Messaging.Agent as A
 import Simplex.Messaging.Agent.Env.SQLite (AgentConfig (..), InitialAgentServers (..))
@@ -73,6 +73,12 @@ smpProxyTests = do
       xit "no SMP service at host/port" todo
       xit "bad SMP fingerprint" todo
     xit "batching proxy requests" todo
+    it "relay rejects forwarded command with changed version" $ \_ ->
+      testChangedFwdVersion
+    it "proxy rejects forwarded correlation ID that is not 24 bytes" $ \_ ->
+      testFwdCorrIdSize
+    it "relay delays AUTH errors to forwarded commands" $ \_ ->
+      testForwardedAuthErrorDelay
   describe "deliver message via SMP proxy" $ do
     let srv1 = SMPServer testHost testPort testKeyHash
         srv2 = SMPServer testHost2 testPort2 testKeyHash
@@ -95,6 +101,11 @@ smpProxyTests = do
           deliverMessageViaProxy proxyServ relayServ C.SEd25519 msg1 msg2
         it "max message size, X25519 keys" . twoServersFirstProxy $
           deliverMessageViaProxy proxyServ relayServ C.SX25519 msg1 msg2
+      describe "version compatibility" $ do
+        let deliver clientVR = deliverMessagesViaProxyVR clientVR srv1 srv2 C.SEd448 ["hello 1"] ["hello 2"]
+        it "prev client" . twoServersFirstProxy $ deliver (prevRange supportedClientSMPRelayVRange)
+        it "prev proxy" . twoServersPrevProxy $ deliver supportedClientSMPRelayVRange
+        it "prev relay" . twoServersPrevRelay $ deliver supportedClientSMPRelayVRange
       describe "stress test 1k" $ do
         let deliver n = deliverMessagesViaProxy srv1 srv2 C.SEd448 [] (map bshow [1 :: Int .. n])
         it "1x1000" . twoServersFirstProxy $ deliver 1000
@@ -155,6 +166,9 @@ smpProxyTests = do
     twoServersFirstProxy test msType = twoServers_ (proxyCfgMS msType) (updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 128, maxJournalMsgCount = 256}) test msType
     twoServersMoreConc test msType = twoServers_ (updateCfg (proxyCfgMS msType) $ \cfg_ -> cfg_ {serverClientConcurrency = 128}) (updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 128, maxJournalMsgCount = 256}) test msType
     twoServersNoConc test msType = twoServers_ (updateCfg (proxyCfgMS msType) $ \cfg_ -> cfg_ {serverClientConcurrency = 1}) (updateCfg (cfgMS msType) $ \cfg_ -> cfg_ {msgQueueQuota = 128, maxJournalMsgCount = 256}) test msType
+    twoServersPrevProxy test msType = twoServers_ (proxyCfgVPrev msType) (cfgMS msType) test msType
+    twoServersPrevRelay test msType = twoServers_ (proxyCfgMS msType) (prevServerVRange $ cfgMS msType) test msType
+    prevServerVRange cfg' = updateCfg cfg' $ \cfg_ -> cfg_ {smpServerVRange = prevRange $ smpServerVRange cfg_}
     twoServers_ :: AServerConfig -> AServerConfig -> IO () -> AStoreType -> IO ()
     twoServers_ cfg1 cfg2 runTest (ASType qsType _) =
       withSmpServerConfigOn (transport @TLS) cfg1 testPort $ \_ ->
@@ -167,11 +181,14 @@ deliverMessageViaProxy :: (C.AlgorithmI a, C.AuthAlgorithm a) => SMPServer -> SM
 deliverMessageViaProxy proxyServ relayServ alg msg msg' = deliverMessagesViaProxy proxyServ relayServ alg [msg] [msg']
 
