fix(node): close review residuals: coalesce cold downloads to one in-flight task, compensate a timed-out upload with the rolled-back tree, aggregate partial ref restores, race-test the anchor lock, fail-closed snapshot test

Auto-deref (&mut tx) on the lock_repo_slug calls and rustfmt of the new test
lines. No behavior change.

.env.example

@@ -13,9 +13,16 @@ GITLAWB_PUBLIC_URL=https://your-node.example.com
 # ── Server ────────────────────────────────────────────────────────────────
 GITLAWB_HOST=0.0.0.0
 GITLAWB_PORT=7545
+# Optional address to bind a Prometheus /metrics exposition endpoint on (e.g.
+# 127.0.0.1:9091). Leave empty (default) to disable. Bind to localhost or a
+# private interface — the metrics endpoint is unauthenticated.
+GITLAWB_METRICS_ADDR=
 
 # ── Storage ───────────────────────────────────────────────────────────────
 GITLAWB_REPOS_DIR=/data/repos
+# Tigris (S3-compatible) bucket for repo storage. Leave empty (default) to
+# disable Tigris and use local-only storage.
+GITLAWB_TIGRIS_BUCKET=
 
 # PostgreSQL connection URL. Required.
 # When using the bundled docker-compose, this is wired automatically.
@@ -26,6 +33,12 @@ DATABASE_URL=postgresql://gitlawb:changeme@localhost:5432/gitlawb
 # connections open lazily. Size against the DB server's max_connections,
 # remembering admin tooling opens its own pool.
 GITLAWB_DB_MAX_CONNECTIONS=20
+# Maximum connections in the dedicated advisory-lock pool (separate from the
+# pool above). Each in-flight repo write pins one connection here for its whole
+# lifetime, so size it to peak concurrent distinct-repo writers — keeping it
+# separate is what stops a push burst from starving request handlers. Keep
+# (main pool + lock pool) within the DB server's max_connections.
+GITLAWB_DB_LOCK_POOL_MAX_CONNECTIONS=32
 # Seconds a request waits for a pool connection before failing with 503.
 GITLAWB_DB_ACQUIRE_TIMEOUT_SECS=5
 # Upper bound on each startup connect+migrate attempt, in seconds. Keep it
@@ -39,6 +52,9 @@ GITLAWB_DB_RETRY_INITIAL_SECS=5
 GITLAWB_DB_RETRY_MAX_SECS=60
 
 # ── IPFS pinning (Pinata) ─────────────────────────────────────────────────
+# URL of a local IPFS/Kubo node HTTP API (e.g. http://127.0.0.1:5001). Leave
+# empty (default) to disable local IPFS.
+GITLAWB_IPFS_API=
 # Get a JWT at https://app.pinata.cloud/developers/api-keys
 GITLAWB_PINATA_JWT=
 GITLAWB_PINATA_UPLOAD_URL=https://uploads.pinata.cloud/v3/files
@@ -127,6 +143,20 @@ GITLAWB_PUSH_RATE_LIMIT=600
 # the client IP. 0 disables. Default 120.
 GITLAWB_CREATE_RATE_LIMIT=120
 
+# ── Write rate limiting (non-creation authenticated writes) ───────────────
+# Max non-creation write requests per client IP per hour: issue/PR comments,
+# labels, stars, merges, protect/unprotect, replicas, visibility, tasks,
+# bounties, profile, and all GraphQL HTTP requests. The brake wraps the whole
+# /graphql route, so queries and the playground GET consume this bucket too, not
+# only mutations (GraphQL WebSocket subscriptions are excluded). Its own bucket,
+# separate from the creation and push brakes. Uses GITLAWB_TRUSTED_PROXY to
+# resolve the client IP.
+# NOTE: this is a per-IP aggregate across ALL those write actions, so behind a
+# shared NAT/egress IP (or with GITLAWB_TRUSTED_PROXY unset) many users collapse
+# onto one bucket — raise this for automation-heavy or multi-user single-IP
+# deployments. 0 disables. Default 600.
+GITLAWB_WRITE_RATE_LIMIT=600
+
 # ── Peer-sync rate limiting (per client IP, uses GITLAWB_TRUSTED_PROXY below) ─
 # /api/v1/peers/announce and /api/v1/sync/notify accept unsigned requests from
 # known peers and run at higher frequency, so a generous bucket. Separate from

.github/workflows/pr-checks.yml

@@ -6,13 +6,6 @@ on:
   # result before it lands. Guards against semantic merge races where two
   # individually-green PRs break main in combination (#113 + #145 → #229).
   merge_group:
-  # Release-branch heads are created and pushed by GITHUB_TOKEN (release-please
-  # and the lock-sync job), and GitHub's recursion guard means those pushes
-  # trigger no workflow runs. workflow_dispatch is the documented carve-out that
-  # GITHUB_TOKEN CAN trigger, so the release flow dispatches this workflow at
-  # the release branch to attach check runs to its head. Read-only workflow
-  # (contents: read), so the dispatch surface adds no privilege.
-  workflow_dispatch:
   push:
     branches:
       - main
@@ -53,7 +46,7 @@ jobs:
         run: cargo fmt --all -- --check
 
       - name: cargo clippy
-        run: cargo clippy --locked --workspace --all-targets -- -D warnings
+        run: cargo clippy --workspace --all-targets -- -D warnings
 
   test:
     name: test (${{ matrix.toolchain }})
@@ -101,7 +94,7 @@ jobs:
       - name: cargo test
         env:
           DATABASE_URL: postgres://postgres:postgres@localhost:5432/gitlawb_test
-        run: cargo test --locked --workspace
+        run: cargo test --workspace
 
   build-release:
     name: build --release
@@ -124,7 +117,7 @@ jobs:
           key: release
 
       - name: cargo build --release
-        run: cargo build --locked --release --workspace
+        run: cargo build --release --workspace
 
   audit:
     name: cargo audit
@@ -206,7 +199,7 @@ jobs:
           key: msrv
 
       - name: cargo check on MSRV
-        run: cargo check --locked --workspace --all-targets
+        run: cargo check --workspace --all-targets
 
   docker-build:
     name: Docker build smoke test
@@ -253,33 +246,4 @@ jobs:
           key: windows
 
       - name: cargo test (shipped Windows crates)
-        run: cargo test --locked -p gl -p git-remote-gitlawb
-
-  # gitlawb-core is embedded by every consumer (gl, git-remote-gitlawb, the node
-  # daemon), so it must stay lean. This gate fails if gitlawb-core's normal
-  # (non-dev, non-build) dependency tree gains any crate not on the allowlist in
-  # ci/gitlawb-core-allowed-deps.txt. Exhaustive-by-construction: a new heavy
-  # dependency reds CI whether or not anyone thought to ban it, which a denylist
-  # cannot do. Regen instructions live in the allowlist header.
-  core-deps-purity:
-    name: gitlawb-core dependency purity
-    runs-on: ubuntu-latest
-    timeout-minutes: 15
-    steps:
-      - name: Check out repository
-        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-        with:
-          persist-credentials: false
-
-      - name: Set up Rust toolchain
-        uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # stable
-        with:
-          toolchain: stable
-
-      - name: Cache cargo
-        uses: Swatinem/rust-cache@98c8021b550208e191a6a3145459bfc9fb29c4c0 # v2.8.0
-        with:
-          key: core-deps-purity
-
-      - name: Check gitlawb-core dependency allowlist
-        run: bash scripts/check-gitlawb-core-deps.sh
+        run: cargo test -p gl -p git-remote-gitlawb

.github/workflows/release.yml

@@ -41,7 +41,6 @@ jobs:
       release_created: ${{ steps.release.outputs.release_created }}
       tag_name: ${{ steps.release.outputs.tag_name }}
       version: ${{ steps.release.outputs.version }}
-      pr: ${{ steps.release.outputs.pr }}
     steps:
       - name: Run release-please
         id: release
@@ -49,143 +48,6 @@ jobs:
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
 
-  # The release PR bumps the workspace crate versions in the Cargo.tomls but
-  # release-please cannot update Cargo.lock, so without this job every release
-  # merge ships a tag whose lockfile disagrees with its manifests (a --locked
-  # build from the tag fails, and pr-checks builds --locked). Sync the lock on
-  # the release branch so the tagged tree is internally consistent.
-  #
-  # Runs on every push-triggered release run, not only when release-please
-  # reports the PR: the `pr` output is set only on runs that CREATE or UPDATE
-  # the release PR, so gating on it would leave an already-open release PR
-  # (opened before this job existed, or untouched by a non-releasable push)
-  # permanently without the sync and without check runs. The resolve step
-  # discovers any open release-please PR itself and no-ops when there is none.
-  sync-release-lock:
-    name: Sync Cargo.lock on the release PR
-    needs: release-please
-    if: ${{ !cancelled() && github.repository == 'Gitlawb/node' && github.event_name != 'workflow_dispatch' }}
-    runs-on: ubuntu-latest
-    permissions:
-      contents: write
-      pull-requests: read
-      # actions: write lets the final step dispatch pr-checks on the release
-      # branch. Note this grant is repo-wide dispatch rights (GitHub cannot
-      # scope it to one workflow); acceptable here because reaching this job's
-      # steps already requires write to main. Needed because every push to the
-      # release branch (release-please's and the lock sync below) is made with
-      # GITHUB_TOKEN, which triggers no workflow runs, so without an explicit
-      # dispatch the release PR's head carries no check runs and the --locked
-      # gate never validates it.
-      actions: write
-    steps:
-      # Prefer the fresh `pr` output when this run just created/updated the
-      # release PR; otherwise fall back to querying for an open release-please
-      # branch (prefix match: the branch carries `--components--` suffixes).
-      - name: Resolve the open release PR branch
-        id: resolve
-        env:
-          GH_TOKEN: ${{ github.token }}
-          OWNER: ${{ github.repository_owner }}
-          BRANCH_FROM_OUTPUT: ${{ needs.release-please.outputs.pr && fromJSON(needs.release-please.outputs.pr).headBranchName || '' }}
-        run: |
-          branch="$BRANCH_FROM_OUTPUT"
-          if [ -z "$branch" ]; then
-            # The branch name alone is not a trust signal. `gh pr list` includes
-            # fork PRs, and a fork's headRefName is bare (no owner prefix), so a
-            # prefix match by itself lets any fork nominate the branch this job
-            # checks out, commits to, and dispatches checks on. Require the head
-            # repo to be THIS repository and the author to be the release-please
-            # app, so neither a fork nor a same-repo human branch named
-            # `release-please--*` can be selected. Then take the highest PR
-            # number: `[0]` depended on gh's default ordering, which is not a
-            # documented guarantee, and picking a stable one keeps a second
-            # (component) release PR from making the choice flap between runs.
-            branch=$(gh pr list --repo "$GITHUB_REPOSITORY" --state open \
-              --json headRefName,headRepositoryOwner,number,author \
-              --jq '[ .[]
-                      | select(.headRepositoryOwner.login == env.OWNER)
-                      | select(.author.login == "app/github-actions")
-                      | select(.headRefName | startswith("release-please--"))
-                    ] | max_by(.number).headRefName // empty')
-          fi
-          if [ -z "$branch" ]; then
-            echo "no open release PR; nothing to sync"
-          fi
-          echo "branch=$branch" >> "$GITHUB_OUTPUT"
-
-      - name: Checkout release branch
-        if: ${{ steps.resolve.outputs.branch != '' }}
-        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-        with:
-          ref: ${{ steps.resolve.outputs.branch }}
-
-      # Resolve on a declared toolchain instead of whatever Rust the runner
-      # image happens to ship, so the lock this job writes comes off the same
-      # stable channel the other cargo jobs use. Not a claim of identical
-      # resolution: `stable` still moves between runs, and the MSRV gate
-      # checks this same lock on 1.91 (pr-checks.yml). It only removes the
-      # runner image as an undeclared input.
-      - name: Install Rust toolchain
-        if: ${{ steps.resolve.outputs.branch != '' }}
-        uses: dtolnay/rust-toolchain@b3b07ba8b418998c39fb20f53e8b695cdcc8de1b # stable
-        with:
-          toolchain: stable
-
-      - name: Sync workspace crate versions into Cargo.lock
-        if: ${{ steps.resolve.outputs.branch != '' }}
-        run: cargo update --workspace
-
-      - name: Commit and push when the lock changed
-        if: ${{ steps.resolve.outputs.branch != '' }}
-        run: |
-          if git diff --quiet Cargo.lock; then
-            echo "Cargo.lock already in sync"
-            exit 0
-          fi
-          git config user.name "github-actions[bot]"
-          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
-          git add Cargo.lock
-          git commit -m "chore: sync Cargo.lock with the release version bump"
-          git push
-
-      # Every push to the release branch is made with GITHUB_TOKEN (release-
-      # please's own pushes and the lock sync above), and events created by
-      # GITHUB_TOKEN start no workflow runs. workflow_dispatch is the documented
-      # exception GITHUB_TOKEN may trigger, so dispatch pr-checks at the release
-      # branch to attach check runs to its current head. Unconditional (also on
-      # the already-in-sync path) because the branch's ORIGINAL head has the
-      # same no-runs problem. Loud on failure by design: a silent fallback
-      # would recreate the checkless-head problem, and a job rerun is safe
-      # (sync no-ops, dispatch retries). Loop-safe: pr-checks is read-only and
-      # dispatches nothing. Runs after the push ack, so the dispatched ref tip
-      # is the sync commit. Merge-queue note: this validates the branch tip;
-      # the enforced merge queue's merge_group run still re-validates the true
-      # merged result before landing.
-      - name: Dispatch PR checks on the release branch
-        if: ${{ steps.resolve.outputs.branch != '' }}
-        env:
-          GH_TOKEN: ${{ github.token }}
-          BRANCH: ${{ steps.resolve.outputs.branch }}
-        run: |
-          # Every dispatch failure fails the job, including the 422 a release
-          # branch cut before pr-checks gained its workflow_dispatch trigger
-          # returns ("does not have a workflow_dispatch trigger", reported
-          # against the dispatched ref's copy of the file). An earlier revision
-          # tolerated exactly that 422 on the theory that release-please would
-          # rewrite the branch and heal it, but a rewrite only happens when a
-          # later releasable commit lands: an open release PR nobody touches is
-          # never rewritten. Tolerating it therefore left the lock-sync commit
-          # this job had just pushed sitting on the release head with no
-          # --locked validation, which is the precise state this workflow exists
-          # to prevent. The tolerance is also no longer needed: once this lands,
-          # every release branch is cut from a main that already carries the
-          # trigger, so the only branch that could 422 is one open at merge time
-          # (there is none), and a job rerun after a manual rebase is safe (the
-          # sync no-ops, the dispatch retries).
-          gh api "repos/${GITHUB_REPOSITORY}/actions/workflows/pr-checks.yml/dispatches" \
-            -f "ref=$BRANCH"
-
   # Each architecture builds NATIVELY (amd64 on ubuntu-latest, arm64 on
   # ubuntu-24.04-arm) and pushes by digest; docker-manifest below stitches the
   # digests into the tagged multi-arch image. No QEMU: emulating the arm64
@@ -446,7 +308,7 @@ jobs:
           set -euo pipefail
           args=""
           for b in $BINS; do args="$args -p $b"; done
-          cargo build --release --locked --target ${{ matrix.target }} $args
+          cargo build --release --target ${{ matrix.target }} $args
 
       # Run each packaged binary's --version so a broken release artifact fails the
       # build instead of shipping. Every target builds on a matching-arch runner,

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "0.7.1"
+  ".": "0.6.0"
 }

CHANGELOG.md

@@ -1,49 +1,5 @@
 # Changelog
 
-## [0.7.1](https://github.com/Gitlawb/node/compare/v0.7.0...v0.7.1) (2026-08-03)
-
-
-### Bug Fixes
-
-* **ci:** make the docker dependency cache work and unbreak Dockerfile.bins ([99b2b9a](https://github.com/Gitlawb/node/commit/99b2b9a16662761a3959a2941798e6823621bda5))
-* **ci:** make the docker dependency cache work and unbreak Dockerfile.bins ([a49e04d](https://github.com/Gitlawb/node/commit/a49e04ddcd23f3b666421486ae60c3c4329e2242)), closes [#265](https://github.com/Gitlawb/node/issues/265) [#266](https://github.com/Gitlawb/node/issues/266)
-* **db:** pin profile DID matching with lone non-key fixture and boundary matrix ([b4016dc](https://github.com/Gitlawb/node/commit/b4016dce9d1fe450603173f9d92bd880127620fa)), closes [#298](https://github.com/Gitlawb/node/issues/298)
-* **node:** add a shared owner/name slug validator ([#272](https://github.com/Gitlawb/node/issues/272)) ([51d82de](https://github.com/Gitlawb/node/commit/51d82ded57a776d9c7e2fe65895efdbf9e227dc8))
-* **node:** bound the slug owner by NAME_MAX, not the DID limit ([#272](https://github.com/Gitlawb/node/issues/272)) ([27b0495](https://github.com/Gitlawb/node/commit/27b0495a6f7edeecddbe62b0f075a7e3b1ef5053))
-* **node:** fail a sync row terminally when a non-directory holds the owner path ([2e664b7](https://github.com/Gitlawb/node/commit/2e664b7885036ee72ce175af95bb567b18222583))
-* **node:** make a stalled sync queue visible, and fix two test claims ([9aa4c17](https://github.com/Gitlawb/node/commit/9aa4c17aefd1885d134b1334b54551366cdd8ef0))
-* **node:** move the attempted_at migration to v17, clear of two open branches ([fbc5868](https://github.com/Gitlawb/node/commit/fbc5868bd16946ea184dad26abcdb0e89d68e99a))
-* **node:** prove the mirror path is inside repos_dir before running git ([#272](https://github.com/Gitlawb/node/issues/272)) ([3ca9473](https://github.com/Gitlawb/node/commit/3ca94739ed7b90e4c265a0947025dc3b50e7583b))
-* **node:** reject a malformed slug at the notify boundary ([#272](https://github.com/Gitlawb/node/issues/272)) ([c587a40](https://github.com/Gitlawb/node/commit/c587a40a508acea9f74c3468d42ff037397c75d0))
-* **node:** reset a peer's reachability gate when an announce changes its http_url ([#270](https://github.com/Gitlawb/node/issues/270)) ([1964f88](https://github.com/Gitlawb/node/commit/1964f88378fe6942ffca5836de3a092b5b8f67b4))
-* **node:** reset a peer's reachability gate when an announce changes its http_url ([#270](https://github.com/Gitlawb/node/issues/270)) ([e2560ee](https://github.com/Gitlawb/node/commit/e2560ee46c28e4e966d87867a86dc4bad4ca7a3a))
-* **node:** stamp every dequeued sync row so a stuck one yields the batch ([6e64d0d](https://github.com/Gitlawb/node/commit/6e64d0d0a7fdd38ef3a56820c1808dd7c9bbd658))
-* **node:** stop a peer-supplied repo slug from escaping repos_dir ([#272](https://github.com/Gitlawb/node/issues/272)) ([c70b1df](https://github.com/Gitlawb/node/commit/c70b1df42d5c3f70ec4c272886c85fcc16ed92d1))
-* **node:** validate the queued slug before the worker derives a path ([#272](https://github.com/Gitlawb/node/issues/272)) ([9f00e24](https://github.com/Gitlawb/node/commit/9f00e24aec217e4b8256dd437cff4741b1478b96))
-* **review:** correct two comment claims and guard two more wrong shapes ([#270](https://github.com/Gitlawb/node/issues/270)) ([811ad50](https://github.com/Gitlawb/node/commit/811ad504bc268c8e9cddac93822adcd00bfe9ff8))
-* **review:** stop the dequeue overclaiming, and pin the starvation premise ([457be11](https://github.com/Gitlawb/node/commit/457be11d728a2a34775cf96be2bc5af9a58cb250))
-* strict DID matching in get_profile ([acb4714](https://github.com/Gitlawb/node/commit/acb47145beec8ce7b20b7a2479f7a55ed68e2a89))
-* strict DID matching in get_profile ([13b565b](https://github.com/Gitlawb/node/commit/13b565bcac09f075a7fd555a6e9acdd27de6c7bb))
-* target stored did on profile update aliases ([9836c3a](https://github.com/Gitlawb/node/commit/9836c3a21bc96a2ce95cb1fd12bd84b22142787d))
-
-## [0.7.0](https://github.com/Gitlawb/node/compare/v0.6.0...v0.7.0) (2026-07-22)
-
-
-### Features
-
-* **gl:** actually verify certificate signatures in gl cert show ([b0a2ed8](https://github.com/Gitlawb/node/commit/b0a2ed830821b5a77b0dafdcf5ce4b1ad904ec25))
-* **gl:** doctor warns when a shell alias shadows the gl binary ([0765c42](https://github.com/Gitlawb/node/commit/0765c42feff47b29054b9a277881a7cd5bfd9135))
-
-
-### Bug Fixes
-
-* **gl:** address review findings — non-fatal DID check, exact-host loopback, detached HEAD, multi-URL remotes, command-aware unalias ([3fa38dc](https://github.com/Gitlawb/node/commit/3fa38dce4e435ef017f607d355266c2892161f8d))
-* **gl:** anchor cert --verify to a trusted issuer; pin canonical payload form ([c681af7](https://github.com/Gitlawb/node/commit/c681af70dba3a76b11795d324f2ca0fecf7e42fd))
-* **gl:** detect gitlawb remotes beyond origin in gl status ([2d88937](https://github.com/Gitlawb/node/commit/2d88937fbabf21013f28debb4d4f93270a4bb8cd))
-* **gl:** doctor treats a reachable local node as configuration, not failure ([94c61f7](https://github.com/Gitlawb/node/commit/94c61f79421fa26ecc1bdf7c255e0c48e66a2048))
-* **gl:** make gl init's push guidance match the repo's actual state ([cb42a48](https://github.com/Gitlawb/node/commit/cb42a4830d0d070165e2212f316d009ec716dbee))
-* warn about the oh-my-zsh gl alias at install time ([e3df51d](https://github.com/Gitlawb/node/commit/e3df51d76c0275fde33954ff0af0aab95fd0cb7d))
-
 ## [0.6.0](https://github.com/Gitlawb/node/compare/v0.5.1...v0.6.0) (2026-07-22)
 
 

Cargo.lock

@@ -770,23 +770,6 @@ dependencies = [
  "zeroize",
 ]
 
-[[package]]
-name = "ark-ff"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f7a806ac6c8307b929df4645776290a50ee2aac754ad09d8bdf73391309e43af"
-dependencies = [
- "ark-ff-asm 0.6.0",
- "ark-ff-macros 0.6.0",
- "ark-serialize 0.6.0",
- "ark-std 0.6.0",
- "digest 0.10.7",
- "educe",
- "num-bigint",
- "num-traits",
- "zeroize",
-]
-
 [[package]]
 name = "ark-ff-asm"
 version = "0.3.0"
@@ -817,16 +800,6 @@ dependencies = [
  "syn 2.0.117",
 ]
 
-[[package]]
-name = "ark-ff-asm"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1479009684adc073dff49a1025d3a7065b317a9ead25aaaca38cdc70058ba8a2"
-dependencies = [
- "quote",
- "syn 2.0.117",
-]
-
 [[package]]
 name = "ark-ff-macros"
 version = "0.3.0"
@@ -865,19 +838,6 @@ dependencies = [
  "syn 2.0.117",
 ]
 
-[[package]]
-name = "ark-ff-macros"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4a0691ed21ef00ef89c1e9bda832eba493dda3ec2f8d892fb25b705f73f06bb8"
-dependencies = [
- "num-bigint",
- "num-traits",
- "proc-macro2",
- "quote",
- "syn 2.0.117",
-]
-
 [[package]]
 name = "ark-serialize"
 version = "0.3.0"
@@ -911,30 +871,6 @@ dependencies = [
  "num-bigint",
 ]
 
-[[package]]
-name = "ark-serialize"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a74dd304fd536fb95d0a328e72be759209cc496a9da094c5bc56e5fea4f9e86b"
-dependencies = [
- "ark-serialize-derive",
- "ark-std 0.6.0",
- "digest 0.10.7",
- "num-bigint",
- "serde_with",
-]
-
-[[package]]
-name = "ark-serialize-derive"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4f153690697a2b91e5e1251ff98411ee5371500a111a0fd317a70e588eb300f9"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn 2.0.117",
-]
-
 [[package]]
 name = "ark-std"
 version = "0.3.0"
@@ -965,16 +901,6 @@ dependencies = [
  "rand 0.8.6",
 ]
 
-[[package]]
-name = "ark-std"
-version = "0.6.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "367c9c827ed431bff6868b7aa926e05b16eb46603cc8b6e768e4a5553fa1d155"
-dependencies = [
- "num-traits",
- "rand 0.8.6",
-]
-
 [[package]]
 name = "arrayref"
 version = "0.3.9"
@@ -2630,7 +2556,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7ab67060fc6b8ef687992d439ca0fa36e7ed17e9a0b16b25b601e8757df720de"
 dependencies = [
  "data-encoding",
- "syn 2.0.117",
+ "syn 1.0.109",
 ]
 