 deliverMessagesViaProxy :: (C.AlgorithmI a, C.AuthAlgorithm a) => SMPServer -> SMPServer -> C.SAlgorithm a -> [ByteString] -> [ByteString] -> IO ()
-deliverMessagesViaProxy proxyServ relayServ alg unsecuredMsgs securedMsgs = do
+deliverMessagesViaProxy = deliverMessagesViaProxyVR $ mkVersionRange minServerSMPRelayVersion currentClientSMPRelayVersion
+
+deliverMessagesViaProxyVR :: (C.AlgorithmI a, C.AuthAlgorithm a) => VersionRangeSMP -> SMPServer -> SMPServer -> C.SAlgorithm a -> [ByteString] -> [ByteString] -> IO ()
+deliverMessagesViaProxyVR clientVR proxyServ relayServ alg unsecuredMsgs securedMsgs = do
   g <- C.newRandom
   -- set up proxy
   ts <- getCurrentTime
-  pc' <- getProtocolClient g NRMInteractive (1, proxyServ, Nothing) defaultSMPClientConfig {serverVRange = mkVersionRange minServerSMPRelayVersion currentClientSMPRelayVersion} [] Nothing ts (\_ -> pure ())
+  pc' <- getProtocolClient g NRMInteractive (1, proxyServ, Nothing) defaultSMPClientConfig {serverVRange = clientVR, proxiedRelayVRange = clientVR} [] Nothing ts (\_ -> pure ())
   pc <- either (fail . show) pure pc'
   THAuthClient {} <- maybe (fail "getProtocolClient returned no thAuth") pure $ thAuth $ thParams pc
   -- set up relay
@@ -191,6 +208,8 @@ deliverMessagesViaProxy proxyServ relayServ alg unsecuredMsgs securedMsgs = do
   forM_ unsecuredMsgs $ \msg -> do
     runExceptT' (proxySMPMessage pc NRMInteractive sess Nothing sndId noMsgFlags msg) `shouldReturn` Right ()
     runExceptT' (proxySMPMessage pc NRMInteractive sess {prSessionId = "bad session"} Nothing sndId noMsgFlags msg) `shouldReturn` Left (ProxyProtocolError $ SMP.PROXY SMP.NO_SESSION)
+    forM_ ([prevVersion minServerSMPRelayVersion, nextVersion currentServerSMPRelayVersion] :: [VersionSMP]) $ \v ->
+      runExceptT' (proxySMPMessage pc NRMInteractive sess {prVersion = v} Nothing sndId noMsgFlags msg) `shouldReturn` Left (ProxyProtocolError $ SMP.PROXY $ SMP.PROTOCOL $ SMP.PROXY $ SMP.BROKER $ SMP.TRANSPORT TEVersion)
     -- receive 1
     (_tSess, _, [(_entId, STEvent (Right (SMP.MSG RcvMessage {msgId, msgBody = EncRcvMsgBody encBody})))]) <- atomically $ readTBQueue msgQ
     dec msgId encBody `shouldBe` Right msg
@@ -229,7 +248,7 @@ agentDeliverMessageViaProxy aTestCfg@(aSrvs, _, aViaProxy) bTestCfg@(bSrvs, _, b
   withAgent 1 aCfg (servers aTestCfg) testDB $ \alice ->
     withAgent 2 aCfg (servers bTestCfg) testDB2 $ \bob -> runRight_ $ do
       (bobId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMInvitation Nothing Nothing CR.IKPQOn False SMSubscribe
-      aliceId <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+      (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
       sqSecured <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
       liftIO $ sqSecured `shouldBe` True
       ("", _, A.CONF confId pqSup' _ "bob's connInfo") <- get alice
@@ -285,7 +304,7 @@ agentDeliverMessagesViaProxyConc agentServers msgs =
     -- otherwise the CONF messages would get mixed with MSG
     prePair alice bob = do
       (bobId, CCLink qInfo Nothing) <- runExceptT' $ A.createConnection alice NRMInteractive 1 True True SCMInvitation Nothing Nothing CR.IKPQOn False SMSubscribe
-      aliceId <- runExceptT' $ A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+      (aliceId, _) <- runExceptT' $ A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
       sqSecured <- runExceptT' $ A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
       liftIO $ sqSecured `shouldBe` True
       confId <-
@@ -344,7 +363,7 @@ agentViaProxyRetryOffline = do
       withServer $ \_ -> do
         (aliceId, bobId) <- withServer2 $ \_ -> runRight $ do
           (bobId, CCLink qInfo Nothing) <- A.createConnection alice NRMInteractive 1 True True SCMInvitation Nothing Nothing CR.IKPQOn False SMSubscribe
-          aliceId <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
+          (aliceId, _) <- A.prepareConnectionToJoin bob 1 True qInfo PQSupportOn
           sqSecured <- A.joinConnection bob NRMInteractive 1 aliceId True qInfo "bob's connInfo" PQSupportOn SMSubscribe
           liftIO $ sqSecured `shouldBe` True
           ("", _, A.CONF confId pqSup' _ "bob's connInfo") <- get alice
@@ -446,6 +465,46 @@ requestRelaySession =
   testSMPClient_ "localhost" testPort supportedServerSMPRelayVRange Nothing $ \(th :: THandleSMP TLS 'TClient) ->
     (\(_, _, reply) -> reply) <$> sendRecv th (Nothing, "1", NoEntity, SMP.PRXY testSMPServer2 Nothing)
 
+testChangedFwdVersion :: IO ()
+testChangedFwdVersion =
+  withSmpServerConfigOn (transport @TLS) cfg testPort $ \_ -> do
+    g <- C.newRandom
+    ts <- getCurrentTime
+    rc <- either (fail . show) pure =<< getProtocolClient g NRMInteractive (1, testSMPServer, Nothing) defaultSMPClientConfig [] Nothing ts (\_ -> pure ())
+    THAuthClient {peerServerPubKey} <- maybe (fail "getProtocolClient returned no thAuth") pure $ thAuth $ thParams rc
+    (cmdPubKey, cmdPrivKey) <- atomically $ C.generateKeyPair g
+    nonce@(C.CbNonce corrId) <- atomically $ C.randomCbNonce g
+    let v = currentClientSMPRelayVersion
+    et <- either (fail . show) (pure . SMP.EncTransmission) $ C.cbEncrypt (C.dh' peerServerPubKey cmdPrivKey) (SMP.encTransmissionNonce v nonce) "" SMP.paddedProxiedTLength
+    let forward fwdVersion = forwardSMPTransmission rc (either error id $ C.corrCbNonce corrId) fwdVersion cmdPubKey et
+    _ <- runExceptT' $ forward v
+    runExceptT (forward $ prevVersion v) `shouldReturn` Left (PCEProtocolError SMP.CRYPTO)
+
+testFwdCorrIdSize :: IO ()
+testFwdCorrIdSize =
+  withSmpServerConfigOn (transport @TLS) proxyCfg testPort $ \_ ->
+    withSmpServerConfigOn (transport @TLS) cfgJ2 testPort2 $ \_ ->
+      testSMPClient_ "localhost" testPort supportedServerSMPRelayVRange Nothing $ \(th :: THandleSMP TLS 'TClient) -> do
+        (_, _, Right (SMP.PKEY sessId _ _)) <- sendRecv th (Nothing, "1", NoEntity, SMP.PRXY testSMPServer2 Nothing)
+        g <- C.newRandom
+        (cmdPubKey, _) <- atomically $ C.generateKeyPair g
+        forM_ (["", "2", "3333333333333333333333333"] :: [ByteString]) $ \corrId -> do
+          (_, _, reply) <- sendRecv th (Nothing, corrId, SMP.EntityId sessId, SMP.PFWD currentClientSMPRelayVersion cmdPubKey (SMP.EncTransmission ""))
+          reply `shouldBe` Right (SMP.ERR $ SMP.CMD SMP.SYNTAX)
+
+testForwardedAuthErrorDelay :: IO ()
+testForwardedAuthErrorDelay =
+  withSmpServerConfigOn (transport @TLS) proxyCfg testPort $ \_ ->
+    withSmpServerConfigOn (transport @TLS) (updateCfg cfgJ2 $ \cfg_ -> cfg_ {authErrorDelay = 500000}) testPort2 $ \_ -> do
+      g <- C.newRandom
+      ts <- getCurrentTime
+      pc <- either (fail . show) pure =<< getProtocolClient g NRMInteractive (1, testSMPServer, Nothing) defaultSMPClientConfig [] Nothing ts (\_ -> pure ())
+      sess <- runExceptT' $ connectSMPProxiedRelay pc NRMInteractive testSMPServer2 Nothing
+      (sendTime, Left (PCEProtocolError SMP.AUTH)) <- responseTime $ runExceptT $ proxySMPMessage pc NRMInteractive sess Nothing (SMP.EntityId "1234") noMsgFlags "hello"
+      (getLinkTime, Left (PCEProtocolError SMP.AUTH)) <- responseTime $ runExceptT $ proxyGetSMPQueueLink pc NRMInteractive sess (SMP.EntityId "1234")
+      sendTime `shouldSatisfy` (>= 0.5)
+      getLinkTime `shouldSatisfy` (>= 0.5)
+
 -- Shared "phase 2" of the reconnection tests: start a healthy relay, confirm it is reachable
 -- directly (PING, not via the proxy) so a proxy failure can only mean the proxy didn't reconnect,
 -- let any stored connection error expire, then require the proxy to establish the session (PKEY).