 [[package]]
@@ -3002,7 +2928,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "39cab71617ae0d63f51a36d69f866391735b51691dbda63cf6f96d042b63efeb"
 dependencies = [
  "libc",
- "windows-sys 0.61.2",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -3374,7 +3300,7 @@ dependencies = [
 
 [[package]]
 name = "git-remote-gitlawb"
-version = "0.7.1"
+version = "0.6.0"
 dependencies = [
  "anyhow",
  "gitlawb-core",
@@ -3386,7 +3312,7 @@ dependencies = [
 
 [[package]]
 name = "gitlawb-attest"
-version = "0.7.1"
+version = "0.6.0"
 dependencies = [
  "base64",
  "ed25519-dalek",
@@ -3403,7 +3329,7 @@ dependencies = [
 
 [[package]]
 name = "gitlawb-core"
-version = "0.7.1"
+version = "0.6.0"
 dependencies = [
  "anyhow",
  "base64",
@@ -3430,13 +3356,14 @@ dependencies = [
 
 [[package]]
 name = "gitlawb-node"
-version = "0.7.1"
+version = "0.6.0"
 dependencies = [
  "alloy",
  "anyhow",
  "async-compression",
  "async-graphql",
  "async-graphql-axum",
+ "async-trait",
  "aws-config",
  "aws-sdk-s3",
  "axum",
@@ -3486,7 +3413,7 @@ dependencies = [
 
 [[package]]
 name = "gl"
-version = "0.7.1"
+version = "0.6.0"
 dependencies = [
  "alloy",
  "anyhow",
@@ -5815,15 +5742,14 @@ dependencies = [
 
 [[package]]
 name = "ruint"
-version = "1.20.0"
+version = "1.17.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f5e99bff0393163bb25029a6af25d3d8d202ba5b5438a74d1bd8789f5c822970"
+checksum = "c141e807189ad38a07276942c6623032d3753c8859c146104ac2e4d68865945a"
 dependencies = [
  "alloy-rlp",
  "ark-ff 0.3.0",
  "ark-ff 0.4.2",
  "ark-ff 0.5.0",
- "ark-ff 0.6.0",
  "bytes",
  "fastrlp 0.3.1",
  "fastrlp 0.4.0",
@@ -5897,7 +5823,7 @@ dependencies = [
  "errno",
  "libc",
  "linux-raw-sys",
- "windows-sys 0.61.2",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]
@@ -6791,7 +6717,7 @@ dependencies = [
  "getrandom 0.4.2",
  "once_cell",
  "rustix",
- "windows-sys 0.61.2",
+ "windows-sys 0.52.0",
 ]
 
 [[package]]

Dockerfile

@@ -8,33 +8,20 @@ FROM rust:1.91-bookworm AS builder
 
 WORKDIR /build
 
-# Cache dependencies first for faster rebuilds.
-# EVERY workspace member's manifest must be copied here, not just the ones this
-# image ships: cargo loads the whole workspace before it resolves anything, so a
-# missing member's Cargo.toml (or a path dependency pointing at one) aborts the
-# layer before a single crate is fetched.
+# Cache dependencies first for faster rebuilds
 COPY Cargo.toml Cargo.lock ./
 COPY crates/gitlawb-core/Cargo.toml crates/gitlawb-core/
 COPY crates/gitlawb-node/Cargo.toml crates/gitlawb-node/
 COPY crates/gl/Cargo.toml crates/gl/
 COPY crates/git-remote-gitlawb/Cargo.toml crates/git-remote-gitlawb/
-COPY crates/gitlawb-attest/Cargo.toml crates/gitlawb-attest/
-COPY crates/icaptcha-client/Cargo.toml crates/icaptcha-client/
 
-# Fetch deps (this layer caches until Cargo.{toml,lock} change).
-# No `|| true`: this layer is expected to succeed, and swallowing its exit code
-# is what let it sit dead from the moment gl gained a path dependency on
-# icaptcha-client. A failure here should stop the build, not silently turn every
-# image build into a cold one.
-RUN mkdir -p crates/gitlawb-core/src crates/gitlawb-node/src crates/gl/src \
-        crates/git-remote-gitlawb/src crates/gitlawb-attest/src crates/icaptcha-client/src && \
+# Fetch deps (this layer caches until Cargo.{toml,lock} change)
+RUN mkdir -p crates/gitlawb-core/src crates/gitlawb-node/src crates/gl/src crates/git-remote-gitlawb/src && \
     echo 'fn main() {}' > crates/gitlawb-node/src/main.rs && \
     echo 'fn main() {}' > crates/gl/src/main.rs && \
     echo 'fn main() {}' > crates/git-remote-gitlawb/src/main.rs && \
     echo '' > crates/gitlawb-core/src/lib.rs && \
-    echo '' > crates/gitlawb-attest/src/lib.rs && \
-    echo '' > crates/icaptcha-client/src/lib.rs && \
-    cargo build --release --locked -p gitlawb-node -p gl -p git-remote-gitlawb
+    cargo build --release -p gitlawb-node -p gl -p git-remote-gitlawb || true
 
 # Now copy real sources and build for real.
 # Force-bump mtimes so cargo's fingerprint check rebuilds — without this,
@@ -47,7 +34,7 @@ RUN find crates -name "*.rs" -exec touch {} + && \
     rm -rf target/release/.fingerprint/gitlawb-node-* \
            target/release/.fingerprint/gl-* \
            target/release/.fingerprint/git-remote-gitlawb-* && \
-    cargo build --release --locked -p gitlawb-node -p gl -p git-remote-gitlawb && \
+    cargo build --release -p gitlawb-node -p gl -p git-remote-gitlawb && \
     strip target/release/gitlawb-node target/release/gl target/release/git-remote-gitlawb
 
 # ── Runtime stage ───────────────────────────────────────────────────────────

Dockerfile.bins

@@ -3,10 +3,7 @@
 # Used by scripts/build-bins.sh — not used for node deployment.
 ARG TARGET=x86_64-unknown-linux-musl
 
-# Must stay at or above the workspace `rust-version` in Cargo.toml. An unmet
-# MSRV is a hard cargo error, and nothing in CI builds this file, so a stale
-# base image here fails only when someone runs scripts/build-bins.sh.
-FROM rust:1.91-bookworm AS builder
+FROM rust:1.85-bookworm AS builder
 ARG TARGET
 
 WORKDIR /build
@@ -26,7 +23,7 @@ COPY crates/ crates/
 # Configure cross-linker for aarch64
 ENV CARGO_TARGET_AARCH64_UNKNOWN_LINUX_MUSL_LINKER=aarch64-linux-gnu-gcc
 
-RUN cargo build --release --locked --target "$TARGET" -p gl -p git-remote-gitlawb
+RUN cargo build --release --target "$TARGET" -p gl -p git-remote-gitlawb
 
 FROM debian:bookworm-slim
 ARG TARGET

ci/gitlawb-core-allowed-deps.txt

@@ -1,112 +0,0 @@
-# Dependency allowlist for the `gitlawb-core` crate.
-#
-# gitlawb-core is the shared-primitives crate every consumer embeds (the `gl`
-# CLI, git-remote-gitlawb, the node daemon). It must stay lean so those
-# consumers do not inherit daemon-weight dependencies. This file is the source
-# of truth for the crates gitlawb-core's NORMAL (non-dev, non-build) transitive
-# dependency closure is allowed to contain. CI (`scripts/check-gitlawb-core-deps.sh`,
-# wired into the `core-deps-purity` job in .github/workflows/pr-checks.yml)
-# fails if gitlawb-core's tree contains any crate not listed here.
-#
-# The allowlist is exhaustive-by-construction: a NEW dependency reds CI whether
-# or not anyone thought to ban it, which a denylist cannot do.
-#
-# Measured in-workspace, so feature unification across the workspace can, in
-# rare cases, surface a crate here that gitlawb-core would not pull standalone.
-# An unexpected new entry is a prompt to check what pulled it in, not noise to
-# rubber-stamp.
-#
-# Regenerate (from repo root, after an INTENTIONAL dependency change):
-#   cargo tree -p gitlawb-core --edges normal --prefix none \
-#     | sed -E 's/ v[0-9].*$//' \
-#     | grep -v '^gitlawb-core$' \
-#     | sort -u
-# then paste the result below this header block. (The checker snapshots and
-# restores Cargo.lock; a manual regen may refresh it — `git checkout Cargo.lock`
-# afterward if you did not intend that.)
-#
-# Lines starting with `#` and blank lines are ignored by the checker.
-aead
-anyhow
-base-x
-base256emoji
-base64
-base64ct
-block-buffer
-cfg-if
-chacha20
-chacha20poly1305
-chrono
-cid
-cipher
-const-oid
-const-str
-core2
-cpufeatures
-crypto-common
-crypto_box
-crypto_secretbox
-curve25519-dalek
-curve25519-dalek-derive
-data-encoding
-data-encoding-macro
-data-encoding-macro-internal
-der
-digest
-ed25519
-ed25519-dalek
-equivalent
-generic-array
-getrandom
-hashbrown
-hex
-iana-time-zone
-indexmap
-inout
-itoa
-libc
-match-lookup
-memchr
-multibase
-multihash
-multihash-codetable
-multihash-derive
-multihash-derive-impl
-num-traits
-opaque-debug
-pem-rfc7468
-pkcs8
-poly1305
-ppv-lite86
-proc-macro-crate
-proc-macro2
-quote
-rand
-rand_chacha
-rand_core
-salsa20
-serde
-serde_core
-serde_derive
-serde_json
-sha2
-signature
-spki
-subtle
-syn
-synstructure
-thiserror
-thiserror-impl
-toml_datetime
-toml_edit
-toml_parser
-typenum
-unicode-ident
-universal-hash
-unsigned-varint
-uuid
-winnow
-zerocopy
-zeroize
-zeroize_derive
-zmij

crates/git-remote-gitlawb/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "git-remote-gitlawb"
 description = "Git remote helper — enables 'git clone gitlawb://did:gitlawb:...'"
-version = "0.7.1" # x-release-please-version
+version = "0.6.0" # x-release-please-version
 edition.workspace = true
 rust-version.workspace = true
 license.workspace = true

crates/gitlawb-attest/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "gitlawb-attest"
 description = "External Attestation v1: pluggable provenance attachments for gitlawb ref-update certs"
-version = "0.7.1" # x-release-please-version
+version = "0.6.0" # x-release-please-version
 edition.workspace = true
 rust-version.workspace = true
 license.workspace = true

crates/gitlawb-core/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "gitlawb-core"
 description = "Core cryptographic primitives for the gitlawb network: DIDs, CIDs, UCAN, HTTP Signatures, ref-update certificates"
-version = "0.7.1" # x-release-please-version
+version = "0.6.0" # x-release-please-version
 edition.workspace = true
 rust-version.workspace = true
 license.workspace = true

crates/gitlawb-node/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "gitlawb-node"
 description = "The gitlawb node daemon — git hosting over HTTP with DID auth"
-version = "0.7.1" # x-release-please-version
+version = "0.6.0" # x-release-please-version
 edition.workspace = true
 rust-version.workspace = true
 license.workspace = true
@@ -12,6 +12,7 @@ path = "src/main.rs"
 
 [dependencies]
 gitlawb-core = { path = "../gitlawb-core" }
+async-trait = "0.1"
 ed25519-dalek = { workspace = true }
 base64 = { workspace = true }
 tokio = { workspace = true }

crates/gitlawb-node/src/api/issues.rs

@@ -70,11 +70,37 @@ pub async fn create_issue(
 
     let create_result = git_issues::create_issue(&disk_path, &issue_id, &json_str);
 
-    // Always release the advisory lock — even on error; upload to Tigris only on success.
-    guard.release(create_result.is_ok()).await;
+    // Always release the advisory lock, even on error; upload to Tigris only on
+    // success. A false return means the durable upload failed/timed out: the
+    // issue ref applied to local disk but never reached object storage, so the
+    // next acquire_write would revert it from the stale archive. Report 5xx so
+    // the client retries rather than trusting an issue that never landed durably,
+    // and skip the trust bump below. (Same P1 data-loss guard as receive-pack.)
+    // On that failure, roll back the local ref BEFORE the lock releases — an
+    // orphan ref would make the retry a duplicate, and in an unlock-to-delete
+    // window a concurrent same-repo write could upload an archive still
+    // carrying it. Best-effort: a failed delete logs and the 5xx still returns.
+    let upload_ok = guard
+        .release_with_failure_cleanup(create_result.is_ok(), |path| {
+            if let Err(e) = git_issues::delete_issue_ref(path, &issue_id) {
+                tracing::warn!(issue = %issue_id, err = %e,
+                    "failed to roll back local issue ref after failed durable upload; \
+                     a retry may duplicate this issue");
+            }
+        })
+        .await;
 
     create_result.map_err(|e| AppError::Git(e.to_string()))?;
 
+    if !upload_ok {
+        tracing::error!(repo = %record.name,
+            "durable upload failed after issue create; reporting failure so the client retries");
+        return Err(AppError::Git(format!(
+            "durable storage upload failed for {}",
+            record.name
+        )));
+    }
+
     // Bump trust score for the issue author — increment current score by 0.05
     // (avoids the push_count=0 stuck-at-0.05 bug for agents who only file issues)
     if let Some(ref author_did) = issue.author {
@@ -244,11 +270,14 @@ pub async fn close_issue(
             .ok()
             .and_then(|i| i.author),
         Ok(None) => {
-            guard.release(false).await;
+            // release(false): no upload attempted, so the bool is always true
+            // (nothing to fail on). Ignore it on this pre-write error path.
+            let _ = guard.release(false).await;
             return Err(AppError::NotFound(format!("issue {issue_id} not found")));
         }
         Err(e) => {
-            guard.release(false).await;
+            // release(false): no upload attempted, nothing to fail on.
+            let _ = guard.release(false).await;
             return Err(AppError::Git(e.to_string()));
         }
     };
@@ -257,21 +286,57 @@ pub async fn close_issue(
         .as_deref()
         .is_some_and(|a| crate::api::did_matches(&auth.0, a));
     if !is_owner && !is_author {
-        guard.release(false).await;
+        // release(false): no upload attempted, nothing to fail on.
+        let _ = guard.release(false).await;
         return Err(AppError::Forbidden(
             "only the repo owner or the issue author can close this issue".into(),
         ));
     }
 
+    // Snapshot the issue ref's pre-close OID so a failed durable upload can roll
+    // the close back on local disk, mirroring create_issue. Best-effort: if the
+    // snapshot read fails we proceed without a rollback closure (the close still
+    // applies and a failed upload still surfaces as a 5xx below).
+    let prior_ref = git_issues::issue_ref_oid(&disk_path, &issue_id)
+        .ok()
+        .flatten();
+
     let close_result = git_issues::close_issue(&disk_path, &issue_id);
 
-    // Always release the advisory lock — even on error; upload to Tigris only on success.
-    guard.release(close_result.is_ok()).await;
+    // Always release the advisory lock, even on error; upload to Tigris only on
+    // success. A false return means the durable upload failed: the close applied
+    // to local disk but never reached object storage, so a later acquire_write
+    // reverts it. Report 5xx so the client retries. (P1 data-loss guard.)
+    // On that failure, roll the local close BACK to the captured open ref BEFORE
+    // the lock releases — otherwise the local fast path in RepoStore::acquire
+    // serves the "closed" state until a later write refreshes the archive, even
+    // though the client got a 5xx. The cleanup runs only when an upload was
+    // attempted and failed, while the advisory lock is still held.
+    let upload_ok = guard
+        .release_with_failure_cleanup(close_result.is_ok(), |path| {
+            if let Some((full_id, oid)) = &prior_ref {
+                if let Err(e) = git_issues::restore_issue_ref(path, full_id, oid) {
+                    tracing::warn!(issue = %issue_id, err = %e,
+                        "failed to roll back local issue close after failed durable upload; \
+                         a local-fast-path read may serve a stale closed state");
+                }
+            }
+        })
+        .await;
 
     let updated = close_result
         .map_err(|e| AppError::Git(e.to_string()))?
         .ok_or_else(|| AppError::RepoNotFound(format!("issue {issue_id} not found")))?;
 
+    if !upload_ok {
+        tracing::error!(repo = %repo,
+            "durable upload failed after issue close; reporting failure so the client retries");
+        return Err(AppError::Git(format!(
+            "durable storage upload failed for {}",
+            record.name
+        )));
+    }
+
     let issue: serde_json::Value = serde_json::from_str(&updated)
         .map_err(|e| AppError::BadRequest(format!("invalid issue data: {e}")))?;
 
@@ -279,3 +344,421 @@ pub async fn close_issue(
 
     Ok(Json(issue))
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use axum::extract::{Path as AxPath, State};
+    use axum::Extension;
+    use std::path::Path as StdPath;
+    use std::time::Duration;
+
+    // Object store whose upload() parks forever, so release()'s bounded timeout
+    // is the only thing that completes it, modeling a stalled durable PUT. exists()
+    // is false so acquire_write never downloads over the fresh local bare repo.
+    struct StallStore;
+    #[async_trait::async_trait]
+    impl crate::git::tigris::ObjectStore for StallStore {
+        async fn exists(&self, _o: &str, _r: &str) -> anyhow::Result<bool> {
+            Ok(false)
+        }
+        async fn upload(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            std::future::pending::<()>().await;
+            Ok(())
+        }
+        async fn download(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            Ok(())
+        }
+        async fn delete(&self, _o: &str, _r: &str) -> anyhow::Result<()> {
+            Ok(())
+        }
+    }
+
+    // Object store whose upload succeeds immediately. exists() is false so
+    // acquire_write never performs a reverting refresh over the local dir —
+    // the retry test below pins that contract explicitly.
+    struct OkStore;
+    #[async_trait::async_trait]
+    impl crate::git::tigris::ObjectStore for OkStore {
+        async fn exists(&self, _o: &str, _r: &str) -> anyhow::Result<bool> {
+            Ok(false)
+        }
+        async fn upload(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            Ok(())
+        }
+        async fn download(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            Ok(())
+        }
+        async fn delete(&self, _o: &str, _r: &str) -> anyhow::Result<()> {
+            Ok(())
+        }
+    }
+
+    // Shared setup for the failed-upload rollback tests: a state whose repo
+    // store stalls uploads (200ms release timeout), a public repo row for
+    // `owner/name`, and an initialized bare repo at the returned path.
+    async fn stall_state_with_repo(
+        pool: sqlx::PgPool,
+        repos_dir: &tempfile::TempDir,
+        owner: &str,
+        name: &str,
+    ) -> (crate::state::AppState, std::path::PathBuf) {
+        let mut state = crate::test_support::test_state(pool.clone()).await;
+        state.repo_store = crate::git::repo_store::RepoStore::new(
+            repos_dir.path().to_path_buf(),
+            Some(std::sync::Arc::new(StallStore)),
+            pool.clone(),
+        )
+        .with_release_upload_timeout(Duration::from_millis(200));
+
+        let owner_slug = owner.replace([':', '/'], "_");
+        let bare = repos_dir
+            .path()
+            .join(&owner_slug)
+            .join(format!("{name}.git"));
+
+        state
+            .db
+            .upsert_mirror_repo(owner, name, &bare.to_string_lossy(), None, false)
+            .await
+            .unwrap();
+
+        std::fs::create_dir_all(&bare).unwrap();
+        let out = std::process::Command::new("git")
+            .args(["init", "--bare", "-q", &bare.to_string_lossy()])
+            .output()
+            .unwrap();
+        assert!(out.status.success(), "git init --bare failed");
+
+        (state, bare)
+    }
+
+    async fn create_issue_status(
+        state: &crate::state::AppState,
+        owner: &str,
+        name: &str,
+    ) -> axum::http::StatusCode {
+        use axum::response::IntoResponse;
+        let resp = super::create_issue(
+            State(state.clone()),
+            Extension(crate::auth::AuthenticatedDid(owner.to_string())),
+            AxPath((owner.to_string(), name.to_string())),
+            axum::Json(CreateIssueRequest {
+                title: "t".into(),
+                body: None,
+                signed_payload: None,
+            }),
+        )
+        .await;
+        match resp {
+            Ok((code, _)) => code,
+            Err(e) => e.into_response().status(),
+        }
+    }
+
+    async fn close_issue_status(
+        state: &crate::state::AppState,
+        owner: &str,
+        name: &str,
+        issue_id: &str,
+    ) -> axum::http::StatusCode {
+        use axum::response::IntoResponse;
+        let resp = super::close_issue(
+            State(state.clone()),
+            Extension(crate::auth::AuthenticatedDid(owner.to_string())),
+            AxPath((owner.to_string(), name.to_string(), issue_id.to_string())),
+        )
+        .await;
+        match resp {
+            Ok(_) => axum::http::StatusCode::OK,
+            Err(e) => e.into_response().status(),
+        }
+    }
+
+    // Seed an OPEN issue directly on local disk, bypassing the (stalling) upload
+    // path so the close test starts from a durably-consistent open issue.
+    fn seed_open_issue(bare: &StdPath, owner: &str, issue_id: &str) {
+        let issue = IssueRecord {
+            id: issue_id.to_string(),
+            title: "t".into(),
+            body: None,
+            author: Some(owner.to_string()),
+            created_at: Utc::now().to_rfc3339(),
+            status: "open".to_string(),
+            signed_payload: None,
+        };
+        let json = serde_json::to_string(&issue).unwrap();
+        git_issues::create_issue(bare, issue_id, &json).unwrap();
+    }
+
+    fn issue_status(bare: &StdPath, issue_id: &str) -> String {
+        let raw = git_issues::get_issue(bare, issue_id).unwrap().unwrap();
+        let v: serde_json::Value = serde_json::from_str(&raw).unwrap();
+        v["status"].as_str().unwrap().to_string()
+    }
+
+    // U2 [F2, P1] — the failed-upload rollback on the close path. close_issue
+    // applies the "closed" ref to LOCAL disk, then release() uploads to durable
+    // storage. When that upload times out the local close is NOT persisted, so a
+    // later acquire_write reverts it from the stale archive — but until then, a
+    // local-fast-path read via RepoStore::acquire serves the stale "closed"
+    // state even though the client got a 5xx. The fix rolls the close back to the
+    // captured open ref while the advisory lock is held. RED with the cleanup
+    // closure reverted to `|_| {}`: the read sees "closed".
+    #[sqlx::test]
+    async fn issue_close_failed_upload_rolls_back_to_open(pool: sqlx::PgPool) {
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        let owner = "z6issuecloserollbackowner";
+        let name = "issuecloserollback";
+        let (state, bare) = stall_state_with_repo(pool, &repos_dir, owner, name).await;
+
+        let issue_id = "aaaa1111-0000-0000-0000-000000000000";
+        seed_open_issue(&bare, owner, issue_id);
+        assert_eq!(issue_status(&bare, issue_id), "open", "seed must be open");
+
+        // Close via the handler — the durable upload stalls and times out (5xx).
+        let status = close_issue_status(&state, owner, name, issue_id).await;
+        assert!(
+            status.is_server_error(),
+            "a stalled durable upload on issue-close must 500, got {status}"
+        );
+
+        // The local fast path must still see the issue OPEN: the failed upload
+        // rolled the close back while the lock was held.
+        assert_eq!(
+            issue_status(&bare, issue_id),
+            "open",
+            "a failed durable upload must roll the local close back to open; a \
+             local-fast-path read served the stale closed state"
+        );
+    }
+
+    // Success path: a working store closes the issue (2xx) and the read reflects
+    // the close. Confirms the rollback wiring does not fire on success.
+    #[sqlx::test]
+    async fn issue_close_success_persists_close(pool: sqlx::PgPool) {
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        let owner = "z6issueclosesuccessowner";
+        let name = "issueclosesuccess";
+        let (mut state, bare) = stall_state_with_repo(pool.clone(), &repos_dir, owner, name).await;
+        // Swap the stalling store for one whose upload succeeds immediately.
+        state.repo_store = crate::git::repo_store::RepoStore::new(
+            repos_dir.path().to_path_buf(),
+            Some(std::sync::Arc::new(OkStore)),
+            pool.clone(),
+        )
+        .with_release_upload_timeout(Duration::from_millis(200));
+
+        let issue_id = "bbbb2222-0000-0000-0000-000000000000";
+        seed_open_issue(&bare, owner, issue_id);
+
+        let status = close_issue_status(&state, owner, name, issue_id).await;
+        assert_eq!(status, StatusCode::OK, "working store must close (200)");
+        assert_eq!(
+            issue_status(&bare, issue_id),
+            "closed",
+            "a successful close must persist as closed"
+        );
+    }
+
+    // Pre-write error path: closing a nonexistent issue returns 404 and never
+    // attempts an upload (release(false)), so the stalling store cannot hang it.
+    #[sqlx::test]
+    async fn issue_close_not_found_releases_without_upload(pool: sqlx::PgPool) {
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        let owner = "z6issueclosemissingowner";
+        let name = "issueclosemissing";
+        let (state, _bare) = stall_state_with_repo(pool, &repos_dir, owner, name).await;
+
+        let status =
+            close_issue_status(&state, owner, name, "cccc3333-0000-0000-0000-000000000000").await;
+        assert_eq!(
+            status,
+            StatusCode::NOT_FOUND,
+            "closing a missing issue must 404 without attempting an upload"
+        );
+    }
+
+    fn issue_refs(bare: &StdPath) -> Vec<String> {
+        let out = std::process::Command::new("git")
+            .args([
+                "for-each-ref",
+                "--format=%(refname)",
+                "refs/gitlawb/issues/",
+            ])
+            .current_dir(bare)
+            .output()
+            .unwrap();
+        assert!(out.status.success(), "git for-each-ref failed");
+        String::from_utf8_lossy(&out.stdout)
+            .lines()
+            .filter(|l| !l.trim().is_empty())
+            .map(|l| l.trim().to_string())
+            .collect()
+    }
+
+    // Rollback half of the #196 round-four issue-create fix. When the durable
+    // upload fails/times out, the handler 500s (covered below) — but the issue
+    // ref it wrote to LOCAL disk must also be rolled back, while the advisory
+    // lock is still held. An orphan ref makes the 500 dishonest: a retry
+    // re-creates the issue under a new uuid while the orphan is still listed,
+    // and a concurrent same-repo write can upload an archive carrying it.
+    // RED pre-fix: the orphan ref survives the failed create.
+    #[sqlx::test]
+    async fn issue_create_failed_upload_rolls_back_local_ref(pool: sqlx::PgPool) {
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        let owner = "z6issuerollbackowner";
+        let name = "issuerollback";
+        let (state, bare) = stall_state_with_repo(pool, &repos_dir, owner, name).await;
+
+        let status = create_issue_status(&state, owner, name).await;
+        assert!(
+            status.is_server_error(),
+            "stalled upload must 500, got {status}"
+        );
+
+        let refs = issue_refs(&bare);
+        assert!(
+            refs.is_empty(),
+            "a failed durable upload must roll back the local issue ref while \
+             the advisory lock is held; orphan ref(s) survived: {refs:?}"
+        );
+    }
+
+    // Retry half: after the failed-upload 500, a retry with a WORKING store
+    // must yield exactly one issue. Pre-fix the orphan ref from the failed
+    // attempt makes the retry a duplicate (two issues in list_issues).
+    #[sqlx::test]
+    async fn issue_create_failed_upload_retry_does_not_duplicate(pool: sqlx::PgPool) {
+        use crate::git::tigris::ObjectStore as _;
+
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        let owner = "z6issueretryowner";
+        let name = "issueretry";
+        let (state, bare) = stall_state_with_repo(pool.clone(), &repos_dir, owner, name).await;
+
+        let status = create_issue_status(&state, owner, name).await;
+        assert!(
+            status.is_server_error(),
+            "stalled upload must 500, got {status}"
+        );
+
+        // PIN the double's contract: exists() stays false after the failed
+        // upload, so the retry's acquire_write performs no reverting refresh
+        // from a stale archive. Without this pin, a download could revert the
+        // orphan ref and the assertion below would pass without the rollback
+        // fix — it must rest on the rollback, not a stale-archive revert.
+        let owner_slug = owner.replace([':', '/'], "_");
+        assert!(
+            !OkStore.exists(&owner_slug, name).await.unwrap(),
+            "test double contract broken: exists() must stay false"
+        );
+
+        // Retry with a store whose upload succeeds.
+        let mut retry_state = state.clone();
+        retry_state.repo_store = crate::git::repo_store::RepoStore::new(
+            repos_dir.path().to_path_buf(),
+            Some(std::sync::Arc::new(OkStore)),
+            pool.clone(),
+        )
+        .with_release_upload_timeout(Duration::from_millis(200));
+
+        let status = create_issue_status(&retry_state, owner, name).await;
+        assert_eq!(
+            status,
+            StatusCode::CREATED,
+            "retry with working store must 201"
+        );
+
+        let issues = crate::git::issues::list_issues(&bare).unwrap();
+        assert_eq!(
+            issues.len(),
+            1,
+            "retry after a failed-upload 500 must yield exactly ONE issue; the \
+             failed attempt's orphan ref duplicated it (got {} issues)",
+            issues.len()
+        );
+    }
+
+    // P1-class data-loss regression, issue-create path. create_issue writes the
+    // issue ref to local disk, then release(true) uploads the repo to durable
+    // storage. When that upload times out the local write is NOT persisted, so
+    // the next acquire_write reverts it from the stale archive. The pre-fix
+    // handler ignored release()'s bool and still returned 201 AND bumped the
+    // author's trust score, so the client trusted an issue that never landed
+    // durably. The fix surfaces a failed/timed-out upload as a 5xx and skips the
+    // success tail (the trust bump). Stall the upload past a tiny release timeout
+    // and assert (a) the handler returns 5xx and (b) the trust bump did not run.
+    // RED pre-fix: 201 + trust score bumped to 0.05.
+    #[sqlx::test]
+    async fn issue_create_durable_upload_timeout_fails_and_skips_side_effects(pool: sqlx::PgPool) {
+        use axum::response::IntoResponse;
+
+        let mut state = crate::test_support::test_state(pool.clone()).await;
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        state.repo_store = crate::git::repo_store::RepoStore::new(
+            repos_dir.path().to_path_buf(),
+            Some(std::sync::Arc::new(StallStore)),
+            pool.clone(),
+        )
+        .with_release_upload_timeout(Duration::from_millis(200));
+
+        let owner = "z6issueownerdurablefail";
+        let name = "issuedurablefail";
+        let owner_slug = owner.replace([':', '/'], "_");
+        let bare = repos_dir
+            .path()
+            .join(&owner_slug)
+            .join(format!("{name}.git"));
+
+        // Public mirror repo so the owner clears authorize_repo_read, and a real
+        // agent row so the post-write trust bump is an observable durable side
+        // effect (register_agent seeds trust_score = 0.0).
+        state
+            .db
+            .upsert_mirror_repo(owner, name, &bare.to_string_lossy(), None, false)
+            .await
+            .unwrap();
+        state.db.register_agent(owner, &[]).await.unwrap();
+
+        std::fs::create_dir_all(&bare).unwrap();
+        let out = std::process::Command::new("git")
+            .args(["init", "--bare", "-q", &bare.to_string_lossy()])
+            .output()
+            .unwrap();
+        assert!(out.status.success(), "git init --bare failed");
+
+        // acquire_write → git_issues::create_issue (writes the ref) → release
+        // (upload stalls, times out after 200ms). The client MUST see a failure.
+        let resp = super::create_issue(
+            State(state.clone()),
+            Extension(crate::auth::AuthenticatedDid(owner.to_string())),
+            AxPath((owner.to_string(), name.to_string())),
+            axum::Json(CreateIssueRequest {
+                title: "t".into(),
+                body: None,
+                signed_payload: None,
+            }),
+        )
+        .await;
+        let status = match resp {
+            Ok((code, _)) => code,
+            Err(e) => e.into_response().status(),
+        };
+        assert!(
+            status.is_server_error(),
+            "a timed-out durable upload on issue-create must fail (5xx) so the \
+             client retries (got {status}), which the client trusts as a landed \
+             issue that a later acquire_write would silently revert"
+        );
+
+        // The success tail (trust bump) must NOT have run: the issue was never
+        // durably accepted. register_agent seeded 0.0; a bump would make it 0.05.
+        let score = state.db.get_trust_score(owner).await.unwrap();
+        assert_eq!(
+            score, 0.0,
+            "the trust bump must NOT run when the durable upload failed; the \
+             issue was not durably accepted (score bumped to {score})"
+        );
+    }
+}