tests/ServerTests.hs

@@ -36,6 +36,7 @@ import Data.List.NonEmpty (NonEmpty)
 import Data.List (isPrefixOf)
 import Data.Maybe (catMaybes)
 import Data.String (IsString (..))
+import Data.Time.Clock (NominalDiffTime, diffUTCTime, getCurrentTime)
 import Text.Read (readMaybe)
 import Data.Type.Equality
 import qualified Data.X509.Validation as XV
@@ -58,7 +59,6 @@ import Simplex.Messaging.Transport.Credentials
 import Simplex.Messaging.Util (whenM)
 import System.Directory (doesDirectoryExist, doesFileExist, removeDirectoryRecursive, removeFile)
 import System.IO (IOMode (..), withFile)
-import System.TimeIt (timeItT)
 import System.Timeout
 import Test.HUnit
 import Test.Hspec hiding (fit, it)
@@ -97,6 +97,7 @@ serverTests = do
   describe "Restore messages (old / v2)" testRestoreExpireMessages
   describe "Save prometheus metrics" testPrometheusMetrics
   describe "Timing of AUTH error" testTiming
+  describe "Delay of AUTH error" testAuthErrorDelay
   describe "Message notifications" $ do
     testMessageNotifications
     testMessageServiceNotifications
@@ -166,7 +167,7 @@ signSend_ h@THandle {params} (C.APrivateAuthKey a pk) serviceKey_ (corrId, qId,
     authorize t = (,(`C.sign'` t) <$> serviceKey_) <$> case a of
       C.SEd25519 -> Just . TASignature . C.ASignature C.SEd25519 $ C.sign' pk t'
       C.SEd448 -> Just . TASignature . C.ASignature C.SEd448 $ C.sign' pk t'
-      C.SX25519 -> (\THAuthClient {peerServerPubKey = k} -> TAAuthenticator $ C.cbAuthenticate k pk (C.cbNonce corrId) t') <$> thAuth params
+      C.SX25519 -> (\THAuthClient {peerServerPubKey = k} -> TAAuthenticator $ C.cbAuthenticate k pk (C.unsafeCbNonce corrId) t') <$> thAuth params
 #if !MIN_VERSION_base(4,18,0)
       _sx448 -> undefined -- ghc8107 fails to the branch excluded by types
 #endif
@@ -194,12 +195,12 @@ _SEND = SEND noMsgFlags
 _SEND' :: MsgBody -> Command 'Sender
 _SEND' = SEND MsgFlags {notification = True}
 
-decryptMsgV2 :: C.DhSecret 'C.X25519 -> ByteString -> ByteString -> Either C.CryptoError ByteString
-decryptMsgV2 dhShared = C.cbDecrypt dhShared . C.cbNonce
+decryptMsgV2 :: C.DhSecret 'C.X25519 -> MsgId -> ByteString -> Either C.CryptoError ByteString
+decryptMsgV2 = C.cbDecrypt
 
-decryptMsgV3 :: C.DhSecret 'C.X25519 -> ByteString -> ByteString -> Either String MsgBody
+decryptMsgV3 :: C.DhSecret 'C.X25519 -> MsgId -> ByteString -> Either String MsgBody
 decryptMsgV3 dhShared nonce body =
-  case parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt dhShared (C.cbNonce nonce) body) of
+  case parseAll clientRcvMsgBodyP =<< first show (C.cbDecrypt dhShared nonce body) of
     Right ClientRcvMsgBody {msgBody} -> Right msgBody
     Right ClientRcvMsgQuota {} -> Left "ClientRcvMsgQuota"
     Left e -> Left e
@@ -425,7 +426,7 @@ testCreateDelete =
       Resp "dabc" _ err8 <- sendRecv sh ("", "dabc", sId, _SEND "hello")
       (err8, ERR AUTH) #== "rejects unsigned SEND too when deleted"
 
-      Resp "abcd" _ err11 <- signSendRecv rh rKey ("abcd", rId, ACK "")
+      Resp "abcd" _ err11 <- signSendRecv rh rKey ("abcd", rId, ACK mId2)
       (err11, ERR AUTH) #== "rejects ACK when conn deleted - the second message is deleted"
 
       Resp "bcda" _ err9 <- signSendRecv rh rKey ("bcda", rId, OFF)
@@ -609,7 +610,7 @@ testGetSubCommands =
       Resp "4" _ (Msg mId2 msg2) <- signSendRecv rh1 rKey ("4", rId, ACK mId1)
       (dec mId2 msg2, Right "hello 2") #== "received from queue via SUB"
       -- bad msgId returns error
-      Resp "5" _ (ERR NO_MSG) <- signSendRecv rh2 rKey ("5", rId, ACK "1234")
+      Resp "5" _ (ERR NO_MSG) <- signSendRecv rh2 rKey ("5", rId, ACK mId2)
       -- already ACK'd by subscriber, but still returns OK when msgId matches
       Resp "5a" _ OK <- signSendRecv rh2 rKey ("5a", rId, ACK mId1)
       -- msg2 is not lost - even if subscriber does not ACK it, it is delivered to getter
@@ -904,7 +905,7 @@ testServiceSubsTotalCount =
                 _ -> pure Nothing
             ]
         threadDelay 1500000
-        readFile testPrometheusMetricsFile >>= \m -> readServiceSubsMetric m `shouldBe` Just 2
+        B.readFile testPrometheusMetricsFile >>= \m -> readServiceSubsMetric (B.unpack m) `shouldBe` Just 2
 