crates/gitlawb-node/src/api/mod.rs

@@ -22,7 +22,6 @@ pub mod replicas;
 pub mod repos;
 pub mod resolve;
 pub mod stars;
-pub mod status;
 pub mod tasks;
 pub mod visibility;
 pub mod webhooks;
@@ -68,14 +67,17 @@ pub(crate) async fn authorize_repo_read(
 /// representation only within `did:key`; never let a bare id match across methods —
 /// `did:web` / `did:gitlawb` share the base58 space with `did:key`, so a
 /// trailing-segment compare would treat `did:key:X` and `did:gitlawb:X` as equal.
-///
-/// The collapse itself lives in exactly one place, [`crate::db::normalize_owner_key`],
-/// which is also the function the stored `owner_did` / `authorizing_did` columns
-/// are normalized through and which `OWNER_KEY_CASE_SQL` mirrors byte for byte.
-/// Two identities match when they normalize to the same key; a second copy of the
-/// rule here is how the Rust gate and the SQL filters would drift apart.
 pub(crate) fn did_matches(a: &str, b: &str) -> bool {
-    crate::db::normalize_owner_key(a) == crate::db::normalize_owner_key(b)
+    if a == b {
+        return true;
+    }
+    fn key_id(d: &str) -> &str {
+        d.strip_prefix("did:key:").unwrap_or(d)
+    }
+    let (ka, kb) = (key_id(a), key_id(b));
+    // After stripping `did:key:`, a value still containing ':' is a non-key full
+    // DID — do not let it match a bare `did:key` id.
+    !ka.contains(':') && !kb.contains(':') && ka == kb
 }
 
 /// 403 unless `caller` is the repo owner. Uses [`did_matches`] so the owner check
@@ -173,7 +175,6 @@ mod authz_guard {
         let events = include_str!("events.rs");
         let tasks = include_str!("tasks.rs");
         let stars = include_str!("stars.rs");
-        let status = include_str!("status/mod.rs");
         let protect = include_str!("protect.rs");
         let visibility = include_str!("visibility.rs");
         let profiles = include_str!("profiles.rs");
@@ -194,13 +195,6 @@ mod authz_guard {
             // (existence hiding) and require_repo_owner guards the owner half.
             (webhooks, "list_webhooks", "authorize_repo_read("),
             (webhooks, "list_webhooks", "require_repo_owner("),
-            // Same two-half shape as list_webhooks: authorize_repo_read runs
-            // first (a quarantined or unreadable repo gets the missing-repo
-            // not-found, never a 403 that would confirm existence), then
-            // require_repo_owner 403s a non-owner of a readable repo. Both halves
-            // are pinned, because dropping either changes what a stranger learns.
-            (status, "create_status", "authorize_repo_read("),
-            (status, "create_status", "require_repo_owner("),
             (labels, "add_label", "require_repo_owner("),
             (labels, "remove_label", "require_repo_owner("),
             // Bucket A' — owner OR author (did_matches against the author)
@@ -223,19 +217,6 @@ mod authz_guard {
             (protect, "list_protected_branches", "authorize_repo_read("),
             (labels, "list_labels", "authorize_repo_read("),
             (events, "list_repo_events", "authorize_repo_read("),
-            // The catch-up poll surface gates before any push-event row is read,
-            // so a stranger cannot use it to learn that a private repo exists or
-            // is being pushed to.
-            (events, "list_repo_push_events", "authorize_repo_read("),
-            // The status read gates before any claim data is loaded, so its deny
-            // is the repo's own not-found rather than an existence oracle over
-            // which commits were reported on.
-            (status, "commit_status", "authorize_repo_read("),
-            // The rollup gates on the same helper before the pull request row is
-            // loaded, so a caller who cannot read the repo cannot learn which
-            // pull request numbers exist on it — and the fallback's branch
-            // resolve and head persist sit behind that same gate.
-            (status, "pull_request_status", "authorize_repo_read("),
             // Bucket C — signer-self: the acting DID is matched/bound to auth.0
             (tasks, "create_task", "did_matches("),
             (tasks, "claim_task", "did_matches("),
@@ -509,7 +490,6 @@ mod authz_guard {
             (include_str!("replicas.rs"), "replicas.rs"),
             (include_str!("repos.rs"), "repos.rs"),
             (include_str!("stars.rs"), "stars.rs"),
-            (include_str!("status/mod.rs"), "status/mod.rs"),
             (include_str!("visibility.rs"), "visibility.rs"),
             (include_str!("webhooks.rs"), "webhooks.rs"),
         ];

crates/gitlawb-node/src/api/peers.rs

@@ -381,14 +381,6 @@ pub async fn notify_sync(
         )));
     }
 
-    // #272: the slug reaches a PathBuf::join in the sync worker, where an extra
-    // separator or a traversal segment places the mirror outside repos_dir. Reject
-    // it here, before the queue row and the ref-update row are written. The message
-    // names the field and nothing about where repos live on disk.
-    if let Err(e) = crate::git::repo_store::validate_repo_slug(&req.repo) {
-        return Err(AppError::BadRequest(format!("invalid repo field: {e}")));
-    }
-
     state
         .db
         .enqueue_sync(&req.repo, &req.node_did, &req.ref_name, &req.new_sha, None)
@@ -592,9 +584,7 @@ mod tests {
         state.config = Arc::new(cfg);
     }
 
-    // The slug must be a well-formed owner/name (#272); these brake tests are about
-    // the rate limiter, so the body carries a valid slug and reaches the brake.
-    const NOTIFY_BODY: &str = r#"{"repo":"z6Mkfoo/hello","ref_name":"refs/heads/main","new_sha":"0000000000000000000000000000000000000000","node_did":"PEER_DID"}"#;
+    const NOTIFY_BODY: &str = r#"{"repo":"demo","ref_name":"refs/heads/main","new_sha":"0000000000000000000000000000000000000000","node_did":"PEER_DID"}"#;
 
     // ── trigger: mandatory signature ──────────────────────────────────────────
 
@@ -876,265 +866,6 @@ mod tests {
         );
     }
 
-    // ── notify: repo slug validation (#272) ───────────────────────────────────
-    //
-    // An unsigned caller who has announced their own DID is a "known peer", so
-    // the slug they send reaches the sync worker's path join. `a//tmp/probe`
-    // resolved to /tmp/probe.git, outside repos_dir entirely. notify must reject
-    // anything that is not a well-formed owner/name slug before it is queued.
-
-    /// Every slug shape the boundary must refuse. `a//tmp/gitlawb-probe` is the
-    /// verified escape; the rest are the traversal and empty-half neighbours.
-    const MALFORMED_SLUGS: [&str; 9] = [
-        "a//tmp/gitlawb-probe",
-        "../hello",
-        "./hello",
-        "../../etc/evil",
-        "a/../../x",
-        "..",
-        "demo",
-        "/hello",
-        "z6Mkfoo/",
-    ];
-
-    fn notify_body(repo: &str, peer_did: &str) -> String {
-        serde_json::json!({
-            "repo": repo,
-            "ref_name": "refs/heads/main",
-            "new_sha": "0000000000000000000000000000000000000000",
-            "node_did": peer_did,
-        })
-        .to_string()
-    }
-
-    #[sqlx::test]
-    async fn sync_notify_rejects_malformed_repo_slugs(pool: PgPool) {
-        let state = test_state(pool).await;
-        let db = state.db.clone();
-        let peer_did = seed_peer(&state).await;
-        let router = crate::server::build_router(state);
-        for (i, slug) in MALFORMED_SLUGS.iter().enumerate() {
-            // Distinct source IPs so the per-IP brake never masks the 400.
-            let ip = format!("198.51.100.{}:5000", i + 1);
-            let resp = router
-                .clone()
-                .oneshot(unsigned_post(
-                    "/api/v1/sync/notify",
-                    &notify_body(slug, &peer_did),
-                    &ip,
-                ))
-                .await
-                .unwrap();
-            assert_eq!(
-                resp.status(),
-                StatusCode::BAD_REQUEST,
-                "slug {slug:?} must be rejected"
-            );
-        }
-        let queued = db.dequeue_pending_syncs(100).await.unwrap();
-        assert!(
-            queued.is_empty(),
-            "no malformed slug may be queued, got {:?}",
-            queued.iter().map(|q| &q.repo).collect::<Vec<_>>()
-        );
-    }
-
-    #[sqlx::test]
-    async fn sync_notify_rejection_body_names_field_not_path(pool: PgPool) {
-        // R3: the client learns which field was bad, never where repos live on
-        // disk. A message carrying the resolved path would hand an unsigned
-        // caller the server's layout.
-        let state = test_state(pool).await;
-        let repos_dir = state.config.repos_dir.display().to_string();
-        let peer_did = seed_peer(&state).await;
-        let router = crate::server::build_router(state);
-        let resp = router
-            .oneshot(unsigned_post(
-                "/api/v1/sync/notify",
-                &notify_body("a//tmp/gitlawb-probe", &peer_did),
-                "198.51.100.20:5000",
-            ))
-            .await
-            .unwrap();
-        assert_eq!(resp.status(), StatusCode::BAD_REQUEST);
-        let bytes = axum::body::to_bytes(resp.into_body(), usize::MAX)
-            .await
-            .unwrap();
-        let body = String::from_utf8(bytes.to_vec()).unwrap();
-        assert!(body.contains("repo"), "body must name the field: {body}");
-        assert!(
-            !body.contains(&repos_dir),
-            "body must not leak repos_dir ({repos_dir}): {body}"
-        );
-        for trimmed in repos_dir.trim_start_matches("./").split('/') {
-            assert!(
-                trimmed.is_empty() || !body.contains(trimmed),
-                "body must not leak a repos_dir segment ({trimmed}): {body}"
-            );
-        }
-        // A rooted path is a leading '/' with something after it. A bare quoted
-        // '/' is the separator the message is allowed to talk about.
-        assert!(
-            !body
-                .split(|c: char| c.is_whitespace() || c == '"' || c == '\'')
-                .any(|tok| tok.starts_with('/') && tok.len() > 1),
-            "body must not contain a rooted filesystem path: {body}"
-        );
-    }
-
-    #[sqlx::test]
-    async fn sync_notify_rejection_writes_no_ref_update(pool: PgPool) {
-        // The second half of R1: the ref-update feed row sits after enqueue_sync,
-        // so the early return must skip it too. Asserted, not argued.
-        let state = test_state(pool.clone()).await;
-        let peer_did = seed_peer(&state).await;
-        let router = crate::server::build_router(state);
-        let resp = router
-            .oneshot(unsigned_post(
-                "/api/v1/sync/notify",
-                &notify_body("a//tmp/gitlawb-probe", &peer_did),
-                "198.51.100.21:5000",
-            ))
-            .await
-            .unwrap();
-        assert_eq!(resp.status(), StatusCode::BAD_REQUEST);
-        let count: i64 = sqlx::query_scalar("SELECT COUNT(*) FROM received_ref_updates")
-            .fetch_one(&pool)
-            .await
-            .unwrap();
-        assert_eq!(
-            count, 0,
-            "a rejected notify must not seed the ref-update feed"
-        );
-    }
-
-    #[sqlx::test]
-    async fn sync_notify_accepts_valid_slug_and_records_ref_update(pool: PgPool) {
-        // must-not-break, and the control that keeps the two negative assertions
-        // above from being vacuous: a well-formed slug still queues and still
-        // writes its ref-update row.
-        let state = test_state(pool.clone()).await;
-        let db = state.db.clone();
-        let peer_did = seed_peer(&state).await;
-        let router = crate::server::build_router(state);
-        let resp = router
-            .oneshot(unsigned_post(
-                "/api/v1/sync/notify",
-                &notify_body("z6Mkfoo/hello", &peer_did),
-                "198.51.100.22:5000",
-            ))
-            .await
-            .unwrap();
-        assert_eq!(resp.status(), StatusCode::OK);
-        let queued = db.dequeue_pending_syncs(100).await.unwrap();
-        assert_eq!(queued.len(), 1);
-        assert_eq!(queued[0].repo, "z6Mkfoo/hello");
-        let count: i64 =
-            sqlx::query_scalar("SELECT COUNT(*) FROM received_ref_updates WHERE repo = $1")
-                .bind("z6Mkfoo/hello")
-                .fetch_one(&pool)
-                .await
-                .unwrap();
-        assert_eq!(
-            count, 1,
-            "the valid path must still write the ref-update row"
-        );
-    }
-
-    #[sqlx::test]
-    async fn sync_notify_unknown_peer_wins_over_malformed_slug(pool: PgPool) {
-        // Ordering: the slug check sits after the peer gate, so an unknown peer
-        // still gets the unknown-peer message rather than a slug complaint.
-        let state = test_state(pool).await;
-        let stranger = Keypair::generate().did().to_string();
-        let router = crate::server::build_router(state);
-        let resp = router
-            .oneshot(unsigned_post(
-                "/api/v1/sync/notify",
-                &notify_body("a//tmp/gitlawb-probe", &stranger),
-                "198.51.100.23:5000",
-            ))
-            .await
-            .unwrap();
-        assert_eq!(resp.status(), StatusCode::BAD_REQUEST);
-        let bytes = axum::body::to_bytes(resp.into_body(), usize::MAX)
-            .await
-            .unwrap();
-        let body = String::from_utf8(bytes.to_vec()).unwrap();
-        assert!(
-            body.contains("unknown peer DID"),
-            "peer gate must still fire first: {body}"
-        );
-    }
-
-    #[sqlx::test]
-    async fn unsigned_announce_then_notify_cannot_queue_a_repos_dir_escape(pool: PgPool) {
-        // The reachability case for #272, committed in its attack form. Two
-        // unsigned requests through the shipped router, no signature and no
-        // config change: announce the attacker's own DID (which is what makes
-        // them a "known peer"), then notify with `a/<abs>/gitlawb-probe`. That
-        // slug used to reach PathBuf::join in the sync worker, where the
-        // absolute second component discarded repos_dir and planted a mirror at
-        // /<abs>/gitlawb-probe.git. The notify must now be refused outright.
-        //
-        // This has to run over crate::server::build_router rather than a
-        // directly mounted handler: the point of the test is that the default
-        // layer stack (optional_signature, not require_signature) leaves the
-        // route open, so mounting the handler alone would assert nothing about
-        // reachability.
-        let state = test_state(pool).await;
-        let db = state.db.clone();
-        let attacker = Keypair::generate().did().to_string();
-        let router = crate::server::build_router(state);
-
-        let announce =
-            format!(r#"{{"did":"{attacker}","http_url":"https://attacker.example.com"}}"#);
-        let resp = router
-            .clone()
-            .oneshot(unsigned_post(
-                "/api/v1/peers/announce",
-                &announce,
-                "198.51.100.40:5000",
-            ))
-            .await
-            .unwrap();
-        // Asserted, not assumed: this plan does not change announce, and the
-        // whole attack rests on an unsigned announce still succeeding.
-        assert_eq!(
-            resp.status(),
-            StatusCode::OK,
-            "unsigned announce still admits an attacker-generated DID"
-        );
-
-        let outside = tempfile::TempDir::new().unwrap();
-        let escape_dir = outside.path().join("gitlawb-probe");
-        let slug = format!("a/{}", escape_dir.display());
-        let resp = router
-            .oneshot(unsigned_post(
-                "/api/v1/sync/notify",
-                &notify_body(&slug, &attacker),
-                "198.51.100.41:5000",
-            ))
-            .await
-            .unwrap();
-        assert_eq!(
-            resp.status(),
-            StatusCode::BAD_REQUEST,
-            "notify must refuse the escaping slug {slug:?}"
-        );
-
-        let queued = db.dequeue_pending_syncs(100).await.unwrap();
-        assert!(
-            queued.is_empty(),
-            "sync_queue must stay empty, got {:?}",
-            queued.iter().map(|q| &q.repo).collect::<Vec<_>>()
-        );
-        assert!(
-            !escape_dir.with_extension("git").exists(),
-            "nothing may be written at the escape target"
-        );
-    }
-
     #[sqlx::test]
     async fn announce_still_accepts_unsigned_in_default_mode(pool: PgPool) {
         // must-not-over-reach: adding the brake must not tighten announce's

crates/gitlawb-node/src/api/pulls.rs

@@ -65,7 +65,6 @@ pub async fn create_pr(
         status: "open".to_string(),
         merged_by_did: None,
         merged_at: None,
-        head_commit: None,
         created_at: now.clone(),
         updated_at: now,
     };
@@ -218,17 +217,12 @@ pub async fn merge_pr(
     let disk_path = guard.path().to_path_buf();
     let merger_did = auth.0;
 
-    // The source head this merge is about to consume, read under the write lock.
-    // The PR row above was loaded before the lock, so a push landing in between
-    // already moved the branch and the row's own `head_commit` is stale; the
-    // value frozen on the merged PR has to be what was merged. `None` when the
-    // ref cannot be resolved, which leaves whatever the push path last stored.
-    let merged_source_head = store::list_refs(&disk_path).ok().and_then(|refs| {
-        let want = format!("refs/heads/{}", pr.source_branch);
-        refs.into_iter()
-            .find(|(name, _)| *name == want)
-            .map(|(_, sha)| sha)
-    });
+    // Snapshot the target branch's pre-merge tip so a failed durable upload can
+    // roll the merge commit back on local disk (mirrors create_issue). Without
+    // this the local fast path in RepoStore::acquire serves the merged tree until
+    // a later write refreshes the archive, even though the client got a 5xx.
+    let target_ref = format!("refs/heads/{}", pr.target_branch);
+    let prior_target = store::ref_oid(&disk_path, &target_ref).ok().flatten();
 
     let merge_result = store::merge_branch(
         &disk_path,
@@ -238,15 +232,38 @@ pub async fn merge_pr(
         &pr.title,
     );
 
-    // Always release the advisory lock — even on error; upload to Tigris only on success.
-    guard.release(merge_result.is_ok()).await;
+    // Always release the advisory lock, even on error; upload to Tigris only on
+    // success. A false return means the durable upload failed/timed out: the merge
+    // commit applied to local disk but never reached object storage, so the next
+    // acquire_write reverts it from the stale archive. Report 5xx so the client
+    // retries rather than marking the PR merged in the DB while the merged tree
+    // was silently lost. Fail BEFORE merge_pr / touch_repo / webhooks. (P1
+    // data-loss guard, same as receive-pack.) On that failure, roll the local
+    // merge back to the captured tip BEFORE the lock releases.
+    let upload_ok = guard
+        .release_with_failure_cleanup(merge_result.is_ok(), |path| {
+            if let Some(oid) = &prior_target {
+                if let Err(e) = store::set_ref(path, &target_ref, oid) {
+                    tracing::warn!(repo = %record.name, err = %e,
+                        "failed to roll back local merge commit after failed durable upload; \
+                         a local-fast-path read may serve the un-uploaded merged tree");
+                }
+            }
+        })
+        .await;
 
     let merge_sha = merge_result.map_err(|e| AppError::Git(e.to_string()))?;
 
-    state
-        .db
-        .merge_pr(&pr.id, &merger_did, merged_source_head.as_deref())
-        .await?;
+    if !upload_ok {
+        tracing::error!(repo = %record.name,
+            "durable upload failed after merge; reporting failure so the client retries");
+        return Err(AppError::Git(format!(
+            "durable storage upload failed for {}",
+            record.name
+        )));
+    }
+
+    state.db.merge_pr(&pr.id, &merger_did).await?;
     let _ = state.db.touch_repo(&record.id).await;
 
     webhooks::fire_event(
@@ -444,156 +461,166 @@ pub async fn list_comments(
 
 #[cfg(test)]
 mod tests {
+    // super::* brings in the handler plus the axum Extension/Path/State
+    // extractors and the Utc/Uuid/PullRequest types the seed uses.
     use super::*;
-    use axum::body::Body;
-    use axum::http::{Method, StatusCode};
-    use axum::routing::post;
-    use axum::Router;
-    use std::process::Command;
-    use tower::ServiceExt;
+    use std::path::Path as StdPath;
+    use std::time::Duration;
 
-    const OWNER_DID: &str = "did:key:z6MkpTHR8VNsBxYAAWHut2Geadd9jSwuBV8xRoAnwWsdvktH";
-    const STALE_SHA: &str = "9999999999999999999999999999999999999999";
-
-    fn git(dir: &std::path::Path, args: &[&str]) {
-        let out = Command::new("git")
-            .args(args)
-            .current_dir(dir)
-            .env("GIT_AUTHOR_NAME", "t")
-            .env("GIT_AUTHOR_EMAIL", "[email protected]")
-            .env("GIT_COMMITTER_NAME", "t")
-            .env("GIT_COMMITTER_EMAIL", "[email protected]")
-            .output()
-            .expect("git runs");
-        assert!(
-            out.status.success(),
-            "git {args:?} failed: {}",
-            String::from_utf8_lossy(&out.stderr)
-        );
+    // Object store whose upload() parks forever, so release()'s bounded timeout
+    // is the only thing that completes it, modeling a stalled durable PUT.
+    // exists() is false so acquire_write never downloads over the local bare repo.
+    struct StallStore;
+    #[async_trait::async_trait]
+    impl crate::git::tigris::ObjectStore for StallStore {
+        async fn exists(&self, _o: &str, _r: &str) -> anyhow::Result<bool> {
+            Ok(false)
+        }
+        async fn upload(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            std::future::pending::<()>().await;
+            Ok(())
+        }
+        async fn download(&self, _o: &str, _r: &str, _p: &StdPath) -> anyhow::Result<()> {
+            Ok(())
+        }
+        async fn delete(&self, _o: &str, _r: &str) -> anyhow::Result<()> {
+            Ok(())
+        }
     }
 
-    /// A bare repo with `main` and a `feature` branch one commit ahead.
-    /// Returns the bare path and the real `feature` tip.
-    fn bare_repo_with_feature(
-        root: &std::path::Path,
-        owner_slug: &str,
-    ) -> (std::path::PathBuf, String) {
-        let work = root.join("work");
-        std::fs::create_dir_all(&work).unwrap();
-        git(&work, &["init", "-q", "-b", "main", "."]);
-        std::fs::write(work.join("a.txt"), "one").unwrap();
-        git(&work, &["add", "."]);
-        git(&work, &["commit", "-qm", "one"]);
-        git(&work, &["checkout", "-qb", "feature"]);
-        std::fs::write(work.join("b.txt"), "two").unwrap();
-        git(&work, &["add", "."]);
-        git(&work, &["commit", "-qm", "two"]);
-        git(&work, &["checkout", "-q", "main"]);
-
-        let bare_dir = root.join(owner_slug);
-        std::fs::create_dir_all(&bare_dir).unwrap();
-        let bare = bare_dir.join("demo.git");
-        git(
-            root,
-            &[
-                "clone",
-                "-q",
-                "--bare",
-                work.to_str().unwrap(),
-                bare.to_str().unwrap(),
-            ],
-        );
-
-        let tip = Command::new("git")
-            .args(["rev-parse", "refs/heads/feature"])
-            .current_dir(&bare)
-            .output()
-            .unwrap();
-        let tip = String::from_utf8_lossy(&tip.stdout).trim().to_string();
-        assert_eq!(tip.len(), 40, "feature tip must resolve");
-        (bare, tip)
-    }
-
-    /// Merging freezes the stored head at the source commit the merge actually
-    /// consumed. The PR row is loaded before the write lock is taken, so a push
-    /// that lands in between leaves the row's own `head_commit` stale; seeding a
-    /// value that matches nothing on disk stands in for that race, and the
-    /// merged PR must still come out pointing at the real `feature` tip.
+    // The failed-upload rollback on the merge path. merge_pr writes the merge
+    // commit to LOCAL disk, then release() uploads to durable storage. When
+    // that upload times out the handler must 5xx (before merge_pr/webhooks)
+    // AND roll the target branch back to its pre-merge tip while the advisory
+    // lock is held, so a local-fast-path read never serves the un-uploaded
+    // merged tree. RED with the cleanup closure at ~pulls.rs:244 reverted to
+    // `|_| {}`: main stays at the merge commit.
     #[sqlx::test]
-    async fn merge_stamps_the_source_head_it_merged_over_a_racing_value(pool: sqlx::PgPool) {
-        let tmp = tempfile::tempdir().unwrap();
-        let owner_slug = OWNER_DID.replace([':', '/'], "_");
-        let (bare, feature_tip) = bare_repo_with_feature(tmp.path(), &owner_slug);
+    async fn merge_failed_upload_rolls_back_target_ref(pool: sqlx::PgPool) {
+        use axum::response::IntoResponse;
+        use std::process::Command;
 
         let mut state = crate::test_support::test_state(pool.clone()).await;
-        state.repo_store =
-            crate::git::repo_store::RepoStore::for_testing(tmp.path().to_path_buf(), pool);
+        let repos_dir = tempfile::TempDir::new().unwrap();
+        state.repo_store = crate::git::repo_store::RepoStore::new(
+            repos_dir.path().to_path_buf(),
+            Some(std::sync::Arc::new(StallStore)),
+            pool.clone(),
+        )
+        .with_release_upload_timeout(Duration::from_millis(200));
 
-        let now = Utc::now();
-        let repo = crate::db::RepoRecord {
-            id: uuid::Uuid::new_v4().to_string(),
-            name: "demo".into(),
-            owner_did: OWNER_DID.into(),
-            description: None,
-            is_public: true,
-            default_branch: "main".into(),
-            created_at: now,
-            updated_at: now,
-            disk_path: bare.to_string_lossy().into_owned(),
-            forked_from: None,
-            machine_id: None,
-        };
-        state.db.create_repo(&repo).await.unwrap();
+        let owner = "z6mergerollbackowner";
+        let name = "mergerollback";
+        let owner_slug = owner.replace([':', '/'], "_");
+        let bare = repos_dir
+            .path()
+            .join(&owner_slug)
+            .join(format!("{name}.git"));
 
-        let pr = PullRequest {
-            id: Uuid::new_v4().to_string(),
-            repo_id: repo.id.clone(),
-            number: 1,
-            title: "add b".into(),
-            body: None,
-            author_did: OWNER_DID.into(),
-            source_branch: "feature".into(),
-            target_branch: "main".into(),
-            status: "open".into(),
-            merged_by_did: None,
-            merged_at: None,
-            head_commit: None,
-            created_at: now.to_rfc3339(),
-            updated_at: now.to_rfc3339(),
-        };
-        state.db.create_pr(&pr).await.unwrap();
-        // The stale value the racing push scenario leaves on the row.
         state
             .db
-            .set_open_pr_heads(&repo.id, "feature", STALE_SHA)
+            .upsert_mirror_repo(owner, name, &bare.to_string_lossy(), None, false)
             .await
             .unwrap();
 
-        let db = state.db.clone();
-        let app = Router::new()
-            .route(
-                "/api/v1/repos/{owner}/{repo}/pulls/{number}/merge",
-                post(merge_pr),
-            )
-            .with_state(state);
+        fn git(args: &[&str], dir: &StdPath) -> String {
+            let out = Command::new("git")
+                .args(args)
+                .current_dir(dir)
+                .output()
+                .unwrap();
+            assert!(
+                out.status.success(),
+                "git {args:?} failed: {}",
+                String::from_utf8_lossy(&out.stderr)
+            );
+            String::from_utf8_lossy(&out.stdout).trim().to_string()
+        }
 
-        let response = app
-            .oneshot(crate::test_support::signed_request_as(
-                OWNER_DID,
-                Method::POST,
-                "/api/v1/repos/z6MkpTHR8VNsBxYAAWHut2Geadd9jSwuBV8xRoAnwWsdvktH/demo/pulls/1/merge",
-                Body::empty(),
-            ))
+        // Scratch repo: c1 on main, c2 on feature; the bare clone carries both
+        // branches, so the merge of feature into main has real work to do.
+        let work = tempfile::TempDir::new().unwrap();
+        git(&["init", "-q", "-b", "main", "."], work.path());
+        git(&["config", "user.email", "t@t"], work.path());
+        git(&["config", "user.name", "t"], work.path());
+        std::fs::write(work.path().join("f.txt"), "one").unwrap();
+        git(&["add", "f.txt"], work.path());
+        git(&["commit", "-q", "-m", "c1"], work.path());
+        git(&["checkout", "-q", "-b", "feature"], work.path());
+        std::fs::write(work.path().join("g.txt"), "two").unwrap();
+        git(&["add", "g.txt"], work.path());
+        git(&["commit", "-q", "-m", "c2"], work.path());
+
+        std::fs::create_dir_all(bare.parent().unwrap()).unwrap();
+        let out = Command::new("git")
+            .args([
+                "clone",
+                "--bare",
+                "-q",
+                &work.path().to_string_lossy(),
+                &bare.to_string_lossy(),
+            ])
+            .output()
+            .unwrap();
+        assert!(
+            out.status.success(),
+            "clone --bare failed: {}",
+            String::from_utf8_lossy(&out.stderr)
+        );
+
+        let pre_merge_tip = store::ref_oid(&bare, "refs/heads/main")
+            .unwrap()
+            .expect("main exists before the merge");
+
+        // Seed the open PR row the handler looks up.
+        let record = state.db.get_repo(owner, name).await.unwrap().unwrap();
+        let now = Utc::now().to_rfc3339();
+        state
+            .db
+            .create_pr(&PullRequest {
+                id: Uuid::new_v4().to_string(),
+                repo_id: record.id.clone(),
+                number: 1,
+                title: "t".into(),
+                body: None,
+                author_did: owner.to_string(),
+                source_branch: "feature".into(),
+                target_branch: "main".into(),
+                status: "open".to_string(),
+                merged_by_did: None,
+                merged_at: None,
+                created_at: now.clone(),
+                updated_at: now,
+            })
             .await
             .unwrap();
-        assert_eq!(response.status(), StatusCode::OK);
 
-        let merged = db.get_pr(&repo.id, 1).await.unwrap().unwrap();
-        assert_eq!(merged.status, "merged");
+        // Merge as the owner: the merge commit applies locally, the durable
+        // upload stalls and times out, and the handler must report 5xx.
+        let resp = super::merge_pr(
+            State(state.clone()),
+            Extension(AuthenticatedDid(owner.to_string())),
+            Path((owner.to_string(), name.to_string(), 1)),
+        )
+        .await;
+        let status = match resp {
+            Ok(_) => StatusCode::OK,
+            Err(e) => e.into_response().status(),
+        };
+        assert!(
+            status.is_server_error(),
+            "a stalled durable upload on merge must 5xx, got {status}"
+        );
+
+        // The target branch must still point at the pre-merge tip: the failed
+        // upload rolled the merge commit back while the lock was held.
+        let tip = store::ref_oid(&bare, "refs/heads/main")
+            .unwrap()
+            .expect("main still exists after the rollback");
         assert_eq!(
-            merged.head_commit,
-            Some(feature_tip),
-            "the merge must stamp the source head it consumed, not the racing value"
+            tip, pre_merge_tip,
+            "a failed durable upload must roll the target branch back to its \
+             pre-merge tip; a local-fast-path read served the un-uploaded merge"
         );
     }
 }