       -- Phase 2: associate 1 more queue via SUB, reconnect with SUBS 3, check metric = 3
       (rPub3, rKey3) <- atomically $ C.generateAuthKeyPair C.SEd25519 g
@@ -934,7 +935,7 @@ testServiceSubsTotalCount =
                 _ -> pure Nothing
             ]
         threadDelay 1500000
-        readFile testPrometheusMetricsFile >>= \m -> readServiceSubsMetric m `shouldBe` Just 3
+        B.readFile testPrometheusMetricsFile >>= \m -> readServiceSubsMetric (B.unpack m) `shouldBe` Just 3
 
 readServiceSubsMetric :: String -> Maybe Int
 readServiceSubsMetric content =
@@ -1274,21 +1275,21 @@ testTiming =
         smpTest2Cfg (cfgMS msType) supportedServerSMPRelayVRange t $ \rh sh ->
           testSameTiming rh sh tst msType
   where
-    testName :: (C.AuthAlg, C.AuthAlg, Int) -> String
-    testName (C.AuthAlg goodKeyAlg, C.AuthAlg badKeyAlg, _) = unwords ["queue key:", show goodKeyAlg, "/ used key:", show badKeyAlg]
-    timingTests :: [(C.AuthAlg, C.AuthAlg, Int)]
+    testName :: (C.AuthAlg, C.AuthAlg) -> String
+    testName (C.AuthAlg goodKeyAlg, C.AuthAlg badKeyAlg) = unwords ["queue key:", show goodKeyAlg, "/ used key:", show badKeyAlg]
+    timingTests :: [(C.AuthAlg, C.AuthAlg)]
     timingTests =
-      [ (C.AuthAlg C.SEd25519, C.AuthAlg C.SEd25519, 200), -- correct key type
-      -- (C.AuthAlg C.SEd25519, C.AuthAlg C.SEd448, 150),
-      -- (C.AuthAlg C.SEd25519, C.AuthAlg C.SX25519, 200),
-        (C.AuthAlg C.SEd448, C.AuthAlg C.SEd25519, 200),
-        (C.AuthAlg C.SEd448, C.AuthAlg C.SEd448, 150), -- correct key type
-        (C.AuthAlg C.SEd448, C.AuthAlg C.SX25519, 200),
-        (C.AuthAlg C.SX25519, C.AuthAlg C.SEd25519, 200),
-        (C.AuthAlg C.SX25519, C.AuthAlg C.SEd448, 150),
-        (C.AuthAlg C.SX25519, C.AuthAlg C.SX25519, 200) -- correct key type
+      [ (C.AuthAlg C.SEd25519, C.AuthAlg C.SEd25519), -- correct key type
+      -- (C.AuthAlg C.SEd25519, C.AuthAlg C.SEd448),
+      -- (C.AuthAlg C.SEd25519, C.AuthAlg C.SX25519),
+        (C.AuthAlg C.SEd448, C.AuthAlg C.SEd25519),
+        (C.AuthAlg C.SEd448, C.AuthAlg C.SEd448), -- correct key type
+        (C.AuthAlg C.SEd448, C.AuthAlg C.SX25519),
+        (C.AuthAlg C.SX25519, C.AuthAlg C.SEd25519),
+        (C.AuthAlg C.SX25519, C.AuthAlg C.SEd448),
+        (C.AuthAlg C.SX25519, C.AuthAlg C.SX25519) -- correct key type
       ]
-    timeRepeat n = fmap fst . timeItT . forM_ (replicate n ()) . const
+    timeRepeat = fmap fst . responseTime . replicateM_ 5
     similarTime t1 t2 msType
       | t1 <= t2 = abs (1 - t1 / t2) < diff
       | otherwise = similarTime t2 t1 msType
@@ -1297,14 +1298,14 @@ testTiming =
         diff = case msType of
           ASType SQSPostgres _ -> 0.45
           _ -> 0.3
-    testSameTiming :: forall c. Transport c => THandleSMP c 'TClient -> THandleSMP c 'TClient -> (C.AuthAlg, C.AuthAlg, Int) -> AStoreType -> Expectation
-    testSameTiming rh sh (C.AuthAlg goodKeyAlg, C.AuthAlg badKeyAlg, n) msType = do
+    testSameTiming :: forall c. Transport c => THandleSMP c 'TClient -> THandleSMP c 'TClient -> (C.AuthAlg, C.AuthAlg) -> AStoreType -> Expectation
+    testSameTiming rh sh (C.AuthAlg goodKeyAlg, C.AuthAlg badKeyAlg) msType = do
       g <- C.newRandom
       (rPub, rKey) <- atomically $ C.generateAuthKeyPair goodKeyAlg g
       (dhPub, dhPriv :: C.PrivateKeyX25519) <- atomically $ C.generateKeyPair g
-      Resp "abcd" NoEntity (Ids rId sId srvDh) <- signSendRecv rh rKey ("abcd", NoEntity, New rPub dhPub)
+      Resp "abcdabcdabcdabcdabcdabcd" NoEntity (Ids rId sId srvDh) <- signSendRecv rh rKey ("abcdabcdabcdabcdabcdabcd", NoEntity, New rPub dhPub)
       let dec = decryptMsgV3 $ C.dh' srvDh dhPriv
-      Resp "cdab" _ resp <- signSendRecv rh rKey ("cdab", rId, SUB)
+      Resp "cdabcdabcdabcdabcdabcdab" _ resp <- signSendRecv rh rKey ("cdabcdabcdabcdabcdabcdab", rId, SUB)
       case resp of
         OK -> pure ()
         SOK Nothing -> pure ()
@@ -1314,9 +1315,9 @@ testTiming =
       runTimingTest rh badKey rId SUB
 