crates/gitlawb-node/src/api/status/mod.rs

@@ -1,657 +0,0 @@
-//! Commit status claims: the write path and the combined read.
-//!
-//! POST /api/v1/repos/:owner/:repo/statuses/:sha — append one claim (owner only)
-//! GET  /api/v1/repos/:owner/:repo/commits/:sha/status — the combined projection
-//!
-//! Claims are append-only: a producer reporting twice for the same context
-//! leaves both rows and the visible status is a projection over the history.
-//! Reporting twice means two REQUESTS, though. An exact repeat of one already
-//! accepted is answered with 200 and the row it wrote, not a second row: the
-//! projection elects the highest `seq`, so a replayed claim would not duplicate
-//! a verdict but overturn the one that superseded it.
-
-use axum::extract::{Extension, Path, State};
-use axum::http::StatusCode;
-use axum::Json;
-use chrono::Utc;
-use serde::{Deserialize, Serialize};
-use uuid::Uuid;
-
-use crate::auth::AuthenticatedDid;
-use crate::db::StatusClaim;
-use crate::error::{AppError, Result};
-use crate::state::AppState;
-
-/// Claims for one (repo, commit, producer, context). Generous for a real
-/// producer; a producer past it is misbehaving.
-const MAX_CLAIMS_PER_TUPLE: i64 = 100;
-/// Distinct contexts for one (repo, commit). The context string is caller-chosen
-/// and free-form, so the tuple cap alone bounds nothing.
-const MAX_CONTEXTS_PER_COMMIT: i64 = 50;
-/// Claim rows for one repo within the trailing window the db layer counts over.
-/// The commit SHA is caller-chosen and never existence-checked, so a writer at
-/// the two caps above can still fan out over fresh 40-hex strings without this
-/// one. It bounds the RATE of that fan-out rather than a lifetime total, because
-/// nothing prunes the table: a lifetime bound would close the surface on a repo
-/// permanently while answering with a status every client retries.
-const MAX_CLAIMS_PER_REPO_PER_WINDOW: i64 = 10_000;
-
-#[derive(Deserialize)]
-pub struct CreateStatusRequest {
-    pub state: String,
-    pub context: String,
-    pub target_url: Option<String>,
-    pub description: Option<String>,
-}
-
-/// The wire state set (KTD-1): GitHub's four commit-status states, so absence
-/// stays distinguishable without a fifth value no client understands.
-const CLAIM_STATES: [&str; 4] = ["error", "failure", "pending", "success"];
-
-/// Bounds on the signature material this write path persists verbatim. Every one
-/// of the four is caller-influenced: the headers come off the request, the
-/// signing string grows with the component list the caller's Signature-Input
-/// chose, and the body is the caller's. Without a bound the claim log's row size
-/// is set by whoever writes to it, so these are explicit and generous rather than
-/// implicit and absent. A conforming `sign_request` produces roughly 100, 200 and
-/// 400 bytes for the first three.
-const MAX_SIGNATURE_CHARS: usize = 512;
-const MAX_SIGNATURE_INPUT_CHARS: usize = 1024;
-const MAX_SIGNING_STRING_CHARS: usize = 4096;
-/// The body is a `CreateStatusRequest`, whose own fields are already capped at
-/// roughly 3.3 KB in total by the three limits below; this leaves room for JSON
-/// framing and nothing more.
-///
-/// Enforced twice, at two different moments, because the handler's check alone
-/// is not the bound it advertises: `require_signature` buffers the whole body
-/// before any handler runs, so by the time [`create_status`] measures it the
-/// allocation has already happened. `build_router` applies the same number as a
-/// `RequestBodyLimitLayer` at the transport, outside the auth layers, which is
-/// what makes it a bound on what is ALLOCATED and not only on what is STORED.
-/// The handler keeps its own check regardless — it is the one that holds for a
-/// handler reached by any other route table, and it is what the material size
-/// tests drive.
-pub(crate) const MAX_REQUEST_BODY_BYTES: usize = 8192;
-
-const MAX_CONTEXT_CHARS: usize = 255;
-const MAX_TARGET_URL_CHARS: usize = 2048;
-const MAX_DESCRIPTION_CHARS: usize = 1024;
-
-/// The 201 body for an accepted claim: what the client needs to identify the row
-/// it just wrote, and nothing else.
-///
-/// Deliberately not the stored [`StatusClaim`]. That struct carries the signature
-/// material, and serializing it echoed the signature, the signing string and the
-/// whole request body back on every write — the body as a JSON array of integers.
-/// Write-time provenance belongs in the row, not on the wire, which is the same
-/// line [`StatusEntry`] draws on the read side.
-#[derive(Serialize)]
-pub struct CreatedStatus {
-    pub id: String,
-    /// The database-assigned ordering key, so the client can name its own row.
-    pub seq: i64,
-    pub repo_id: String,
-    pub commit_sha: String,
-    pub state: String,
-    pub context: String,
-    pub target_url: Option<String>,
-    pub description: Option<String>,
-    pub producer_did: String,
-    pub created_at: String,
-}
-
-impl CreatedStatus {
-    fn from_claim(claim: StatusClaim, seq: i64) -> Self {
-        Self {
-            id: claim.id,
-            seq,
-            repo_id: claim.repo_id,
-            commit_sha: claim.commit_sha,
-            state: claim.state,
-            context: claim.context,
-            target_url: claim.target_url,
-            description: claim.description,
-            producer_did: claim.producer_did,
-            created_at: claim.created_at,
-        }
-    }
-}
-
-/// POST /api/v1/repos/:owner/:repo/statuses/:sha
-pub async fn create_status(
-    State(state): State<AppState>,
-    Extension(auth): Extension<AuthenticatedDid>,
-    Path((owner, name, sha)): Path<(String, String, String)>,
-    material: Option<Extension<crate::auth::SignatureMaterial>>,
-    Json(req): Json<CreateStatusRequest>,
-) -> Result<(StatusCode, Json<CreatedStatus>)> {
-    // Read-visibility first, then owner, and the order is the security property.
-    // authorize_repo_read denies a quarantined repo before the visibility gate and
-    // answers with the repo's own not-found, byte-identical to a missing repo, so
-    // a caller who cannot read the repo cannot learn it exists. Loading the repo
-    // and comparing the owner would answer 403 there, turning this endpoint into
-    // an existence oracle. require_repo_owner then 403s a non-owner of a repo the
-    // caller can read, where existence is not secret.
-    let (record, _rules) =
-        crate::api::authorize_repo_read(&state, &owner, &name, Some(&auth.0), "/").await?;
-    crate::api::require_repo_owner(&record, &auth.0)?;
-
-    let commit_sha = normalize_sha(&sha)?;
-    let claim_state = validate_state(&req.state)?;
-    let context = validate_context(&req.context)?;
-    let target_url = validate_target_url(req.target_url.as_deref())?;
-    let description = validate_description(req.description.as_deref())?;
-
-    // KTD-5: the signature material is captured here because the request carries
-    // the only copy, and `signing_string` is the exact byte sequence the
-    // middleware verified. Its absence means this handler was reached without
-    // `require_signature` — a server misconfiguration (a route group that lost
-    // its auth layer, a handler mounted somewhere new), not anything the client
-    // did, so it is a 500 and no row is written. Storing an empty payload instead
-    // would leave every claim from that moment on unverifiable with nothing going
-    // red, which is exactly the absence-renders-as-success shape.
-    let Some(Extension(material)) = material else {
-        return Err(AppError::Internal(anyhow::anyhow!(
-            "status claim write reached without verified signature material — \
-             the route is not behind require_signature"
-        )));
-    };
-    // The body is carried only behind the `PersistsSignedBody` marker, which
-    // this route's group applies outside the auth layers. Its absence means the
-    // marker did not reach the middleware, because the layer was dropped or
-    // reordered. That is the same class of misconfiguration as the branch above
-    // and gets the same answer. Storing an empty body instead would write claims
-    // that carry a signature over a digest of bytes nobody kept, which is
-    // exactly the case `stored_claim_re_verifies_and_a_tampered_row_does_not`
-    // exists to make impossible.
-    let Some(body) = material.body.clone() else {
-        return Err(AppError::Internal(anyhow::anyhow!(
-            "status claim write reached with signature material carrying no request body: \
-             the route lost its persist-body marker or applied it inside the auth layers"
-        )));
-    };
-    // Bounded before anything is written. The material is verified, which says
-    // the caller holds the key, not that what they signed is a reasonable size.
-    bound("signature", material.signature.len(), MAX_SIGNATURE_CHARS)?;
-    bound(
-        "signature_input",
-        material.signature_input.len(),
-        MAX_SIGNATURE_INPUT_CHARS,
-    )?;
-    bound(
-        "signing_string",
-        material.signing_string.len(),
-        MAX_SIGNING_STRING_CHARS,
-    )?;
-    bound("request body", body.len(), MAX_REQUEST_BODY_BYTES)?;
-    // After the bounds, so the scan below walks a signing string of known size.
-    require_body_bound_signature(&material, &body)?;
-
-    let digest = request_digest(&material, &body);
-    let (signature, signature_input, signing_string, request_body) = (
-        material.signature,
-        material.signature_input,
-        material.signing_string,
-        body.to_vec(),
-    );
-
-    let claim = StatusClaim {
-        id: Uuid::new_v4().to_string(),
-        // Ignored on insert: the database assigns the ordering key (KTD-3).
-        seq: 0,
-        repo_id: record.id.clone(),
-        commit_sha,
-        state: claim_state,
-        context,
-        target_url,
-        description,
-        // Both are the owner identity today (KTD-5); they split when delegated
-        // capabilities land.
-        producer_did: auth.0.clone(),
-        authorizing_did: auth.0,
-        signature,
-        signature_input,
-        signing_string,
-        request_body,
-        request_digest: digest,
-        created_at: Utc::now().to_rfc3339(),
-    };
-
-    let caps = crate::db::ClaimCaps {
-        per_tuple: MAX_CLAIMS_PER_TUPLE,
-        contexts_per_commit: MAX_CONTEXTS_PER_COMMIT,
-        per_repo_window: MAX_CLAIMS_PER_REPO_PER_WINDOW,
-    };
-    match state.db.insert_status_claim_capped(&claim, &caps).await? {
-        crate::db::ClaimInsert::Inserted(seq) => Ok((
-            StatusCode::CREATED,
-            Json(CreatedStatus::from_claim(claim, seq)),
-        )),
-        // Already recorded, so 200 and the original row rather than 201 and a
-        // second one. The claim the caller gets back is the stored one, id and
-        // seq included, which is what makes a retry safe to treat as success.
-        crate::db::ClaimInsert::AlreadyRecorded(existing) => {
-            let seq = existing.seq;
-            Ok((
-                StatusCode::OK,
-                Json(CreatedStatus::from_claim(*existing, seq)),
-            ))
-        }
-        crate::db::ClaimInsert::CapExceeded(which) => Err(AppError::TooManyRequests(format!(
-            "claim limit reached for {which}"
-        ))),
-    }
-}
-
-/// Refuse a claim whose signature does not cover the body about to be stored.
-///
-/// A claim IS provenance, so the signature stored beside it has to bind the
-/// bytes stored beside it. It does not do that directly: `signing_string`
-/// reaches the body only through a `content-digest` line. And a caller can sign
-/// a content-digest the request then never carries. `require_signature` rebuilds
-/// the covered values from the request, so an ABSENT `Content-Digest` header
-/// makes the covered digest the empty string on the signing side and the
-/// verifying side alike; the Ed25519 check passes over method, path and an empty
-/// digest, and any body at all rides along. The row that write produces asserts
-/// a provenance that cannot be re-verified: the signing string covers a digest
-/// of bytes that are not the ones in `request_body`.
-///
-/// The predicate here is the write-time form of the one `re_verify` runs against
-/// a stored row in the tests — recompute the digest of the bytes about to be
-/// persisted and require the signing string to carry exactly that line — so what
-/// this route accepts and what a third party can later re-verify are the same
-/// question, asked at both ends.
-///
-/// This is a route property, checked where the claim becomes history, and it is
-/// deliberately not the only line of defence: requiring the header in
-/// `require_signature` closes the same hole one layer earlier and for every
-/// signed route, and that change belongs there rather than here because no other
-/// route's correctness should depend on the status handler. The two agree and
-/// neither substitutes for the other — without the middleware's requirement the
-/// empty digest is still accepted everywhere else, and without this one a later
-/// relaxation of the middleware would silently resume writing unverifiable
-/// history.
-///
-/// A 400, not a 500: the material is exactly what the client signed, and what
-/// they signed does not cover what they sent.
-fn require_body_bound_signature(
-    material: &crate::auth::SignatureMaterial,
-    body: &[u8],
-) -> Result<()> {
-    let covered = format!(
-        "\"content-digest\": {}",
-        gitlawb_core::http_sig::compute_content_digest(body)
-    );
-    // Line-exact rather than a substring search: `build_signing_string` emits one
-    // component per line, and a `contains` would also be satisfied by the digest
-    // appearing INSIDE some other covered component's value, which proves
-    // nothing about what the signature covered.
-    if material.signing_string.lines().any(|line| line == covered) {
-        return Ok(());
-    }
-    Err(AppError::BadRequest(
-        "the signature does not cover this request body: send a Content-Digest header \
-         computed over the body and name \"content-digest\" among the signature's covered \
-         components"
-            .into(),
-    ))
-}
-
-/// The identity of one signed write, as a hex sha-256.
-///
-/// All four inputs together, because a replay is the same bytes arriving twice:
-/// the signature, the input that names what it covers, the canonical string it
-/// was verified over, and the body that string covers through a content-digest.
-/// The signature alone would nearly do (it covers the other three transitively),
-/// but hashing what is actually stored means the digest describes the row rather
-/// than a claim about it.
-///
-/// Each field is length-prefixed so no two different requests can concatenate to
-/// the same bytes: without it, a signature ending in some prefix of the next
-/// field would collide with the shorter signature that absorbed it.
-///
-/// This is not a nonce, and deliberately so. A nonce table needs pruning and a
-/// pruning window is a second replay window; the claim row IS the record of what
-/// was accepted, so uniqueness on it is self-maintaining.
-fn request_digest(material: &crate::auth::SignatureMaterial, body: &[u8]) -> String {
-    use sha2::{Digest, Sha256};
-    let mut h = Sha256::new();
-    for part in [
-        material.signature.as_bytes(),
-        material.signature_input.as_bytes(),
-        material.signing_string.as_bytes(),
-        body,
-    ] {
-        h.update((part.len() as u64).to_be_bytes());
-        h.update(part);
-    }
-    format!("{:x}", h.finalize())
-}
-
-/// One reported context in the combined response. Signature material is
-/// deliberately absent: it is write-time provenance, not read-surface data.
-#[derive(Serialize)]
-pub struct StatusEntry {
-    pub state: String,
-    pub context: String,
-    pub target_url: Option<String>,
-    pub description: Option<String>,
-    pub producer_did: String,
-    pub created_at: String,
-}
-
-/// The combined commit status. `state` never leaves the four-value set (KTD-1),
-/// so absence is carried by `total_count` 0 with the pending state rather than a
-/// fifth value, and `reported_only` (R19) says out loud that the state covers the
-/// contexts that reported, not every check a caller expected.
-#[derive(Serialize)]
-pub struct CombinedStatus {
-    pub state: String,
-    pub sha: String,
-    pub total_count: usize,
-    pub statuses: Vec<StatusEntry>,
-    pub reported_only: bool,
-}
-
-/// GET /api/v1/repos/:owner/:repo/commits/:sha/status
-///
-/// The auth extension is optional and MUST be last in the extractor list: the
-/// route group carries `optional_signature`, so an unsigned caller reaches here
-/// with no `AuthenticatedDid` at all and a public repo still answers.
-pub async fn commit_status(
-    State(state): State<AppState>,
-    Path((owner, name, sha)): Path<(String, String, String)>,
-    auth: Option<Extension<AuthenticatedDid>>,
-) -> Result<Json<CombinedStatus>> {
-    let caller = auth.as_ref().map(|Extension(a)| a.0.as_str());
-    // The gate runs first and on the requested path, before any claim data is
-    // touched. Its deny is the repo's own not-found, byte-identical to a missing
-    // repo, so the status surface cannot answer "does this repo exist".
-    let (record, _rules) =
-        crate::api::authorize_repo_read(&state, &owner, &name, caller, "/").await?;
-
-    let commit_sha = normalize_sha(&sha)?;
-    let statuses = project_claims(&state, &record, &commit_sha).await?;
-
-    Ok(Json(CombinedStatus {
-        state: combined_state(&statuses).to_string(),
-        sha: commit_sha,
-        total_count: statuses.len(),
-        statuses,
-        reported_only: true,
-    }))
-}
-
-/// The projection, in one place. Both read surfaces call it, so R3's "the rollup
-/// is derived by the same projection as the commit read" holds by construction
-/// rather than by two implementations happening to agree.
-///
-/// KTD-5: current authorization, not write-time authorization. An ownership
-/// transfer drops the prior owner's claims from the projection while the
-/// append-only history keeps them.
-async fn project_claims(
-    state: &AppState,
-    record: &crate::db::RepoRecord,
-    commit_sha: &str,
-) -> Result<Vec<StatusEntry>> {
-    let claims = state
-        .db
-        .latest_status_claims(&record.id, commit_sha, &record.owner_did)
-        .await?;
-    Ok(claims
-        .into_iter()
-        .map(|c| StatusEntry {
-            state: c.state,
-            context: c.context,
-            target_url: c.target_url,
-            description: c.description,
-            producer_did: c.producer_did,
-            created_at: c.created_at,
-        })
-        .collect())
-}
-
-/// The pull request head rollup (R11). `state` stays inside the four wire values
-/// whatever happened to the head (KTD-1): an unresolvable head is carried by
-/// `head_resolved`, alongside the pull request's own state, so a client can tell
-/// "the head could not be resolved" from "the head resolved and nothing reported"
-/// without a fifth state value no client understands.
-#[derive(Serialize)]
-pub struct PullRequestStatus {
-    pub number: i64,
-    /// The pull request's own state: open, closed, or merged.
-    pub pull_request_state: String,
-    pub head_resolved: bool,
-    /// The target commit, absent exactly when `head_resolved` is false.
-    pub sha: Option<String>,
-    pub state: String,
-    pub total_count: usize,
-    pub statuses: Vec<StatusEntry>,
-    pub reported_only: bool,
-}
-
-/// GET /api/v1/repos/:owner/:repo/pulls/:number/status
-///
-/// Same optional-auth shape as the commit read, and the auth extension MUST stay
-/// last in the extractor list.
-pub async fn pull_request_status(
-    State(state): State<AppState>,
-    Path((owner, name, number)): Path<(String, String, i64)>,
-    auth: Option<Extension<AuthenticatedDid>>,
-) -> Result<Json<PullRequestStatus>> {
-    let caller = auth.as_ref().map(|Extension(a)| a.0.as_str());
-    // The gate runs first, on the requested path, before the pull request row is
-    // loaded. Its deny is the repo's own not-found, byte-identical to a missing
-    // repo. Once it passes, a missing pull request number is the plain not-found:
-    // existence is no longer secret.
-    let (record, _rules) =
-        crate::api::authorize_repo_read(&state, &owner, &name, caller, "/").await?;
-
-    let pr = state
-        .db
-        .get_pr(&record.id, number)
-        .await?
-        .ok_or_else(|| AppError::NotFound(format!("PR #{number} not found")))?;
-
-    let head = rollup_head(&state, &record, &pr).await?;
-
-    let statuses = match &head {
-        Some(sha) => project_claims(&state, &record, sha).await?,
-        None => Vec::new(),
-    };
-
-    Ok(Json(PullRequestStatus {
-        number: pr.number,
-        pull_request_state: pr.status,
-        head_resolved: head.is_some(),
-        sha: head,
-        state: combined_state(&statuses).to_string(),
-        total_count: statuses.len(),
-        statuses,
-        reported_only: true,
-    }))
-}
-
-/// Branch-head lookups performed by the rollup fallback, counted so a test can
-/// assert WORK DONE rather than only the answer returned. A fallback that
-/// re-resolved on every read and then discarded the answer would be invisible in
-/// the response and visible here.
-#[cfg(test)]
-pub(crate) static BRANCH_RESOLVES: std::sync::atomic::AtomicUsize =
-    std::sync::atomic::AtomicUsize::new(0);
-
-/// The rollup's target commit (KTD-4).
-///
-/// The stored head wins whenever it is set. It is frozen at close or merge, so a
-/// closed or merged pull request keeps the commit the decision was made against,
-/// and only an OPEN pull request with no stored head falls back to the branch.
-///
-/// The fallback is one database read: the latest push this node recorded for the
-/// pull request's source branch, from `repo_push_events`. It deliberately does
-/// NOT go through `git::store` ref listing, which needs an acquired repository
-/// path — on a cold node that downloads the whole repository from object storage,
-/// and this read group carries no rate limiter, so an anonymous caller could
-/// drive repeated downloads with a URL.
-///
-/// `repo_push_events` is the source rather than `branch_cids` because
-/// `record_push_events` writes it for every ref update unconditionally, whereas
-/// the sole writer of `branch_cids` sits behind a pin CID. A node with no object
-/// pinning configured never writes that table, so a resolve keyed on it could
-/// never answer there and every open pull request without a stored head reported
-/// `head_resolved: false` forever. The residual limit is that only pushes taken
-/// after this shipped have rows, so a branch last pushed before then still does
-/// not resolve.
-///
-/// The resolved head is persisted, which makes an unauthenticated GET write. That
-/// is only acceptable because the write is self-limiting: it fires exactly when
-/// `head_commit` is absent and it fills it, and the fill is conditioned on that
-/// same absence in SQL, so the PERSIST happens at most once per pull request.
-/// The resolve attempt itself is not once-only — a pull request whose branch does
-/// not resolve stores nothing, so there is nothing to short-circuit on and the
-/// lookup runs again on every read until it succeeds.
-async fn rollup_head(
-    state: &AppState,
-    record: &crate::db::RepoRecord,
-    pr: &crate::db::PullRequest,
-) -> Result<Option<String>> {
-    if let Some(stored) = pr.head_commit.clone() {
-        return Ok(Some(stored));
-    }
-    if pr.status != "open" {
-        return Ok(None);
-    }
-
-    // The push path records the full ref while a pull request stores the bare
-    // branch name, the same mismatch `crate::api::repos::branch_from_ref`
-    // reconciles on the write side; this is that mapping run backwards. The empty
-    // branch is excluded for the same reason it is there: `refs/heads/` is not a
-    // branch and nothing can have written it.
-    if pr.source_branch.is_empty() {
-        return Ok(None);
-    }
-    let ref_name = format!("refs/heads/{}", pr.source_branch);
-    #[cfg(test)]
-    BRANCH_RESOLVES.fetch_add(1, std::sync::atomic::Ordering::SeqCst);
-    let resolved = state
-        .db
-        .latest_push_sha_for_ref(&record.id, &ref_name)
-        .await?
-        // A recorded head that is not a commit SHA is no target at all, and
-        // storing it would leave the rollup pointing at nothing forever.
-        .and_then(|sha| normalize_sha(&sha).ok());
-
-    let Some(sha) = resolved else {
-        return Ok(None);
-    };
-    // Best effort, and deliberately not `?`. The head is already resolved and the
-    // response does not depend on this write landing, so a transient database
-    // failure here must not turn a served read into a 500; the next read resolves
-    // again and re-attempts the fill. Same catch-and-log the sibling writes on the
-    // push path use (`update_open_pr_heads`, `record_push_events`).
-    if let Err(e) = state.db.set_pr_head_if_absent(&pr.id, &sha).await {
-        tracing::warn!(
-            err = %e,
-            pr_id = %pr.id,
-            "failed to persist a resolved pull request head; serving the resolved head anyway"
-        );
-    }
-    Ok(Some(sha))
-}
-
-/// KTD-1's fold: any error or failure yields failure, else any pending yields
-/// pending, else success. An empty set is pending, never success — total absence
-/// of a verdict is its own state and must not read as a pass (R10).
-fn combined_state(statuses: &[StatusEntry]) -> &'static str {
-    if statuses.is_empty() {
-        return "pending";
-    }
-    if statuses
-        .iter()
-        .any(|s| s.state == "error" || s.state == "failure")
-    {
-        "failure"
-    } else if statuses.iter().any(|s| s.state == "pending") {
-        "pending"
-    } else {
-        "success"
-    }
-}
-
-/// 40 hex characters, lowercased. The SHA is never existence-checked (KTD-7):
-/// a claim describes a commit object, and the object may not have arrived yet.
-fn normalize_sha(sha: &str) -> Result<String> {
-    if sha.len() == 40 && sha.bytes().all(|b| b.is_ascii_hexdigit()) {
-        Ok(sha.to_ascii_lowercase())
-    } else {
-        Err(AppError::BadRequest(
-            "commit sha must be exactly 40 hexadecimal characters".into(),
-        ))
-    }
-}
-
-/// One size bound on persisted signature material, measured in bytes.
-fn bound(what: &str, len: usize, max: usize) -> Result<()> {
-    if len > max {
-        return Err(AppError::BadRequest(format!(
-            "{what} must be at most {max} bytes"
-        )));
-    }
-    Ok(())
-}
-
-fn validate_state(state: &str) -> Result<String> {
-    if CLAIM_STATES.contains(&state) {
-        Ok(state.to_string())
-    } else {
-        Err(AppError::BadRequest(format!(
-            "state must be one of {}",
-            CLAIM_STATES.join(", ")
-        )))
-    }
-}
-
-/// The context is the projection key, so it is trimmed, bounded, and required to
-/// be free of control characters.
-fn validate_context(context: &str) -> Result<String> {
-    let trimmed = context.trim();
-    if trimmed.is_empty() || trimmed.chars().count() > MAX_CONTEXT_CHARS {
-        return Err(AppError::BadRequest(format!(
-            "context must be 1 to {MAX_CONTEXT_CHARS} characters"
-        )));
-    }
-    if trimmed.chars().any(char::is_control) {
-        return Err(AppError::BadRequest(
-            "context must not contain control characters".into(),
-        ));
-    }
-    Ok(trimmed.to_string())
-}
-
-fn validate_target_url(url: Option<&str>) -> Result<Option<String>> {
-    let Some(url) = url else { return Ok(None) };
-    if url.chars().count() > MAX_TARGET_URL_CHARS {
-        return Err(AppError::BadRequest(format!(
-            "target_url must be at most {MAX_TARGET_URL_CHARS} characters"
-        )));
-    }
-    let lower = url.to_ascii_lowercase();
-    if lower.starts_with("http://") || lower.starts_with("https://") {
-        Ok(Some(url.to_string()))
-    } else {
-        Err(AppError::BadRequest(
-            "target_url must be an http or https URL".into(),
-        ))
-    }
-}
-
-fn validate_description(description: Option<&str>) -> Result<Option<String>> {
-    match description {
-        Some(d) if d.chars().count() > MAX_DESCRIPTION_CHARS => Err(AppError::BadRequest(format!(
-            "description must be at most {MAX_DESCRIPTION_CHARS} characters"
-        ))),
-        other => Ok(other.map(str::to_string)),
-    }
-}
-
-#[cfg(test)]
-mod tests;

crates/gitlawb-node/src/auth/mod.rs

@@ -17,53 +17,6 @@ use crate::state::AppState;
 #[derive(Clone, Debug)]
 pub struct AuthenticatedDid(pub String);
 
-/// The RFC 9421 material this request was actually verified against, injected
-/// alongside [`AuthenticatedDid`] by `require_signature`.
-///
-/// A handler that records a signed claim as history has to store it: the header
-/// values and the canonical signing string are gone once the request ends, and a
-/// stored claim nobody can re-verify is not history. `signing_string` is the exact
-/// byte sequence the Ed25519 verification succeeded over, not a rebuild — a
-/// handler reconstructing it from headers would be verifying a different string
-/// than the middleware did.
-#[derive(Clone, Debug)]
-pub struct SignatureMaterial {
-    /// The `Signature` header value.
-    pub signature: String,
-    /// The `Signature-Input` header value.
-    pub signature_input: String,
-    /// The canonical bytes the signature covered.
-    pub signing_string: String,
-    /// The buffered request body, on the routes that persist it. The signing
-    /// string covers it only through the content-digest, so without these bytes
-    /// a stored claim can be shown to carry *a* valid signature over *some*
-    /// digest and nothing more. Carried here because the body is consumed
-    /// downstream by the extractor and cannot be recovered afterwards.
-    ///
-    /// `Option`, and populated only behind [`PersistsSignedBody`], because a
-    /// route that never stores the body has no use for a second handle to it.
-    /// The unlike-sized field is this one: the three above are header-derived
-    /// and bounded by the server's header limit, while a receive-pack POST
-    /// carries up to GITLAWB_MAX_PACK_BYTES (2 GB by default). `Bytes`, not
-    /// `Vec<u8>`, so the handle the marked routes do take is a refcount bump
-    /// over the buffer the middleware already holds rather than a copy.
-    pub body: Option<axum::body::Bytes>,
-}
-
-/// Marks a route group whose handler persists the signed request body.
-///
-/// Applied as an extension layer OUTSIDE the auth layers, so it is on the
-/// request before [`require_signature`] decides whether to carry the body. A
-/// marker applied inside them is never seen, and the failure is silent: the
-/// handler still compiles, tests that inject material by hand still pass, and
-/// production stores nothing. The status write handler refuses an absent body
-/// for that reason.
-///
-/// A marker rather than a path check because `require_signature` must not learn
-/// which URL persists claims; that knowledge belongs at the route table.
-#[derive(Clone, Copy, Debug)]
-pub struct PersistsSignedBody;
-
 /// Whether `caller` is authorized to push to `record`.
 ///
 /// Phase 1 (`GITLAWB_ENFORCE_OWNER_PUSH`): owner-only, via the canonical
@@ -285,28 +238,10 @@ pub async fn require_signature(request: Request, next: Next) -> Response {
 
     tracing::info!(did = %sig.key_id, "✓ authenticated request");
 
-    // The body travels only where it is stored. On every other signed route the
-    // middleware's handle would keep the buffer alive for the rest of the layer
-    // chain, until the handler's own extractor consumes the request.
-    let body = parts
-        .extensions
-        .get::<PersistsSignedBody>()
-        .map(|_| body_bytes.clone());
-
-    let material = SignatureMaterial {
-        signature: sig_header,
-        signature_input: sig_input,
-        signing_string,
-        body,
-    };
-
     let mut request = Request::from_parts(parts, Body::from(body_bytes));
     request
         .extensions_mut()
         .insert(AuthenticatedDid(sig.key_id.to_string()));
-    // Carry the verified material forward for handlers that persist a signed
-    // claim; none of it can be recovered once the request is gone.
-    request.extensions_mut().insert(material);
     next.run(request).await
 }
 
@@ -580,6 +515,7 @@ mod tests {
             repo_store: crate::git::repo_store::RepoStore::for_testing(PathBuf::from("/tmp"), pool),
             rate_limiter: RateLimiter::new(100, Duration::from_secs(60)),
             create_ip_rate_limiter: RateLimiter::new(1000, Duration::from_secs(3600)),
+            write_rate_limiter: RateLimiter::new(1000, Duration::from_secs(3600)),
             push_rate_limiter: RateLimiter::new(600, Duration::from_secs(3600)),
             push_limiter_trust: crate::rate_limit::TrustedProxy::None,
             sync_trigger_rate_limiter: RateLimiter::new(60, Duration::from_secs(3600)),
@@ -588,100 +524,6 @@ mod tests {
         }
     }
 
-    /// Reports what the signature middleware put in `SignatureMaterial.body`,
-    /// so a test can tell "captured" from "deliberately not captured" apart
-    /// from "the middleware never ran".
-    async fn report_body(material: Option<axum::Extension<SignatureMaterial>>) -> String {
-        match material {
-            Some(axum::Extension(m)) => match m.body {
-                Some(bytes) => format!("captured:{}", bytes.len()),
-                None => "absent".to_string(),
-            },
-            None => "no-material".to_string(),
-        }
-    }
-
-    fn signed_post(path: &str, body: &[u8]) -> Request {
-        let kp = Keypair::generate();
-        let signed = gitlawb_core::http_sig::sign_request(&kp, "POST", path, body);
-        Request::builder()
-            .method("POST")
-            .uri(path)
-            .header("content-digest", signed.content_digest)
-            .header("signature-input", signed.signature_input)
-            .header("signature", signed.signature)
-            .body(axum::body::Body::from(body.to_vec()))
-            .unwrap()
-    }
-
-    /// The body handle travels only on a route that asked for it.
-    ///
-    /// Both arms drive a genuinely signed request through `require_signature`;
-    /// the only difference is the [`PersistsSignedBody`] marker layer. A
-    /// middleware that captured unconditionally would make both arms read
-    /// `captured`, and one that never read the marker would make both read
-    /// `absent`, so the pair pins the decision rather than the mechanism.
-    #[tokio::test]
-    async fn the_body_is_carried_only_when_the_route_marks_itself_as_persisting_it() {
-        let body = br#"{"state":"success"}"#;
-
-        // The marker layer is added AFTER the middleware layer, which is what
-        // makes it the outer of the two and therefore the one that runs first.
-        // This is the same relative position `build_router` uses on the status
-        // write group, where the marker sits outside `add_auth_layers`.
-        let marked = Router::new()
-            .route("/", axum::routing::post(report_body))
-            .layer(middleware::from_fn(require_signature))
-            .layer(axum::Extension(PersistsSignedBody));
-        let resp = marked.oneshot(signed_post("/", body)).await.unwrap();
-        assert_eq!(resp.status(), StatusCode::OK);
-        let seen = axum::body::to_bytes(resp.into_body(), 2048).await.unwrap();
-        assert_eq!(
-            String::from_utf8(seen.to_vec()).unwrap(),
-            format!("captured:{}", body.len()),
-            "a route carrying the persist marker must receive the buffered body"
-        );
-
-        let unmarked = Router::new()
-            .route("/", axum::routing::post(report_body))
-            .layer(middleware::from_fn(require_signature));
-        let resp = unmarked.oneshot(signed_post("/", body)).await.unwrap();
-        assert_eq!(resp.status(), StatusCode::OK);
-        let seen = axum::body::to_bytes(resp.into_body(), 2048).await.unwrap();
-        assert_eq!(
-            String::from_utf8(seen.to_vec()).unwrap(),
-            "absent",
-            "a signed route that never persists the body must not be handed a handle to it"
-        );
-    }
-
-    /// The marker is read from the request the middleware was given, and the
-    /// request it hands downstream still carries it. Both halves matter: the
-    /// first is the capture decision, the second is that rebuilding the request
-    /// from its parts does not drop extensions a later layer may want.
-    #[tokio::test]
-    async fn the_marker_survives_the_request_rebuild() {
-        async fn report_marker(marker: Option<axum::Extension<PersistsSignedBody>>) -> String {
-            match marker {
-                Some(_) => "marked".to_string(),
-                None => "unmarked".to_string(),
-            }
-        }
-
-        let app = Router::new()
-            .route("/", axum::routing::post(report_marker))
-            .layer(middleware::from_fn(require_signature))
-            .layer(axum::Extension(PersistsSignedBody));
-        let resp = app.oneshot(signed_post("/", b"{}")).await.unwrap();
-        assert_eq!(resp.status(), StatusCode::OK);
-        let seen = axum::body::to_bytes(resp.into_body(), 2048).await.unwrap();
-        assert_eq!(
-            String::from_utf8(seen.to_vec()).unwrap(),
-            "marked",
-            "the marker must reach the handler, so the middleware saw it too"
-        );
-    }
-
     #[tokio::test]
     async fn require_ucan_chain_no_header_passes_through() {
         let state = make_test_state(Keypair::generate().did());

crates/gitlawb-node/src/config.rs

@@ -1,18 +1,45 @@
-use clap::Parser;
+use clap::{Parser, Subcommand};
 use std::path::PathBuf;
 
+/// Optional admin subcommands. When none is given, the binary runs the node
+/// daemon as before — the default (no-subcommand) startup path is unchanged.
+#[derive(Subcommand, Debug, Clone)]
+pub enum Command {
+    /// Dry-run (default) or, with --execute, delete empty spam-burst repos owned
+    /// by the known burst DID. Never touches the hard-excluded DIDs, and verifies
+    /// each repo is empty (zero git refs) per repo before selecting it.
+    PurgeSpam {
+        /// Actually delete the candidates. Omit for a dry-run that prints the
+        /// candidate list and deletes nothing.
+        #[arg(long, default_value_t = false)]
+        execute: bool,
+    },
+}
+
 #[derive(Parser, Debug, Clone)]
 #[command(name = "gitlawb-node", about = "gitlawb node daemon", version)]
 pub struct Config {
-    /// Directory where bare git repositories are stored
-    #[arg(long, env = "GITLAWB_REPOS_DIR", default_value = "./data/repos")]
+    /// Admin subcommand to run instead of the node daemon. Absent = run the node.
+    #[command(subcommand)]
+    pub command: Option<Command>,
+
+    /// Directory where bare git repositories are stored. `global` so it can
+    /// follow a subcommand (e.g. `gitlawb-node purge-spam --repos-dir …`).
+    #[arg(
+        long,
+        env = "GITLAWB_REPOS_DIR",
+        default_value = "./data/repos",
+        global = true
+    )]
     pub repos_dir: PathBuf,
 
-    /// PostgreSQL connection URL (Supabase or any Postgres instance)
+    /// PostgreSQL connection URL (Supabase or any Postgres instance). `global` so
+    /// admin subcommands accept it in either position.
     #[arg(
         long,
         env = "DATABASE_URL",
-        default_value = "postgresql://localhost/gitlawb"
+        default_value = "postgresql://localhost/gitlawb",
+        global = true
     )]
     pub database_url: String,
 
@@ -194,6 +221,22 @@ pub struct Config {
     )]
     pub db_max_connections: u32,
 
+    /// Maximum connections in the dedicated advisory-lock pool, separate from
+    /// the main pool above. Every in-flight repo write (git-receive-pack, fork,
+    /// merge) pins one connection here for its whole lifetime — the write, its
+    /// metadata/fan-out tail, and the bounded post-write archive upload — so
+    /// size it to the expected peak number of concurrent distinct-repo writers,
+    /// NOT small. Keeping it separate from GITLAWB_DB_MAX_CONNECTIONS is what
+    /// stops a push burst from starving normal request handlers. Keep
+    /// (main pool + lock pool) within the database server's max_connections.
+    #[arg(
+        long,
+        env = "GITLAWB_DB_LOCK_POOL_MAX_CONNECTIONS",
+        default_value_t = 32,
+        value_parser = clap::value_parser!(u32).range(1..)
+    )]
+    pub db_lock_pool_max_connections: u32,
+
     /// Maximum time a request waits for a pool connection before failing with
     /// 503, in seconds. Bounds queueing when the database is slow or down.
     #[arg(

crates/gitlawb-node/src/error.rs

@@ -35,9 +35,6 @@ pub enum AppError {
     #[error("invalid request: {0}")]
     BadRequest(String),
 
-    #[error("too many requests: {0}")]
-    TooManyRequests(String),
-
     #[error("incomplete: {0}")]
     Incomplete(String),
 
@@ -47,6 +44,9 @@ pub enum AppError {
     #[error("git service timed out: {0}")]
     Timeout(String),
 
+    #[error("service unavailable: {0}")]
+    Unavailable(String),
+
     #[error("database error: {0}")]
     Db(#[from] sqlx::Error),
 
@@ -132,9 +132,6 @@ impl IntoResponse for AppError {
             // IcaptchaProofRequired is handled above (it carries extra headers/fields).
             AppError::IcaptchaProofRequired { .. } => unreachable!("handled before this match"),
             AppError::BadRequest(msg) => (StatusCode::BAD_REQUEST, "bad_request", msg.clone()),
-            AppError::TooManyRequests(msg) => {
-                (StatusCode::TOO_MANY_REQUESTS, "rate_limited", msg.clone())
-            }
             AppError::Incomplete(msg) => {
                 (StatusCode::UNPROCESSABLE_ENTITY, "incomplete", msg.clone())
             }
@@ -142,6 +139,14 @@ impl IntoResponse for AppError {
             // 504, distinct from the 500 git_error and from the read-gate's 404 /
             // the auth 401, so the client can tell a deadline from a failure.
             AppError::Timeout(msg) => (StatusCode::GATEWAY_TIMEOUT, "git_timeout", msg.clone()),
+            // 503, retryable: a transient resource contention (e.g. the repo lock
+            // pool is pinned by a write burst or a purge holds the key), distinct
+            // from the 500 git_error so the client knows to back off and retry.
+            AppError::Unavailable(msg) => (
+                StatusCode::SERVICE_UNAVAILABLE,
+                "service_unavailable",
+                msg.clone(),
+            ),
             AppError::Db(e) if db_unavailable(e) => (
                 StatusCode::SERVICE_UNAVAILABLE,
                 DB_UNAVAILABLE_CODE,
@@ -182,4 +187,20 @@ mod tests {
             StatusCode::INTERNAL_SERVER_ERROR
         );
     }
+
+    #[test]
+    fn unavailable_maps_to_503_distinct_from_git_500() {
+        // The create-repo lock-acquire failure returns Unavailable so a client sees
+        // a retryable 503, not a terminal 500 git_error.
+        assert_eq!(
+            AppError::Unavailable("lock pool pinned".into())
+                .into_response()
+                .status(),
+            StatusCode::SERVICE_UNAVAILABLE
+        );
+        assert_eq!(
+            AppError::Git("x".into()).into_response().status(),
+            StatusCode::INTERNAL_SERVER_ERROR
+        );
+    }
 }

crates/gitlawb-node/src/git/issues.rs

@@ -61,6 +61,51 @@ pub fn create_issue(repo_path: &Path, issue_id: &str, json: &str) -> Result<()>
     Ok(())
 }
 
+/// Delete an issue ref. Used to roll back a locally written issue whose
+/// durable upload failed, so a retry does not duplicate it. The dangling
+/// blob object is harmless.
+pub fn delete_issue_ref(repo_path: &Path, issue_id: &str) -> Result<()> {
+    let ref_name = format!("refs/gitlawb/issues/{issue_id}");
+    let output = Command::new("git")
+        .args(["update-ref", "-d", &ref_name])
+        .current_dir(repo_path)
+        .output()
+        .context("failed to run git update-ref -d")?;
+
+    if !output.status.success() {
+        let stderr = String::from_utf8_lossy(&output.stderr);
+        anyhow::bail!("git update-ref -d failed: {stderr}");
+    }
+
+    Ok(())
+}
+
+/// Read the object id an issue ref currently points to (resolving an 8-char
+/// prefix), returning `(full_id, oid)`. Returns None if no such issue exists.
+/// Used to snapshot an issue ref's pre-mutation state so a failed durable
+/// upload can restore it, mirroring `create_issue`'s rollback.
+pub fn issue_ref_oid(repo_path: &Path, issue_id: &str) -> Result<Option<(String, String)>> {
+    let full_id = match resolve_issue_id(repo_path, issue_id)? {
+        Some(id) => id,
+        None => return Ok(None),
+    };
+    let ref_name = format!("refs/gitlawb/issues/{full_id}");
+    let oid = match crate::git::store::ref_oid(repo_path, &ref_name)? {
+        Some(oid) => oid,
+        None => return Ok(None),
+    };
+    Ok(Some((full_id, oid)))
+}
+
+/// Reset an issue ref back to a captured object id — rolls back a local
+/// mutation (e.g. a close) whose durable upload failed, so a local-fast-path
+/// read does not serve the un-uploaded state. `full_id` must be the resolved
+/// UUID (from [`issue_ref_oid`]). The now-dangling object is harmless.
+pub fn restore_issue_ref(repo_path: &Path, full_id: &str, oid: &str) -> Result<()> {
+    let ref_name = format!("refs/gitlawb/issues/{full_id}");
+    crate::git::store::set_ref(repo_path, &ref_name, oid)
+}
+
 /// List all issue refs and return their JSON content.
 pub fn list_issues(repo_path: &Path) -> Result<Vec<String>> {
     // List all refs under refs/gitlawb/issues/

crates/gitlawb-node/src/git/repo_store/download.rs

@@ -0,0 +1,431 @@
+//! Read-path download coordination (KTD-3) for [`RepoStore`].
+//!
+//! The read path's cache-miss handling lives here: the per-repo async mutex
+//! that serializes concurrent readers of the same repo, the holder's fetch +
+//! extract + under-lock publish, and the RAII temp-dir cleanup. `acquire` and
+//! `acquire_fresh` stay in the parent module and call into `download_published`.
+//! These methods reach the parent's private fields (`download_locks`) and
+//! private methods (`try_lock_repo`) directly, because a child module can see
+//! its parent's private items.
+
+use super::*;
+
+/// Outcome of a coordinated read-path download (KTD-3).
+pub(super) enum DownloadOutcome {
+    /// The local dir is present: this reader downloaded and published it under
+    /// the advisory lock, or a concurrent reader did and this one served it.
+    Published,
+    /// The download was discarded without publishing: the advisory lock was
+    /// contended (a live writer or purge holds it), the lock pool errored, or
+    /// the archive vanished under the lock (purged mid-download). The caller
+    /// degrades to its missing-path or serve-local-copy outcome.
+    Skipped,
+}
+
+/// RAII cleanup for a read-path download temp dir (finding 1). Its `Drop`
+/// removes the dir (best effort), so a handler future cancelled mid-download
+/// (axum drops it on client disconnect, the same hazard `RepoWriteGuard` /
+/// `RepoLockGuard` Drop impls exist for) cannot leak a fully-populated repo
+/// dir. `disarm` forgets the dir once the publish rename has consumed it.
+struct TempDownloadDir {
+    path: PathBuf,
+    armed: bool,
+}
+
+impl TempDownloadDir {
+    fn new(path: PathBuf) -> Self {
+        Self { path, armed: true }
+    }
+
+    fn path(&self) -> &Path {
+        &self.path
+    }
+
+    /// Disarm: the temp has been consumed (renamed into place), so `Drop` must
+    /// not try to remove it.
+    fn disarm(mut self) {
+        self.armed = false;
+    }
+}
+
+impl Drop for TempDownloadDir {
+    fn drop(&mut self) {
+        if self.armed {
+            let _ = std::fs::remove_dir_all(&self.path);
+        }
+    }
+}
+
+/// RAII removal of a download-coordination map entry (finding 4). Its `Drop`
+/// removes the entry from `download_locks` whenever the map still holds THIS
+/// Arc (ptr_eq-guarded so a newer entry inserted after an earlier removal is
+/// never clobbered). Removal is guaranteed on every path (a normal return or
+/// `?` drops it in the holder; once a download is in flight it travels with
+/// the spawned task and drops at settle, covering holder timeout and a
+/// handler future cancelled mid cold-read, which axum triggers on client
+/// disconnect), which is what keeps the map from growing per arbitrary
+/// requested name: the pre-fix explicit removals ran only at the labelled
+/// return points, which a cancelled future skipped, leaking the entry. The
+/// outer map is a std::sync::Mutex so
+/// this Drop can prune it synchronously (Drop cannot be async); it is only ever
+/// held briefly for a get/insert/remove, never across an await.
+struct DownloadEntryGuard {
+    locks: Arc<std::sync::Mutex<HashMap<String, Arc<Mutex<()>>>>>,
+    key: String,
+    entry: Arc<Mutex<()>>,
+}
+
+impl Drop for DownloadEntryGuard {
+    fn drop(&mut self) {
+        // A poisoned map is still safe to prune — recover the guard and remove.
+        let mut map = self.locks.lock().unwrap_or_else(|p| p.into_inner());
+        if map
+            .get(&self.key)
+            .is_some_and(|cur| Arc::ptr_eq(cur, &self.entry))
+        {
+            map.remove(&self.key);
+        }
+    }
+}
+
+impl RepoStore {
+    /// Coordinated read-path download (KTD-3). Serializes concurrent readers
+    /// of the same repo on a per-repo async mutex: the first one in becomes
+    /// the holder and runs [`download_and_publish`](Self::download_and_publish);
+    /// a contended reader awaits the holder, re-checks the local dir on wake,
+    /// and serves what the holder published instead of re-downloading. Only
+    /// callers that already confirmed the archive exists reach this, and the
+    /// map entry is removed on completion, so the map cannot grow per
+    /// arbitrary requested name.
+    pub(super) async fn download_published(
+        &self,
+        owner_did: &str,
+        repo_name: &str,
+        owner_slug: &str,
+        local_path: &Path,
+        store: &Arc<dyn ObjectStore>,
+        local_existed_at_entry: bool,
+    ) -> Result<DownloadOutcome> {
+        let map_key = format!("{owner_slug}/{repo_name}");
+        let entry = {
+            let mut map = self.download_locks.lock().unwrap();
+            Arc::clone(
+                map.entry(map_key.clone())
+                    .or_insert_with(|| Arc::new(Mutex::new(()))),
+            )
+        };
+        let held = match Arc::clone(&entry).try_lock_owned() {
+            Ok(g) => g,
+            Err(_) => {
+                // Bounded coalescing wait: once a holder's download is in
+                // flight, the spawned task owns this lock (and the map entry)
+                // until it settles, whatever happens to the holder (timeout or
+                // cancellation), so an unbounded lock().await here would park
+                // this reader for as long as a stall lasts. On elapse, report
+                // the same timeout error the holder would. Do NOT touch the
+                // map entry on this path: pruning it while the download is in
+                // flight would let the next arrival start a duplicate download.
+                let waited = tokio::time::timeout(
+                    self.release_upload_timeout,
+                    Arc::clone(&entry).lock_owned(),
+                )
+                .await;
+                let g = match waited {
+                    Ok(g) => g,
+                    Err(_) => {
+                        if local_path.exists() {
+                            // A concurrent reader published while we waited.
+                            return Ok(DownloadOutcome::Published);
+                        }
+                        return Err(anyhow::anyhow!("read download timed out"));
+                    }
+                };
+                if local_path.exists() {
+                    // A concurrent reader published while we waited — serve it.
+                    // The holder's own guard prunes the map entry on its exit.
+                    return Ok(DownloadOutcome::Published);
+                }
+                // The holder degraded without publishing; this reader takes over.
+                // The prior holder removed the map entry on its way out, so
+                // re-register THIS entry before downloading, so readers arriving
+                // now coalesce onto it instead of starting a duplicate download
+                // (finding 3). Narrows the duplicate-download window; publishes
+                // stay serialized by the advisory lock regardless.
+                {
+                    let mut map = self.download_locks.lock().unwrap();
+                    map.entry(map_key.clone())
+                        .or_insert_with(|| Arc::clone(&entry));
+                }
+                g
+            }
+        };
+        // RAII: remove the map entry when this holder's work is done (finding
+        // 4). Created only once THIS reader holds the per-repo lock, so a
+        // parked or timed-out waiter never prunes the entry out from under a
+        // live holder; its Drop is ptr_eq-guarded, so it never clobbers a
+        // newer entry. Ownership moves into download_and_publish and from
+        // there into the spawned download task itself, so on EVERY settle path
+        // (success, download error, holder timeout, holder cancellation) the
+        // entry stays registered and the lock stays held until the in-flight
+        // download settles: waiters queue on the ONE download instead of each
+        // timeout or cancel wave starting a fresh full download of the same
+        // object.
+        let entry_guard = DownloadEntryGuard {
+            locks: Arc::clone(&self.download_locks),
+            key: map_key,
+            entry: Arc::clone(&entry),
+        };
+        self.download_and_publish(
+            owner_did,
+            repo_name,
+            owner_slug,
+            local_path,
+            store,
+            local_existed_at_entry,
+            entry_guard,
+            held,
+        )
+        .await
+    }
+
+    /// The holder's half of the coordinated download: fetch and extract the
+    /// archive into a temp sibling with NO advisory lock held (the network
+    /// phase must never pin a lock-pool connection — a cold-read burst across
+    /// distinct repos must not drain the writer-sized pool), then take the
+    /// per-repo advisory lock only around the publish: re-check the archive
+    /// still exists under the lock (a purge deletes it under this same lock,
+    /// so a post-purge downloader discards rather than resurrecting), swap
+    /// the temp copy into place, release. Advisory contention or a lock-pool
+    /// error discards the temp copy and degrades — no blocking lock or pool
+    /// wait exists anywhere on the read path.
+    ///
+    /// `local_existed_at_entry` selects the stale-swap guard (finding 2): a cold
+    /// `acquire` passes `false`, an `acquire_fresh` refresh passes `true`. Under
+    /// the lock, before the swap, we detect a writer that published a fresher
+    /// copy during our unlocked download window and serve theirs instead of
+    /// clobbering it with our now-stale temp.
+    ///
+    /// `entry_guard` and `held` are this holder's per-repo coordination (the
+    /// download_locks map entry and the locked per-repo mutex). Both move into
+    /// the spawned download task, which returns them through its JoinHandle on
+    /// success; on any other settle path (download error, task panic, holder
+    /// timeout, holder cancellation) they drop when the task settles, so
+    /// waiters keep queueing on the SAME in-flight download until it settles
+    /// rather than each starting a duplicate one.
+    #[allow(clippy::too_many_arguments)]
+    async fn download_and_publish(
+        &self,
+        owner_did: &str,
+        repo_name: &str,
+        owner_slug: &str,
+        local_path: &Path,
+        store: &Arc<dyn ObjectStore>,
+        local_existed_at_entry: bool,
+        entry_guard: DownloadEntryGuard,
+        held: tokio::sync::OwnedMutexGuard<()>,
+    ) -> Result<DownloadOutcome> {
+        let parent = local_path.parent().context("repo path has no parent")?;
+        std::fs::create_dir_all(parent).context("creating repo parent dir")?;
+        let file_name = local_path
+            .file_name()
+            .context("repo path has no file name")?
+            .to_string_lossy();
+
+        // Best-effort sweep of leftover temp siblings from a prior download whose
+        // cleanup never ran (a process crash or kill mid-download, where no Drop
+        // and no owning task survive to remove the temp). Serialized per
+        // repo by the download mutex, so no concurrent same-repo download owns a
+        // live temp here; scoped to THIS repo's prefix, so other repos are
+        // untouched.
+        let tmp_prefix = format!(".{file_name}.tmp-download.");
+        if let Ok(entries) = std::fs::read_dir(parent) {
+            for entry in entries.flatten() {
+                if entry.file_name().to_string_lossy().starts_with(&tmp_prefix) {
+                    let _ = std::fs::remove_dir_all(entry.path());
+                }
+            }
+        }
+
+        // Stale-swap signal (finding 2): capture the local dir's mtime BEFORE the
+        // unlocked download. Under the lock, a changed mtime (acquire_fresh,
+        // refreshing over an existing copy) or the dir's mere appearance (cold
+        // acquire) means a writer published a fresher copy during our download
+        // window, so we serve theirs rather than swap our stale temp over it.
+        let entry_mtime = std::fs::metadata(local_path)
+            .and_then(|m| m.modified())
+            .ok();
+
+        // Unique per-download temp target (same parent as local_path so the
+        // publish rename stays on one filesystem); mirrors the extract temp
+        // naming in tigris.rs. Wrapped in an RAII guard so a handler future
+        // cancelled mid-download (axum drops it on client disconnect) cannot
+        // leak the populated temp dir (the same hazard the guard Drop impls
+        // below exist for, finding 1).
+        let temp = TempDownloadDir::new(parent.join(format!(
+            ".{file_name}.tmp-download.{}",
+            uuid::Uuid::new_v4()
+        )));
+
+        // Ownership of the download AND the temp-dir guard moves into a spawned
+        // task, so timeout and handler cancellation share ONE cleanup path
+        // (finding F3 and its cancellation twin). `store.download` runs its tar
+        // extraction in `spawn_blocking`, and neither a tokio timeout nor a
+        // dropped handler future cancels a spawn_blocking task: pre-fix,
+        // dropping the download future (a timed-out await, or axum dropping the
+        // whole handler on client disconnect) detached the extraction, whose
+        // later rename resurrected `temp.path()` AFTER the guard's Drop had
+        // removed it, leaving an orphan only a later same-repo download sweeps.
+        // The spawned task is never dropped mid-flight, so the download future
+        // always runs to settle. On success it returns the still-armed guard
+        // through the JoinHandle: the caller consumes it and publishes below.
+        // If the caller timed out or was cancelled, the JoinHandle is gone, so
+        // the runtime drops the returned guard when the task settles and its
+        // Drop removes the temp dir; cleanup is tied to the extraction's
+        // completion, never racing it (this subsumes the previous detached
+        // janitor). A download error drops the guard inside the task with the
+        // same at-settle timing.
+        let store_dl = Arc::clone(store);
+        let (dl_owner, dl_repo, dl_target) = (
+            owner_slug.to_string(),
+            repo_name.to_string(),
+            temp.path().to_path_buf(),
+        );
+        // The per-repo coordination (`held`, `entry_guard`) moves into the task
+        // alongside the temp guard: whatever happens to THIS caller (timeout
+        // below, or a handler future cancelled by axum on client disconnect),
+        // the lock stays held and the map entry stays registered until the
+        // in-flight download settles, so waiters and new arrivals queue on the
+        // ONE download instead of starting duplicates. On success the tuple
+        // returns through the JoinHandle: the caller publishes under `held` as
+        // before. If the caller timed out or was cancelled, the runtime drops
+        // the returned tuple when the task settles, in field order: `temp`
+        // (dir removed), then `held` (lock released), then `entry_guard` (map
+        // entry pruned), the same unlock-before-prune order as a normal
+        // return.
+        let dl_task = tokio::spawn(async move {
+            // The guards travel as one `(held, entry_guard)` tuple so that
+            // wherever it drops (here on error, in the runtime at settle after
+            // a caller timeout or cancellation, or in the caller after a
+            // normal publish), its field order gives unlock-before-prune.
+            match store_dl.download(&dl_owner, &dl_repo, &dl_target).await {
+                Ok(()) => Ok((temp, (held, entry_guard))),
+                // The download settled, so cleanup is safe. Explicit drops pin
+                // the order (async-block captures have no guaranteed one):
+                // temp removed, lock released, map entry pruned last.
+                Err(e) => {
+                    drop(temp);
+                    drop(held);
+                    drop(entry_guard);
+                    Err(e)
+                }
+            }
+        });
+        // Bound the wait, not the work: a stalled GET would park this reader
+        // indefinitely. On timeout the caller gets the same error as before,
+        // but the coordination does NOT unwind: it lives in the task (see
+        // above) and is released only at settle. Waiters therefore stay queued
+        // on the SAME in-flight download (each bounded by its own timeout in
+        // download_published) instead of each timeout wave starting a fresh
+        // full download of the same object; the task keeps running and cleans
+        // up its temp when the download settles.
+        // `_coordination` is `(held, entry_guard)`, kept alive to the end of
+        // this function exactly as the separate guards were before: the
+        // publish below still runs with the per-repo lock held, and the map
+        // entry is pruned only on exit (unlock first, prune second, per the
+        // tuple's field order).
+        let (temp, _coordination) =
+            match tokio::time::timeout(self.release_upload_timeout, dl_task).await {
+                Ok(Ok(Ok(returned))) => returned,
+                Ok(Ok(Err(e))) => return Err(e),
+                // The task panicked; its unwind dropped the guards, removing
+                // the dir, releasing the lock, and pruning the map entry.
+                Ok(Err(join_err)) => {
+                    return Err(anyhow::Error::from(join_err).context("read download task failed"))
+                }
+                Err(_) => {
+                    warn!(repo = %repo_name, timeout_secs = self.release_upload_timeout.as_secs(),
+                        "read download timed out — discarding");
+                    // Dropping the JoinHandle detaches the task; the runtime
+                    // frees temp dir, lock, and map entry at settle (see the
+                    // spawn comment above).
+                    return Err(anyhow::anyhow!("read download timed out"));
+                }
+            };
+
+        let guard = match self.try_lock_repo(owner_did, repo_name).await {
+            Ok(Some(g)) => g,
+            Ok(None) => {
+                debug!(repo = %repo_name,
+                    "read download discarded — repo lock held by a live writer or purge");
+                return Ok(DownloadOutcome::Skipped);
+            }
+            Err(e) => {
+                warn!(repo = %repo_name, err = %e,
+                    "read download discarded — could not acquire repo lock");
+                return Ok(DownloadOutcome::Skipped);
+            }
+        };
+        // Re-check under the lock: the archive gone here means a purge won the
+        // key mid-download — discard, never publish (the read-side twin of the
+        // upload paths' under-lock dir re-check). Bounded so a stalled HEAD
+        // cannot pin the advisory lock + its lock-pool connection. Both a
+        // timeout and an exists() error collapse to "not present" and take the
+        // discard arm: fail closed rather than publish unconfirmed state.
+        let archive_present = matches!(
+            tokio::time::timeout(
+                self.release_upload_timeout,
+                store.exists(owner_slug, repo_name),
+            )
+            .await,
+            Ok(Ok(true))
+        );
+        if !archive_present {
+            warn!(repo = %repo_name,
+                "read download discarded — archive gone under lock (purged?)");
+            guard.release().await;
+            return Ok(DownloadOutcome::Skipped);
+        }
+
+        // Stale-swap guard (finding 2): a writer that locked, published a fresher
+        // copy, and released during our unlocked download window must not have
+        // its work clobbered by our now-stale temp. The only concurrent mutator
+        // of local_path is a writer or a purge (readers of the same repo
+        // serialize on the download mutex), so the signals below unambiguously
+        // mean "a writer published here."
+        let writer_published = match (local_existed_at_entry, entry_mtime) {
+            // acquire_fresh over an existing copy: presence is always true, so it
+            // cannot signal a writer; a CHANGED mtime can (the writer replaced
+            // the dir via rename, minting a fresh mtime).
+            (true, Some(entry_m)) => std::fs::metadata(local_path)
+                .and_then(|m| m.modified())
+                .map(|now| now != entry_m)
+                .unwrap_or(false),
+            // Cold acquire, or acquire_fresh whose dir was absent at entry: the
+            // dir being present now means a writer published it during the window.
+            _ => local_path.exists(),
+        };
+        if writer_published {
+            warn!(repo = %repo_name,
+                "read download discarded: a writer published a fresher copy during the download window");
+            guard.release().await;
+            // temp guard's Drop removes the stale temp; serve the fresher local copy.
+            return Ok(DownloadOutcome::Published);
+        }
+
+        let swapped = (|| -> std::io::Result<()> {
+            if local_path.exists() {
+                std::fs::remove_dir_all(local_path)?;
+            }
+            std::fs::rename(temp.path(), local_path)
+        })();
+        guard.release().await;
+        match swapped {
+            Ok(()) => {
+                temp.disarm(); // the rename consumed the temp; nothing to clean up
+                Ok(DownloadOutcome::Published)
+            }
+            // temp guard's Drop removes the temp if the rename left it behind.
+            Err(e) => Err(e).context("publishing downloaded repo into place"),
+        }
+    }
+}

crates/gitlawb-node/src/git/smart_http.rs

@@ -86,127 +86,6 @@ pub async fn receive_pack(
         .body(Body::from(output))?)
 }
 