       (sPub, sKey) <- atomically $ C.generateAuthKeyPair goodKeyAlg g
-      Resp "dabc" _ OK <- signSendRecv rh rKey ("dabc", rId, KEY sPub)
+      Resp "dabcdabcdabcdabcdabcdabc" _ OK <- signSendRecv rh rKey ("dabcdabcdabcdabcdabcdabc", rId, KEY sPub)
 
-      Resp "bcda" _ OK <- signSendRecv sh sKey ("bcda", sId, _SEND "hello")
+      Resp "bcdabcdabcdabcdabcdabcda" _ OK <- signSendRecv sh sKey ("bcdabcdabcdabcdabcdabcda", sId, _SEND "hello")
       Resp "" _ (Msg mId msg) <- tGet1 rh
       (dec mId msg, Right "hello") #== "delivered from queue"
 
@@ -1325,17 +1326,17 @@ testTiming =
         runTimingTest :: PartyI p => THandleSMP c 'TClient -> C.APrivateAuthKey -> EntityId -> Command p -> IO ()
         runTimingTest h badKey qId cmd = do
           threadDelay 100000
-          _ <- timeRepeat n $ do
+          _ <- timeRepeat $ do
             -- "warm up" the server
-            Resp "dabc" _ (ERR AUTH) <- signSendRecv h badKey ("dabc", EntityId "1234", cmd)
+            Resp "dabcdabcdabcdabcdabcdabc" _ (ERR AUTH) <- signSendRecv h badKey ("dabcdabcdabcdabcdabcdabc", EntityId "1234", cmd)
             return ()
           threadDelay 100000
-          timeWrongKey <- timeRepeat n $ do
-            Resp "cdab" _ (ERR AUTH) <- signSendRecv h badKey ("cdab", qId, cmd)
+          timeWrongKey <- timeRepeat $ do
+            Resp "cdabcdabcdabcdabcdabcdab" _ (ERR AUTH) <- signSendRecv h badKey ("cdabcdabcdabcdabcdabcdab", qId, cmd)
             return ()
           threadDelay 100000
-          timeNoQueue <- timeRepeat n $ do
-            Resp "dabc" _ (ERR AUTH) <- signSendRecv h badKey ("dabc", EntityId "1234", cmd)
+          timeNoQueue <- timeRepeat $ do
+            Resp "dabcdabcdabcdabcdabcdabc" _ (ERR AUTH) <- signSendRecv h badKey ("dabcdabcdabcdabcdabcdabc", EntityId "1234", cmd)
             return ()
           let ok = similarTime timeNoQueue timeWrongKey msType
           unless ok . putStrLn . unwords $
@@ -1347,6 +1348,33 @@ testTiming =
             ]
           ok `shouldBe` True
 