-// ── receive-pack report-status ────────────────────────────────────────────
-//
-// PORTED VERBATIM from PR #384 (`fix/issue-26-split-1-durable-post-receive`),
-// which introduced these two functions for the same reason: `receive_pack`
-// forwards git's protocol response untouched, and a zero exit says nothing
-// about whether any individual ref was taken. Kept byte-for-byte identical so
-// the two branches converge on one parser rather than growing two that disagree
-// about what "accepted" means; whichever lands second should drop its copy.
-
-/// Parse the git-receive-pack report-status output to determine per-ref
-/// success/failure. Returns `(unpack_ok, per_ref_results)` where
-/// `per_ref_results` is a list of `(ref_name, is_ok)`.
-///
-/// The report-status format (after the sideband framing) is:
-/// ```text
-/// unpack ok\n          (or: unpack fail\n)
-/// ok <refname>\n       (per successful ref)
-/// ng <refname> <reason>\n  (per rejected ref)
-/// \n                   (empty line terminates)
-/// ```
-///
-/// Returns `None` if the output cannot be parsed (e.g. the client did
-/// not request report-status, or the output is truncated). In that
-/// case the caller should treat all refs as uncertain.
-pub fn parse_report_status(output: &[u8]) -> Option<(bool, Vec<(String, bool)>)> {
-    let text = std::str::from_utf8(output).ok()?;
-    // Strip sideband framing: each line starts with a pkt-line length
-    // prefix and a channel byte (1 = stdout, 2 = stderr). The actual
-    // data starts after the first `0000` flush packet or after we
-    // strip sideband bytes.
-    let stripped = strip_sideband(text)?;
-    // DIVERGES FROM #384 by these three lines, and deliberately.
-    //
-    // When the client asked for `side-band-64k` — which `git push` over smart
-    // HTTP does — receive-pack multiplexes the report onto band 1, and what
-    // travels on that band is itself a pkt-line stream. One pass of
-    // `strip_sideband` peels the outer framing and leaves `000eunpack ok\n...`,
-    // whose first line is not `unpack ok`, so the parse below returns None and
-    // every ref falls back to inconclusive: the parser would answer "cannot
-    // tell" for exactly the pushes it exists to classify. A second pass peels
-    // the inner layer. It is a no-op on the single-wrapped shape, because
-    // plain report text does not begin with four hex digits and
-    // `strip_sideband` declines it.
-    let stripped = strip_sideband(&stripped).unwrap_or(stripped);
-    let lines: Vec<&str> = stripped.lines().collect();
-    if lines.is_empty() {
-        return None;
-    }
-
-    // First non-empty line is "unpack ok" or "unpack fail".
-    let unpack_line = lines.iter().find(|l| !l.is_empty())?;
-    let unpack_ok = if unpack_line.starts_with("unpack ok") {
-        true
-    } else if unpack_line.starts_with("unpack fail") {
-        false
-    } else {
-        return None;
-    };
-
-    let mut results = Vec::new();
-    for line in &lines[1..] {
-        let line = line.trim();
-        if line.is_empty() {
-            break;
-        }
-        if let Some(rest) = line.strip_prefix("ok ") {
-            results.push((rest.to_string(), true));
-        } else if let Some(rest) = line.strip_prefix("ng ") {
-            // "ng <refname> <reason>" — skip the reason
-            let ref_name = rest.split_whitespace().next()?.to_string();
-            results.push((ref_name, false));
-        }
-    }
-
-    Some((unpack_ok, results))
-}
-
-/// Strip git sideband framing from a pkt-line encoded output.
-/// Sideband-encoded lines start with a 4-hex-digit length, then a
-/// channel byte (0x01=stdout, 0x02=stderr), then payload. Returns
-/// the decoded payload lines concatenated, or `None` if the framing
-/// is malformed.
-fn strip_sideband(text: &str) -> Option<String> {
-    let mut output = String::new();
-    let mut pos = 0;
-    let bytes = text.as_bytes();
-
-    loop {
-        if pos + 4 > bytes.len() {
-            break;
-        }
-        let len_str = std::str::from_utf8(&bytes[pos..pos + 4]).ok()?;
-        let len = usize::from_str_radix(len_str, 16).ok()?;
-        if len == 0 {
-            // Flush packet — end of sideband stream
-            break;
-        }
-        if len < 4 || pos + len > bytes.len() {
-            break;
-        }
-        let pkt_data = std::str::from_utf8(&bytes[pos + 4..pos + len]).ok()?;
-        pos += len;
-
-        // Sideband: first byte is channel (1=stdout, 2=stderr)
-        if let Some(payload) = pkt_data.strip_prefix('\x01') {
-            output.push_str(payload);
-        } else if pkt_data.starts_with('\x02') {
-            // stderr — skip (git error messages)
-        } else {
-            // Not sideband encoded — pass through
-            output.push_str(pkt_data);
-        }
-    }
-
-    if output.is_empty() {
-        None
-    } else {
-        Some(output)
-    }
-}
-
 /// Sends SIGTERM to a child's whole process group on drop, unless disarmed first.
 ///
 /// A served `git upload-pack`/`receive-pack` forks helpers such as `pack-objects`.

crates/gitlawb-node/src/git/store.rs

@@ -63,6 +63,105 @@ pub fn list_refs(repo_path: &Path) -> Result<Vec<(String, String)>> {
     Ok(refs)
 }
 
+/// Read the object id a single ref currently points to. Returns None if the
+/// ref does not exist. Used to snapshot a ref's pre-write state so a failed
+/// durable upload can roll the local write back.
+pub fn ref_oid(repo_path: &Path, ref_name: &str) -> Result<Option<String>> {
+    let output = Command::new("git")
+        .args(["rev-parse", "--verify", ref_name])
+        .current_dir(repo_path)
+        .output()
+        .context("failed to run git rev-parse")?;
+    if !output.status.success() {
+        return Ok(None);
+    }
+    let oid = String::from_utf8_lossy(&output.stdout).trim().to_string();
+    if oid.is_empty() {
+        return Ok(None);
+    }
+    Ok(Some(oid))
+}
+
+/// Force a single ref back to a captured object id — rolls back a local write
+/// (e.g. a merge commit) whose durable upload failed, so a local-fast-path
+/// read does not serve the un-uploaded state. The now-dangling objects are
+/// harmless.
+pub fn set_ref(repo_path: &Path, ref_name: &str, oid: &str) -> Result<()> {
+    let output = Command::new("git")
+        .args(["update-ref", ref_name, oid])
+        .current_dir(repo_path)
+        .output()
+        .context("failed to run git update-ref")?;
+    if !output.status.success() {
+        let stderr = String::from_utf8_lossy(&output.stderr);
+        bail!("git update-ref failed: {stderr}");
+    }
+    Ok(())
+}
+
+/// Restore a repo's refs to a previously captured snapshot: reset every
+/// snapshot ref to its captured object id, and delete any ref that exists now
+/// but was absent from the snapshot (created by the intervening write). Used
+/// to roll back a receive-pack whose durable upload failed, so the next
+/// local-fast-path read does not serve refs that never reached object storage.
+/// Best-effort per ref: an individual failure is collected and reported, but
+/// the remaining refs are still attempted.
+pub fn restore_refs(repo_path: &Path, snapshot: &[(String, String)]) -> Result<()> {
+    use std::collections::HashSet;
+
+    let mut errors: Vec<String> = Vec::new();
+
+    // Reset every snapshot ref to its captured oid (recreates refs the write
+    // deleted, and rewinds refs the write advanced).
+    for (ref_name, oid) in snapshot {
+        if let Err(e) = set_ref(repo_path, ref_name, oid) {
+            errors.push(format!("{ref_name}: {e}"));
+        }
+    }
+
+    // Delete refs that exist now but were not in the snapshot (created by the
+    // write being rolled back). A failure to LIST the current refs is collected
+    // like any per-ref failure, never returned early: the snapshot-reset half
+    // above already ran, and bailing here would discard its collected errors.
+    let snapshot_names: HashSet<&str> = snapshot.iter().map(|(r, _)| r.as_str()).collect();
+    match list_refs(repo_path) {
+        Ok(current) => {
+            for (ref_name, _) in &current {
+                if !snapshot_names.contains(ref_name.as_str()) {
+                    match Command::new("git")
+                        .args(["update-ref", "-d", ref_name])
+                        .current_dir(repo_path)
+                        .output()
+                    {
+                        Ok(output) if output.status.success() => {}
+                        Ok(output) => {
+                            let stderr = String::from_utf8_lossy(&output.stderr);
+                            errors.push(format!("{ref_name} (delete): {stderr}"));
+                        }
+                        Err(e) => {
+                            errors.push(format!(
+                                "{ref_name} (delete): failed to run git update-ref -d: {e}"
+                            ));
+                        }
+                    }
+                }
+            }
+        }
+        Err(e) => {
+            errors.push(format!("(extra-ref sweep) listing current refs: {e}"));
+        }
+    }
+
+    if !errors.is_empty() {
+        bail!(
+            "failed to restore {} ref(s): {}",
+            errors.len(),
+            errors.join("; ")
+        );
+    }
+    Ok(())
+}
+
 /// Read the current HEAD commit hash of a repository.
 /// Returns None if the repo is empty (no commits yet).
 pub fn head_commit(repo_path: &Path) -> Result<Option<String>> {
@@ -455,10 +554,176 @@ pub fn repo_disk_path(repos_dir: &Path, owner_did: &str, repo_name: &str) -> Pat
 
 #[cfg(test)]
 mod tests {
-    use super::branch_diff_names;
+    use super::{branch_diff_names, list_refs, restore_refs};
     use std::path::Path;
     use std::process::Command;
 
+    fn run_git(dir: &Path, args: &[&str]) -> String {
+        let out = Command::new("git")
+            .args(args)
+            .current_dir(dir)
+            .output()
+            .unwrap();
+        assert!(
+            out.status.success(),
+            "git {args:?} failed: {}",
+            String::from_utf8_lossy(&out.stderr)
+        );
+        String::from_utf8_lossy(&out.stdout).trim().to_string()
+    }
+
+    /// Bare repo with main at c2 and keeper at c2, whose "pre-push" state was
+    /// main and keeper both at c1. Returns (workdir, baredir, bare_path, c1, c2).
+    fn advanced_bare_repo() -> (
+        tempfile::TempDir,
+        tempfile::TempDir,
+        std::path::PathBuf,
+        String,
+        String,
+    ) {
+        let work = tempfile::TempDir::new().unwrap();
+        run_git(work.path(), &["init", "-q", "-b", "main", "."]);
+        run_git(work.path(), &["config", "user.email", "t@t"]);
+        run_git(work.path(), &["config", "user.name", "t"]);
+        std::fs::write(work.path().join("f.txt"), "one").unwrap();
+        run_git(work.path(), &["add", "f.txt"]);
+        run_git(work.path(), &["commit", "-qm", "c1"]);
+        let c1 = run_git(work.path(), &["rev-parse", "HEAD"]);
+        std::fs::write(work.path().join("f.txt"), "two").unwrap();
+        run_git(work.path(), &["add", "f.txt"]);
+        run_git(work.path(), &["commit", "-qm", "c2"]);
+        let c2 = run_git(work.path(), &["rev-parse", "HEAD"]);
+
+        let dir = tempfile::TempDir::new().unwrap();
+        let bare = dir.path().join("repo.git");
+        let out = Command::new("git")
+            .args([
+                "clone",
+                "--bare",
+                "-q",
+                &work.path().to_string_lossy(),
+                &bare.to_string_lossy(),
+            ])
+            .output()
+            .unwrap();
+        assert!(out.status.success(), "clone --bare failed");
+        // "The push" advanced main to c2 (already there from the clone), moved
+        // keeper c1 -> c2, and created feature at c2.
+        run_git(&bare, &["update-ref", "refs/heads/keeper", &c2]);
+        run_git(&bare, &["update-ref", "refs/heads/feature", &c2]);
+        (work, dir, bare, c1, c2)
+    }
+
+    fn sorted_refs(bare: &Path) -> Vec<(String, String)> {
+        let mut refs = list_refs(bare).unwrap();
+        refs.sort();
+        refs
+    }
+
+    /// A single unrestorable snapshot entry (invalid oid, `set_ref` fails) must
+    /// not abort the restore: every OTHER snapshot ref is still reset and the
+    /// push-created extra ref is still deleted, and the fn returns Err carrying
+    /// the failure. RED with the reset loop reverted to `set_ref(..)?`: the
+    /// broken entry is first in the snapshot, so an early return leaves keeper
+    /// at c2 and feature alive.
+    #[test]
+    fn restore_refs_partial_failure_still_resets_rest_and_deletes_extras() {
+        let (_work, _dir, bare, c1, c2) = advanced_bare_repo();
+
+        let snapshot = vec![
+            (
+                "refs/heads/broken".to_string(),
+                "not-a-valid-oid".to_string(),
+            ),
+            ("refs/heads/main".to_string(), c1.clone()),
+            ("refs/heads/keeper".to_string(), c1.clone()),
+        ];
+        let err = restore_refs(&bare, &snapshot).unwrap_err();
+        let msg = format!("{err:#}");
+        assert!(
+            msg.contains("refs/heads/broken"),
+            "the aggregated error names the failed ref: {msg}"
+        );
+
+        let refs = sorted_refs(&bare);
+        assert_eq!(
+            refs,
+            vec![
+                ("refs/heads/keeper".to_string(), c1.clone()),
+                ("refs/heads/main".to_string(), c1.clone()),
+            ],
+            "keeper and main are reset to c1 and feature is deleted despite the \
+             broken entry (c2 was {c2})"
+        );
+    }
+
+    /// Happy path unchanged by the hardening: a fully valid snapshot restores
+    /// exactly (refs rewound, created ref deleted) and returns Ok.
+    #[test]
+    fn restore_refs_happy_path_restores_snapshot_exactly() {
+        let (_work, _dir, bare, c1, _c2) = advanced_bare_repo();
+
+        let snapshot = vec![
+            ("refs/heads/main".to_string(), c1.clone()),
+            ("refs/heads/keeper".to_string(), c1.clone()),
+        ];
+        restore_refs(&bare, &snapshot).unwrap();
+
+        let refs = sorted_refs(&bare);
+        assert_eq!(
+            refs,
+            vec![
+                ("refs/heads/keeper".to_string(), c1.clone()),
+                ("refs/heads/main".to_string(), c1),
+            ],
+            "restore must reproduce the snapshot exactly"
+        );
+    }
+
+    /// When the internal `list_refs` (extra-ref sweep) fails, the snapshot-reset
+    /// half must still have run and its collected failures must surface in the
+    /// aggregated error, not be discarded by an early `?` return. The repo is a
+    /// real bare repo declaring repositoryformatversion=999, so both `set_ref`
+    /// and `list_refs` fail deterministically. RED with the sweep reverted to
+    /// `let current = list_refs(repo_path)?;`: the raw for-each-ref error
+    /// propagates without the "failed to restore" aggregate or the ref name.
+    #[test]
+    fn restore_refs_aggregates_when_internal_list_refs_fails() {
+        let dir = tempfile::TempDir::new().unwrap();
+        let bare = dir.path().join("repo.git");
+        let out = Command::new("git")
+            .args(["init", "--bare", "-q", &bare.to_string_lossy()])
+            .output()
+            .unwrap();
+        assert!(out.status.success(), "git init --bare failed");
+        std::fs::write(
+            bare.join("config"),
+            "[core]\n\trepositoryformatversion = 999\n\tbare = true\n",
+        )
+        .unwrap();
+        assert!(
+            list_refs(&bare).is_err(),
+            "precondition: list_refs must fail on the corrupted repo"
+        );
+
+        let snapshot = vec![("refs/heads/main".to_string(), "a".repeat(40))];
+        let err = restore_refs(&bare, &snapshot).unwrap_err();
+        let msg = format!("{err:#}");
+        assert!(
+            msg.contains("failed to restore"),
+            "must return the aggregated error, not the raw list_refs error: {msg}"
+        );
+        assert!(
+            msg.contains("refs/heads/main"),
+            "the set_ref failure from the reset half must survive the sweep \
+             failure: {msg}"
+        );
+        assert!(
+            msg.contains("listing current refs"),
+            "the sweep failure itself is also collected: {msg}"
+        );
+    }
+
     #[test]
     fn branch_diff_names_lists_changed_paths() {
         let td = tempfile::TempDir::new().unwrap();

crates/gitlawb-node/src/git/tigris.rs

@@ -11,6 +11,21 @@ use anyhow::{Context, Result};
 use aws_sdk_s3::Client as S3Client;
 use tracing::{debug, info};
 
+/// Object storage for git bare-repo archives. The seam that lets `RepoStore`
+/// (and tests) depend on storage behavior without a live bucket. `TigrisClient`
+/// is the production implementation.
+#[async_trait::async_trait]
+pub trait ObjectStore: Send + Sync {
+    /// Whether an archive exists for this repo.
+    async fn exists(&self, owner_slug: &str, repo_name: &str) -> Result<bool>;
+    /// Upload a local bare repo directory as an archive.
+    async fn upload(&self, owner_slug: &str, repo_name: &str, local_path: &Path) -> Result<()>;
+    /// Download and extract a repo archive to local disk.
+    async fn download(&self, owner_slug: &str, repo_name: &str, local_path: &Path) -> Result<()>;
+    /// Delete a repo archive.
+    async fn delete(&self, owner_slug: &str, repo_name: &str) -> Result<()>;
+}
+
 /// Wrapper around the S3 client with the configured bucket.
 #[derive(Clone)]
 pub struct TigrisClient {
@@ -35,9 +50,12 @@ impl TigrisClient {
     fn repo_key(owner_slug: &str, repo_name: &str) -> String {
         format!("repos/v1/{owner_slug}/{repo_name}.tar.zst")
     }
+}
 
+#[async_trait::async_trait]
+impl ObjectStore for TigrisClient {
     /// Check if a repo archive exists in Tigris.
-    pub async fn exists(&self, owner_slug: &str, repo_name: &str) -> Result<bool> {
+    async fn exists(&self, owner_slug: &str, repo_name: &str) -> Result<bool> {
         let key = Self::repo_key(owner_slug, repo_name);
         match self
             .s3
@@ -59,7 +77,7 @@ impl TigrisClient {
     }
 
     /// Upload a local bare repo directory to Tigris as a tar.zst archive.
-    pub async fn upload(&self, owner_slug: &str, repo_name: &str, local_path: &Path) -> Result<()> {
+    async fn upload(&self, owner_slug: &str, repo_name: &str, local_path: &Path) -> Result<()> {
         let key = Self::repo_key(owner_slug, repo_name);
         debug!(key = %key, path = %local_path.display(), "uploading repo to tigris");
 
@@ -89,12 +107,7 @@ impl TigrisClient {
     }
 
     /// Download a repo archive from Tigris and extract to local disk.
-    pub async fn download(
-        &self,
-        owner_slug: &str,
-        repo_name: &str,
-        local_path: &Path,
-    ) -> Result<()> {
+    async fn download(&self, owner_slug: &str, repo_name: &str, local_path: &Path) -> Result<()> {
         let key = Self::repo_key(owner_slug, repo_name);
         debug!(key = %key, path = %local_path.display(), "downloading repo from tigris");
 
@@ -128,8 +141,7 @@ impl TigrisClient {
     }
 
     /// Delete a repo archive from Tigris.
-    #[allow(dead_code)]
-    pub async fn delete(&self, owner_slug: &str, repo_name: &str) -> Result<()> {
+    async fn delete(&self, owner_slug: &str, repo_name: &str) -> Result<()> {
         let key = Self::repo_key(owner_slug, repo_name);
         self.s3
             .delete_object()

crates/gitlawb-node/src/main.rs

@@ -1,3 +1,4 @@
+mod admin;
 mod api;
 mod arweave;
 mod auth;
@@ -70,6 +71,12 @@ async fn main() -> Result<()> {
 
     let mut config = Config::parse();
 
+    // Admin subcommands run out-of-band and exit, never starting the node. The
+    // no-subcommand path below is the unchanged daemon startup.
+    if let Some(command) = config.command.clone() {
+        return run_admin_command(command, &config).await;
+    }
+
     // Merge the embedded seed list of public network nodes into the runtime
     // bootstrap peers. Operators can opt out via GITLAWB_BOOTSTRAP_DISABLE_SEEDS.
     bootstrap::merge_seeds(&mut config);
@@ -279,8 +286,22 @@ async fn main() -> Result<()> {
         None
     };
 
+    let object_store =
+        tigris.map(|t| std::sync::Arc::new(t) as std::sync::Arc<dyn git::tigris::ObjectStore>);
+    // Advisory-lock guards get their own pool, separate from the app pool
+    // (db.pool()) that serves normal handlers, so a concurrent-push burst pins
+    // lock connections here instead of starving request handlers. Sized to peak
+    // writers via GITLAWB_DB_LOCK_POOL_MAX_CONNECTIONS.
+    let lock_pool_acquire_timeout = std::time::Duration::from_secs(config.db_acquire_timeout_secs);
+    let lock_pool = Db::connect_lock_pool(
+        &config.database_url,
+        config.db_lock_pool_max_connections,
+        lock_pool_acquire_timeout,
+    )
+    .await
+    .context("connecting advisory-lock pool")?;
     let repo_store =
-        git::repo_store::RepoStore::new(config.repos_dir.clone(), tigris, db.pool().clone());
+        git::repo_store::RepoStore::new(config.repos_dir.clone(), object_store, lock_pool);
 
     // Per-DID limiter for the creation endpoints. Keyed on the authenticated
     // DID (attacker-varied), so bound its key set to cap memory.
@@ -294,36 +315,24 @@ async fn main() -> Result<()> {
     // capped regardless of how many identities it mints. Sized well above any
     // legitimate per-IP creation rate; GITLAWB_CREATE_RATE_LIMIT overrides, 0
     // disables. Bounded key set — the key is a client-influenced IP.
-    let create_limit = std::env::var("GITLAWB_CREATE_RATE_LIMIT")
-        .ok()
-        .and_then(|v| v.trim().parse::<usize>().ok())
-        .unwrap_or(120);
-    let create_ip_rate_limiter = rate_limit::RateLimiter::new_bounded(
-        create_limit,
-        std::time::Duration::from_secs(3600),
-        200_000,
-    );
-    if create_limit == 0 {
-        tracing::warn!("GITLAWB_CREATE_RATE_LIMIT=0 — per-IP creation rate limiting disabled");
-    }
+    let create_ip_rate_limiter = build_ip_limiter("GITLAWB_CREATE_RATE_LIMIT", 120, "creation");
+
+    // Per-IP brake for the authenticated non-creation write surface (issue/PR
+    // comments, labels, stars, merges, protect, replicas, visibility, tasks,
+    // bounties, profile, GraphQL mutations). Its own bucket, separate from the
+    // creation brake, so a write flood cannot drain the creation budget and vice
+    // versa (same rationale as the sync_trigger / peer_write split). Sized above
+    // any legitimate per-IP write rate — real agent automation makes many small
+    // writes per hour. GITLAWB_WRITE_RATE_LIMIT overrides; 0 disables. Bounded
+    // key set — the key is a client-influenced IP.
+    let write_rate_limiter = build_ip_limiter("GITLAWB_WRITE_RATE_LIMIT", 600, "write");
 