+testAuthErrorDelay :: SpecWith (ASrvTransport, AStoreType)
+testAuthErrorDelay =
+  it "should send AUTH errors after the delay and other responses without it" $ \(ATransport t, msType) ->
+    smpTest2Cfg (updateCfg (cfgMS msType) $ \cfg' -> cfg' {authErrorDelay = 500000}) supportedClientSMPRelayVRange t $ \rh sh -> do
+      g <- C.newRandom
+      (rPub, rKey) <- atomically $ C.generateAuthKeyPair C.SEd25519 g
+      (dhPub, _ :: C.PrivateKeyX25519) <- atomically $ C.generateKeyPair g
+      Resp "1" NoEntity (Ids rId sId _) <- signSendRecv rh rKey ("1", NoEntity, New rPub dhPub)
+      (_, badKey) <- atomically $ C.generateAuthKeyPair C.SEd25519 g
+      (wrongKeyTime, Resp "2" _ (ERR AUTH)) <- responseTime $ signSendRecv rh badKey ("2", rId, SUB)
+      (noQueueTime, Resp "3" _ (ERR AUTH)) <- responseTime $ signSendRecv rh badKey ("3", EntityId "1234", SUB)
+      (noLinkTime, Resp "4" _ (ERR AUTH)) <- responseTime $ sendRecv sh ("", "4", EntityId "1234", LGET)
+      (subTime, Resp "5" _ (SOK Nothing)) <- responseTime $ signSendRecv rh rKey ("5", rId, SUB)
+      Resp "6" _ OK <- signSendRecv rh rKey ("6", rId, OFF)
+      (suspendedTime, Resp "7" _ (ERR AUTH)) <- responseTime $ sendRecv sh ("", "7", sId, _SEND "hello")
+      wrongKeyTime `shouldSatisfy` (>= 0.5)
+      noQueueTime `shouldSatisfy` (>= 0.5)
+      suspendedTime `shouldSatisfy` (>= 0.5)
+      noLinkTime `shouldSatisfy` (>= 0.5)
+      subTime `shouldSatisfy` (< 0.5)
+
+responseTime :: IO a -> IO (NominalDiffTime, a)
+responseTime action = do
+  t <- getCurrentTime
+  r <- action
+  (,r) . (`diffUTCTime` t) <$> getCurrentTime
+
 testMessageNotifications :: SpecWith (ASrvTransport, AStoreType)
 testMessageNotifications =
   it "should create simplex connection, subscribe notifier and deliver notifications" $ \(ATransport t, msType) -> do
@@ -1814,10 +1842,10 @@ serverSyntaxTests (ATransport t) = do
     it "no parameters" $ (sampleSig, "abcd", "12345678", SEND_) >#> ("", "abcd", "12345678", ERR $ CMD SYNTAX)
     it "no queue ID" $ (sampleSig, "bcda", "", (SEND_, ' ', noMsgFlags, ' ', "hello" :: ByteString)) >#> ("", "bcda", "", ERR $ CMD NO_ENTITY)
   describe "ACK" $ do
-    it "valid syntax" $ (sampleSig, "cdab", "12345678", (ACK_, ' ', "1234" :: ByteString)) >#> ("", "cdab", "12345678", ERR AUTH)
+    it "valid syntax" $ (sampleSig, "cdab", "12345678", (ACK_, ' ', "123456789012345678901234" :: ByteString)) >#> ("", "cdab", "12345678", ERR AUTH)
     it "no parameters" $ (sampleSig, "abcd", "12345678", ACK_) >#> ("", "abcd", "12345678", ERR $ CMD SYNTAX)
-    it "no queue ID" $ (sampleSig, "bcda", "", (ACK_, ' ', "1234" :: ByteString)) >#> ("", "bcda", "", ERR $ CMD NO_AUTH)
-    it "no signature" $ ("", "cdab", "12345678", (ACK_, ' ', "1234" :: ByteString)) >#> ("", "cdab", "12345678", ERR $ CMD NO_AUTH)
+    it "no queue ID" $ (sampleSig, "bcda", "", (ACK_, ' ', "123456789012345678901234" :: ByteString)) >#> ("", "bcda", "", ERR $ CMD NO_AUTH)
+    it "no signature" $ ("", "cdab", "12345678", (ACK_, ' ', "123456789012345678901234" :: ByteString)) >#> ("", "cdab", "12345678", ERR $ CMD NO_AUTH)
   describe "PING" $ do
     it "valid syntax" $ ("", "abcd", "", PING_) >#> ("", "abcd", "", PONG)
   describe "broker response not allowed" $ do

tests/Test.hs

@@ -109,7 +109,8 @@ main = do
           describe "SMP server schema dump" $
             postgresSchemaDumpTest
               serverMigrations
-              [ "20250320_short_links" -- snd_secure moves to the bottom on down migration
+              [ "20250320_short_links", -- snd_secure moves to the bottom on down migration
+                "20260918_expire_messages" -- msg_queue_expire moves to the bottom on down migration
               ] -- skipComparisonForDownMigrations
               testStoreDBOpts
               "src/Simplex/Messaging/Server/QueueStore/Postgres/server_schema.sql"

tests/Util.hs

@@ -56,7 +56,7 @@ instance Example a => Example (TestWrapper a) where
     ci <- envCI
     runTest `E.catches` [E.Handler (onTestFailure ci), E.Handler (onTestException ci)]
     where
-      tt = 120
+      tt = 300
       runTest =
         timeout (tt * 1000000) (evaluateExample action params hooks state) `finally` callCommand "sync" >>= \case
           Just r -> pure r

tests/XFTPAgent.hs

@@ -34,7 +34,7 @@ import Simplex.FileTransfer.Server.Env (AFStoreType, XFTPServerConfig (..), defa
 import Simplex.FileTransfer.Server.Store (STMFileStore)
 import Simplex.FileTransfer.Transport (XFTPErrorType (AUTH))
 import Simplex.FileTransfer.Types (RcvFileId, SndFileId)
-import Simplex.Messaging.Agent (AgentClient, testProtocolServer, xftpDeleteRcvFile, xftpDeleteSndFileInternal, xftpDeleteSndFileRemote, xftpReceiveFile, xftpSendDescription, xftpStartWorkers)
+import Simplex.Messaging.Agent (AgentClient, testProtocolServer, xftpDeleteRcvFile, xftpDeleteSndFileInternal, xftpDeleteSndFileRemote, xftpPrepareReceiveFile, xftpPrepareSendFile, xftpReceiveFile, xftpSendDescription, xftpStartReceiveFile, xftpStartSendFile, xftpStartWorkers)
 import qualified Simplex.Messaging.Agent as XA
 import Simplex.Messaging.Agent.Client (ProtocolTestFailure (..), ProtocolTestStep (..))
 import Simplex.Messaging.Agent.Env.SQLite (AgentConfig, xftpCfg)
@@ -84,6 +84,8 @@ xftpAgentTests =
       it "should send and receive with encrypted local files" testXFTPAgentSendReceiveEncrypted
       it "should send and receive large file with a redirect" testXFTPAgentSendReceiveRedirect
       it "should send and receive small file without a redirect" testXFTPAgentSendReceiveNoRedirect
+      it "should receive prepared file only after it is started" testXFTPAgentPrepareReceive
+      it "should send prepared file only after it is started" testXFTPAgentPrepareSend
       it "should extend storage time with an entitlement proof and report the granted expiry" $ \_ -> testXFTPAgentEntitlement
       describe "sending and receiving with version negotiation" $ beforeWith (const (pure ())) testXFTPAgentSendReceiveMatrix
       it "should resume receiving file after restart" $ \_ -> testXFTPAgentReceiveRestore
@@ -278,6 +280,56 @@ testXFTPAgentSendReceiveNoRedirect = withXFTPServer $ do
       inBytes <- B.readFile filePathIn
       B.readFile out `shouldReturn` inBytes
 
+testXFTPAgentPrepareReceive :: HasCallStack => AFStoreType -> IO ()
+testXFTPAgentPrepareReceive = withXFTPServer $ do
+  filePath <- createRandomFile
+  (_, _, rfd1, rfd2) <- withAgent 1 agentCfg initAgentServers testDB $ \sndr -> runRight $ testSend sndr filePath
+  rfId2 <- withAgent 2 agentCfg initAgentServers testDB2 $ \rcp -> runRight $ do
+    xftpStartWorkers rcp (Just recipientFiles)
+    rfId1 <- xftpReceiveFile rcp 1 rfd1 Nothing True
+    rfId2 <- xftpPrepareReceiveFile rcp 1 rfd2 Nothing True
+    rfProgress rcp $ mb 18
+    ("", rfId1', RFDONE _) <- rfGet rcp
+    liftIO $ do
+      rfId1' `shouldBe` rfId1
+      timeout 300000 (rfGet rcp) `shouldReturn` Nothing
+    pure rfId2
+  withAgent 3 agentCfg initAgentServers testDB2 $ \rcp' -> runRight_ $ do
+    xftpStartWorkers rcp' (Just recipientFiles)
+    liftIO $ timeout 300000 (rfGet rcp') `shouldReturn` Nothing
+    xftpStartReceiveFile rcp' rfId2
+    rfProgress rcp' $ mb 18
+    ("", rfId2', RFDONE path) <- rfGet rcp'
+    liftIO $ do
+      rfId2' `shouldBe` rfId2
+      file <- B.readFile filePath
+      B.readFile path `shouldReturn` file
+
+testXFTPAgentPrepareSend :: HasCallStack => AFStoreType -> IO ()
+testXFTPAgentPrepareSend = withXFTPServer $ do
+  filePath1 <- createRandomFile' "testfile1"
+  filePath2 <- createRandomFile' "testfile2"
+  sfId2 <- withAgent 1 agentCfg initAgentServers testDB $ \sndr -> runRight $ do
+    xftpStartWorkers sndr (Just senderFiles)
+    sfId1 <- xftpSendFile sndr 1 (CF.plain filePath1) 1
+    sfId2 <- xftpPrepareSendFile sndr 1 (CF.plain filePath2) 1 Nothing
+    sfProgress sndr $ mb 18
+    ("", sfId1', SFDONE _ _) <- sfGet sndr
+    liftIO $ do
+      sfId1' `shouldBe` sfId1
+      timeout 300000 (sfGet sndr) `shouldReturn` Nothing
+    pure sfId2
+  rfd <- withAgent 2 agentCfg initAgentServers testDB $ \sndr' -> runRight $ do
+    xftpStartWorkers sndr' (Just senderFiles)
+    liftIO $ timeout 300000 (sfGet sndr') `shouldReturn` Nothing
+    xftpStartSendFile sndr' sfId2
+    sfProgress sndr' $ mb 18
+    ("", sfId2', SFDONE _ [rfd]) <- sfGet sndr'
+    liftIO $ sfId2' `shouldBe` sfId2
+    pure rfd
+  withAgent 3 agentCfg initAgentServers testDB2 $ \rcp ->
+    runRight_ . void $ testReceive rcp rfd filePath2
+
 testXFTPAgentSendReceiveMatrix :: Spec
 testXFTPAgentSendReceiveMatrix = do
   describe "old server" $ do