     // Push-path flood brake: max git-receive-pack requests per client IP per
     // hour (counts both the info/refs advertisement and the push POST). Sized
     // for heavy agent automation while still stopping flood traffic (the June
     // 2026 attack pushed several times per second per IP). GITLAWB_PUSH_RATE_LIMIT
     // overrides; 0 disables. Bounded key set — the key is a client-influenced IP.
-    let push_limit = std::env::var("GITLAWB_PUSH_RATE_LIMIT")
-        .ok()
-        .and_then(|v| v.trim().parse::<usize>().ok())
-        .unwrap_or(600);
-    let push_rate_limiter = rate_limit::RateLimiter::new_bounded(
-        push_limit,
-        std::time::Duration::from_secs(3600),
-        200_000,
-    );
-    if push_limit == 0 {
-        tracing::warn!("GITLAWB_PUSH_RATE_LIMIT=0 — per-IP push rate limiting disabled");
-    }
+    let push_rate_limiter = build_ip_limiter("GITLAWB_PUSH_RATE_LIMIT", 600, "push");
 
     // Which forwarded header the edge is trusted to set. Default None (trust
     // nothing, key on the socket peer). Fly nodes set GITLAWB_TRUSTED_PROXY=fly;
@@ -331,7 +340,7 @@ async fn main() -> Result<()> {
     let push_limiter_trust = rate_limit::TrustedProxy::from_env_value(
         &std::env::var("GITLAWB_TRUSTED_PROXY").unwrap_or_default(),
     );
-    tracing::info!(trust = ?push_limiter_trust, push_limit, "push rate limiter configured");
+    tracing::info!(trust = ?push_limiter_trust, "push rate limiter trust configured");
 
     // Peer-sync flood brakes, keyed on the resolved client IP (per-DID is useless
     // here — a did:key farm self-registers). Two buckets so an unsigned notify
@@ -373,6 +382,7 @@ async fn main() -> Result<()> {
         repo_store,
         rate_limiter,
         create_ip_rate_limiter,
+        write_rate_limiter,
         push_rate_limiter,
         push_limiter_trust,
         sync_trigger_rate_limiter,
@@ -408,22 +418,17 @@ async fn main() -> Result<()> {
 
     // Periodic cleanup of expired rate limit entries + consumed-proof ledger
     {
-        let rl = state.rate_limiter.clone();
-        let create_ip_rl = state.create_ip_rate_limiter.clone();
-        let push_rl = state.push_rate_limiter.clone();
-        let sync_trigger_rl = state.sync_trigger_rate_limiter.clone();
-        let peer_write_rl = state.peer_write_rate_limiter.clone();
+        // Clone the whole state so the sweep uses AppState::cleanup_rate_limiters
+        // (the single source of truth that reaps EVERY limiter, including
+        // write_rate_limiter) rather than a hand-maintained list that can omit one.
+        let state = state.clone();
         let db = state.db.clone();
         let mut shutdown_rx = state.subscribe_shutdown();
         tokio::spawn(async move {
             loop {
                 tokio::select! {
                     _ = tokio::time::sleep(std::time::Duration::from_secs(300)) => {
-                        rl.cleanup().await;
-                        create_ip_rl.cleanup().await;
-                        push_rl.cleanup().await;
-                        sync_trigger_rl.cleanup().await;
-                        peer_write_rl.cleanup().await;
+                        state.cleanup_rate_limiters().await;
                         let now = std::time::SystemTime::now()
                             .duration_since(std::time::UNIX_EPOCH)
                             .map(|d| d.as_secs() as i64)
@@ -564,6 +569,110 @@ async fn main() -> Result<()> {
     Ok(())
 }
 
+/// Pure resolver for a usize rate-limit env value. Returns `(resolved, invalid)`:
+/// an absent or empty value resolves to `default` and is NOT invalid; a non-empty
+/// value that fails to parse resolves to `default` and IS invalid (so the caller
+/// warns rather than silently disabling the intended, usually stricter, brake).
+fn resolve_rate_limit(raw: Option<&str>, default: usize) -> (usize, bool) {
+    match raw.map(str::trim) {
+        None | Some("") => (default, false),
+        Some(t) => match t.parse::<usize>() {
+            Ok(n) => (n, false),
+            Err(_) => (default, true),
+        },
+    }
+}
+
+/// Read a rate-limit env var, warning on a present-but-unparseable value. A typo
+/// like `GITLAWB_WRITE_RATE_LIMIT=5O` must not silently revert to the default and
+/// leave the operator believing a stricter cap is in force.
+fn rate_limit_from_env(var: &str, default: usize) -> usize {
+    let raw = std::env::var(var).ok();
+    let (value, invalid) = resolve_rate_limit(raw.as_deref(), default);
+    if invalid {
+        tracing::warn!(
+            var,
+            value = raw.as_deref().unwrap_or(""),
+            default,
+            "invalid rate-limit value — using default; the intended brake is NOT applied"
+        );
+    }
+    value
+}
+
+/// Build a bounded per-client-IP rate limiter from an env var: resolve the limit
+/// (honoring the present-but-unparseable warning in `rate_limit_from_env`), build
+/// the limiter with the shared 1-hour window and 200k-key cap, and warn when the
+/// limit is 0 (disabled). Collapses the identical create/write/push setup.
+fn build_ip_limiter(var: &str, default: usize, label: &str) -> rate_limit::RateLimiter {
+    let limit = rate_limit_from_env(var, default);
+    let limiter =
+        rate_limit::RateLimiter::new_bounded(limit, std::time::Duration::from_secs(3600), 200_000);
+    if limit == 0 {
+        tracing::warn!("{var}=0 — per-IP {label} rate limiting disabled");
+    } else {
+        tracing::info!(%var, limit, "per-IP {label} rate limiter configured");
+    }
+    limiter
+}
+
+/// Dispatch an admin subcommand. Connects the database directly (no server, no
+/// p2p, no metrics) and runs the requested tool to completion, then exits.
+async fn run_admin_command(command: config::Command, config: &Config) -> Result<()> {
+    match command {
+        config::Command::PurgeSpam { execute } => {
+            let acquire_timeout = std::time::Duration::from_secs(config.db_acquire_timeout_secs);
+            let db = Db::connect(
+                &config.database_url,
+                config.db_max_connections,
+                acquire_timeout,
+            )
+            .await
+            .context("connecting to database for purge-spam")?;
+            // Build the object store so purge is authoritative against remote
+            // state: the emptiness recheck consults the archive (not just
+            // possibly-stale local disk) and a successful purge deletes the
+            // archive too. When no bucket is configured this stays None and purge
+            // is local-only, exactly as before.
+            let object_store: Option<std::sync::Arc<dyn git::tigris::ObjectStore>> = if config
+                .tigris_bucket
+                .is_empty()
+            {
+                None
+            } else {
+                match git::tigris::TigrisClient::new(&config.tigris_bucket).await {
+                    Ok(client) => Some(std::sync::Arc::new(client)),
+                    Err(e) => {
+                        // Fail closed: without the configured store we cannot
+                        // do the authoritative recheck, and a local-only purge
+                        // on a Tigris deployment is the stale-view delete this
+                        // guards against. Refuse to run rather than fall back.
+                        anyhow::bail!(
+                                "purge-spam: GITLAWB_TIGRIS_BUCKET is set but the object-store client failed to initialize ({e}); refusing to run a local-only purge on a Tigris deployment"
+                            );
+                    }
+                }
+            };
+            // Give the purge-spam store a dedicated advisory-lock pool, separate
+            // from `db`'s app pool. The guard then holds a lock-pool connection
+            // while `delete_repo_by_id` runs on the app pool, so a purge at
+            // GITLAWB_DB_MAX_CONNECTIONS=1 does not self-deadlock (KTD3).
+            let lock_pool = Db::connect_lock_pool(
+                &config.database_url,
+                config.db_lock_pool_max_connections,
+                acquire_timeout,
+            )
+            .await
+            .context("connecting advisory-lock pool for purge-spam")?;
+            let repo_store =
+                git::repo_store::RepoStore::new(config.repos_dir.clone(), object_store, lock_pool);
+            admin::run_purge_spam(&db, &repo_store, &config.repos_dir, execute)
+                .await
+                .map(|_| ())
+        }
+    }
+}
+
 fn spawn_shutdown_signal(tx: watch::Sender<bool>) {
     tokio::spawn(async move {
         #[cfg(unix)]
@@ -1024,6 +1133,40 @@ fn load_or_create_keypair(config: &Config) -> Result<Keypair> {
     }
 }
 
+#[cfg(test)]
+mod rate_limit_env_tests {
+    use super::resolve_rate_limit;
+
+    #[test]
+    fn absent_uses_default_without_warning() {
+        assert_eq!(resolve_rate_limit(None, 600), (600, false));
+    }
+
+    #[test]
+    fn empty_or_whitespace_uses_default_without_warning() {
+        assert_eq!(resolve_rate_limit(Some(""), 600), (600, false));
+        assert_eq!(resolve_rate_limit(Some("   "), 600), (600, false));
+    }
+
+    #[test]
+    fn valid_value_parses() {
+        assert_eq!(resolve_rate_limit(Some("50"), 600), (50, false));
+        assert_eq!(resolve_rate_limit(Some(" 50 "), 600), (50, false));
+        // 0 is a VALID value (disables the brake); the ==0 warning is handled
+        // separately at the call site. It must NOT be flagged invalid here.
+        assert_eq!(resolve_rate_limit(Some("0"), 600), (0, false));
+    }
+
+    #[test]
+    fn present_but_unparseable_defaults_and_flags_invalid() {
+        // The L8 case: a fat-fingered stricter cap ("5O", letter O) must default
+        // AND be flagged so the caller warns, not silently disable the brake.
+        assert_eq!(resolve_rate_limit(Some("5O"), 600), (600, true));
+        assert_eq!(resolve_rate_limit(Some("abc"), 600), (600, true));
+        assert_eq!(resolve_rate_limit(Some("-1"), 600), (600, true));
+    }
+}
+
 #[cfg(test)]
 mod gossip_ssrf_tests {
     use super::ping_peer_health;

crates/gitlawb-node/src/metrics.rs

@@ -235,11 +235,7 @@ pub fn record_auth_failure(route: &str, reason: &str) {
     }
 }
 
-/// Record one sync_queue item outcome. `status` ∈ {done, failed, skipped,
-/// rejected, deferred}. `rejected` is terminal and caused by a guard (a
-/// malformed slug, or a path resolving outside repos_dir); `deferred` leaves
-/// the row pending for a later pass, so a queue stalled on an operator
-/// condition is visible rather than looking idle.
+/// Record one sync_queue item outcome. `status` ∈ {done, failed, skipped}.
 pub fn record_sync_processed(status: &str) {
     if let Some(c) = SYNC_PROCESSED.get() {
         c.with_label_values(&[status]).inc();

crates/gitlawb-node/src/rate_limit.rs

@@ -70,10 +70,23 @@ impl RateLimiter {
         self.window.min(Duration::from_secs(1))
     }
 
+    /// Returns `true` if the request is allowed. Thin wrapper over
+    /// [`check_retry`](Self::check_retry) for callers that only need the
+    /// allow/deny decision, not the rejection's Retry-After delay. Test-only now:
+    /// production 429 sites use `check_retry` so they can advertise Retry-After.
+    #[cfg(test)]
     pub(crate) async fn check(&self, key: &str) -> bool {
+        self.check_retry(key).await.is_none()
+    }
+
+    /// Record a request against `key`. Returns `None` when it is allowed, or
+    /// `Some(retry_after)` when it is rejected — the delay a client should wait
+    /// before retrying, so a 429 can advertise a value consistent with the
+    /// actual window instead of a constant.
+    pub(crate) async fn check_retry(&self, key: &str) -> Option<Duration> {
         // max_requests == 0 means the limiter is disabled, not "block all".
         if self.max_requests == 0 {
-            return true;
+            return None;
         }
         let now = Instant::now();
         let mut state = self.state.lock().await;
@@ -85,10 +98,25 @@ impl RateLimiter {
                 .timestamps
                 .retain(|t| now.duration_since(*t) < self.window);
             if window.timestamps.len() >= self.max_requests {
-                return false;
+                // Insertion order is preserved, so after the retain the first
+                // element is the oldest live request. It ages out (freeing a
+                // slot) one window after it landed, so that remaining time is
+                // the delay to advertise. Clamp to >= 1 so a 429 never tells a
+                // client to retry immediately.
+                let retry = window
+                    .timestamps
+                    .first()
+                    .map(|oldest| {
+                        self.window
+                            .as_secs()
+                            .saturating_sub(now.duration_since(*oldest).as_secs())
+                    })
+                    .unwrap_or(0)
+                    .max(1);
+                return Some(Duration::from_secs(retry));
             }
             window.timestamps.push(now);
-            return true;
+            return None;
         }
 
         // New key. Enforce the cap BEFORE inserting so a flood of distinct keys
@@ -109,7 +137,11 @@ impl RateLimiter {
             }
             drop(last_sweep);
             if state.len() >= self.max_keys {
-                return false;
+                // The key is untracked (never inserted), so there is no oldest
+                // entry to derive a delay from. Advertise the full window as a
+                // conservative bound: capacity frees as other keys expire, at
+                // the latest one window from now.
+                return Some(self.window.max(Duration::from_secs(1)));
             }
         }
         state.insert(
@@ -118,7 +150,7 @@ impl RateLimiter {
                 timestamps: vec![now],
             },
         );
-        true
+        None
     }
 
     pub async fn cleanup(&self) {
@@ -130,6 +162,41 @@ impl RateLimiter {
             !w.timestamps.is_empty()
         });
     }
+
+    /// Test-only: insert a fully-expired entry (no live timestamps) that the next
+    /// `cleanup()` must reclaim. Models a key whose window has aged out, without
+    /// depending on wall-clock sleeps or a short window. Used cross-module by the
+    /// AppState cleanup guard test (H2).
+    #[cfg(test)]
+    pub(crate) async fn insert_reclaimable_for_test(&self, key: &str) {
+        self.state
+            .lock()
+            .await
+            .insert(key.to_string(), Window { timestamps: vec![] });
+    }
+
+    /// Test-only: number of distinct keys currently tracked.
+    #[cfg(test)]
+    pub(crate) async fn tracked_keys(&self) -> usize {
+        self.state.lock().await.len()
+    }
+
+    /// Test-only: seed a key's bucket with entries of the given ages (how long
+    /// ago each request landed), so a test can drive the Retry-After computation
+    /// off a controlled oldest-entry age without wall-clock sleeps. Ages that
+    /// would underflow the monotonic clock collapse to `now`.
+    #[cfg(test)]
+    pub(crate) async fn seed_aged_entry_for_test(&self, key: &str, ages: &[Duration]) {
+        let now = Instant::now();
+        let timestamps = ages
+            .iter()
+            .map(|a| now.checked_sub(*a).unwrap_or(now))
+            .collect();
+        self.state
+            .lock()
+            .await
+            .insert(key.to_string(), Window { timestamps });
+    }
 }
 
 pub async fn rate_limit_by_did(request: Request, next: Next) -> Response {
@@ -141,13 +208,8 @@ pub async fn rate_limit_by_did(request: Request, next: Next) -> Response {
         .map(|a| a.0.clone());
 
     if let (Some(limiter), Some(did)) = (limiter, did) {
-        if !limiter.check(&did).await {
-            return (
-                StatusCode::TOO_MANY_REQUESTS,
-                [("retry-after", "60")],
-                "rate limit exceeded — try again later",
-            )
-                .into_response();
+        if let Some(retry_after) = limiter.check_retry(&did).await {
+            return too_many_requests(retry_after);
         }
     }
 
@@ -250,13 +312,13 @@ impl<S: Send + Sync> axum::extract::FromRequestParts<S> for PeerAddr {
     }
 }
 
-/// The shared 429 response for the per-IP flood brakes. Route-agnostic: this
-/// middleware now serves the push path AND the peer-sync routes, so the message
-/// stays generic (the offending path is recorded in the warn log below).
-pub fn too_many_requests() -> Response {
+/// The shared 429 response for the rate-limit middlewares (per-DID and per-IP
+/// alike). Route-agnostic: the message stays generic (the offending path, when
+/// logged, is recorded at the call site).
+pub fn too_many_requests(retry_after: Duration) -> Response {
     (
         StatusCode::TOO_MANY_REQUESTS,
-        [("retry-after", "60")],
+        [("retry-after", retry_after.as_secs().to_string())],
         "rate limit exceeded — try again later",
     )
         .into_response()
@@ -274,9 +336,9 @@ pub async fn rate_limit_by_ip(request: Request, next: Next) -> Response {
 
     if let Some(limiter) = limiter {
         if let Some(key) = client_key(request.headers(), peer, limiter.trust) {
-            if !limiter.limiter.check(&key).await {
+            if let Some(retry_after) = limiter.limiter.check_retry(&key).await {
                 tracing::warn!(key = %key, path = %request.uri().path(), "per-IP rate limit exceeded");
-                return too_many_requests();
+                return too_many_requests(retry_after);
             }
         }
     }
@@ -605,4 +667,196 @@ mod tests {
             "an exhausted IP must be braked with 429 before auth runs, not leak to 401"
         );
     }
+
+    // ── Adversarial TrustedProxy verification through the middleware (U5) ──
+    // These complement the client_key unit tests above by driving hostile
+    // headers through rate_limit_by_ip on a real router: the property under test
+    // is that the *bucket key* cannot be rotated by a spoofer.
+
+    async fn post_with(
+        router: &axum::Router,
+        peer: SocketAddr,
+        hdrs: &[(&str, &str)],
+    ) -> StatusCode {
+        use tower::ServiceExt;
+        let mut b = axum::http::Request::builder()
+            .method(axum::http::Method::POST)
+            .uri("/api/v1/repos");
+        for (k, v) in hdrs {
+            b = b.header(*k, *v);
+        }
+        let mut req = b.body(axum::body::Body::empty()).unwrap();
+        req.extensions_mut().insert(ConnectInfo(peer));
+        router.clone().oneshot(req).await.unwrap().status()
+    }
+
+    #[tokio::test]
+    async fn xff_spoofed_leftmost_hop_cannot_rotate_bucket() {
+        // XForwardedFor mode, budget 1. A spoofer varies the client-controlled
+        // leftmost hop every request but the trusted proxy always appends the
+        // same rightmost hop. All requests must key to the rightmost hop, so the
+        // second is braked despite a fresh leftmost value.
+        let router = ip_limited_over_auth_router(IpRateLimiter {
+            limiter: RateLimiter::new(1, Duration::from_secs(60)),
+            trust: TrustedProxy::XForwardedFor,
+        });
+        let p1: SocketAddr = "10.0.0.1:1".parse().unwrap();
+        let p2: SocketAddr = "10.0.0.2:1".parse().unwrap();
+        // First request passes the brake, reaches (failing) auth.
+        assert_eq!(
+            post_with(&router, p1, &[("x-forwarded-for", "9.9.9.1, 203.0.113.5")]).await,
+            StatusCode::UNAUTHORIZED
+        );
+        // Different leftmost + different socket peer, SAME rightmost trusted hop:
+        // braked, because the key is the rightmost hop, not the spoofed value.
+        assert_eq!(
+            post_with(&router, p2, &[("x-forwarded-for", "9.9.9.2, 203.0.113.5")]).await,
+            StatusCode::TOO_MANY_REQUESTS,
+            "a spoofer varying the leftmost XFF hop must not rotate its bucket key"
+        );
+    }
+
+    #[tokio::test]
+    async fn xff_distinct_real_clients_get_distinct_buckets() {
+        // Distinct trusted (rightmost) hops are distinct clients: neither throttles
+        // the other, so a busy legitimate client cannot starve a different one.
+        let router = ip_limited_over_auth_router(IpRateLimiter {
+            limiter: RateLimiter::new(1, Duration::from_secs(60)),
+            trust: TrustedProxy::XForwardedFor,
+        });
+        let peer: SocketAddr = "10.0.0.9:1".parse().unwrap();
+        assert_eq!(
+            post_with(
+                &router,
+                peer,
+                &[("x-forwarded-for", "1.1.1.1, 203.0.113.5")]
+            )
+            .await,
+            StatusCode::UNAUTHORIZED
+        );
+        assert_eq!(
+            post_with(
+                &router,
+                peer,
+                &[("x-forwarded-for", "1.1.1.1, 203.0.113.6")]
+            )
+            .await,
+            StatusCode::UNAUTHORIZED,
+            "a different trusted client hop must get its own bucket"
+        );
+    }
+
+    #[tokio::test]
+    async fn absent_trusted_header_collapses_to_socket_peer() {
+        // The documented fallback: in a trusted-proxy mode, a request with no
+        // forwarded header keys on the socket peer. Behind a real edge every
+        // request shares the proxy's socket, so they collapse onto one bucket —
+        // asserted here by name so any change to this behavior is deliberate.
+        let router = ip_limited_over_auth_router(IpRateLimiter {
+            limiter: RateLimiter::new(1, Duration::from_secs(60)),
+            trust: TrustedProxy::XForwardedFor,
+        });
+        let proxy: SocketAddr = "172.16.0.1:1".parse().unwrap();
+        assert_eq!(
+            post_with(&router, proxy, &[]).await,
+            StatusCode::UNAUTHORIZED
+        );
+        assert_eq!(
+            post_with(&router, proxy, &[]).await,
+            StatusCode::TOO_MANY_REQUESTS,
+            "with no trusted header, requests fall back to the socket peer key (collapse)"
+        );
+    }
+
+    // ── Retry-After derived from the window, not a constant 60 (U5) ───────
+    // A 429 must advertise the delay until the oldest live request ages out of
+    // the limiter's window, so the value tracks the real window instead of the
+    // hardcoded 60 that contradicted every limiter's 3600s window.
+
+    async fn resp_from(router: &axum::Router, peer: SocketAddr) -> Response {
+        use tower::ServiceExt;
+        let mut req = axum::http::Request::builder()
+            .method(axum::http::Method::POST)
+            .uri("/o/r/git-receive-pack")
+            .body(axum::body::Body::empty())
+            .unwrap();
+        req.extensions_mut().insert(ConnectInfo(peer));
+        router.clone().oneshot(req).await.unwrap()
+    }
+
+    fn retry_after_secs(resp: &Response) -> u64 {
+        resp.headers()
+            .get("retry-after")
+            .expect("429 must carry a retry-after header")
+            .to_str()
+            .unwrap()
+            .parse()
+            .expect("retry-after must be an integer number of seconds")
+    }
+
+    #[tokio::test]
+    async fn retry_after_reflects_full_window_for_freshly_filled_bucket() {
+        // A just-filled bucket's oldest entry is ~now, so the advertised delay
+        // must be close to the whole window (100s), NOT the old constant 60.
+        let router = ip_limited_router(IpRateLimiter {
+            limiter: RateLimiter::new(1, Duration::from_secs(100)),
+            trust: TrustedProxy::None,
+        });
+        let peer: SocketAddr = "203.0.113.50:6000".parse().unwrap();
+        assert_eq!(resp_from(&router, peer).await.status(), StatusCode::OK);
+        let resp = resp_from(&router, peer).await;
+        assert_eq!(resp.status(), StatusCode::TOO_MANY_REQUESTS);
+        let retry = retry_after_secs(&resp);
+        assert!(
+            (95..=100).contains(&retry),
+            "freshly-filled bucket must advertise ~window (95..=100s), got {retry} (constant-60 bug if 60)"
+        );
+    }
+
+    #[tokio::test]
+    async fn retry_after_shrinks_as_oldest_entry_ages() {
+        // Seed the sole slot with an entry already 95s into a 100s window: the
+        // next request is rejected and must advertise only the ~5s remaining,
+        // not 60.
+        let limiter = RateLimiter::new(1, Duration::from_secs(100));
+        let peer: SocketAddr = "203.0.113.51:6000".parse().unwrap();
+        limiter
+            .seed_aged_entry_for_test(&peer.ip().to_string(), &[Duration::from_secs(95)])
+            .await;
+        let router = ip_limited_router(IpRateLimiter {
+            limiter,
+            trust: TrustedProxy::None,
+        });
+        let resp = resp_from(&router, peer).await;
+        assert_eq!(resp.status(), StatusCode::TOO_MANY_REQUESTS);
+        let retry = retry_after_secs(&resp);
+        assert!(
+            (1..=10).contains(&retry),
+            "an aged bucket must advertise the small remaining delay (1..=10s), got {retry}"
+        );
+    }
+
+    #[tokio::test]
+    async fn retry_after_clamps_to_at_least_one_second() {
+        // Oldest entry is 99s into a 100s window: the raw remainder rounds toward
+        // 1s and must never be advertised as 0 (a 0/blank retry lets a rejected
+        // client hammer immediately).
+        let limiter = RateLimiter::new(1, Duration::from_secs(100));
+        let peer: SocketAddr = "203.0.113.52:6000".parse().unwrap();
+        limiter
+            .seed_aged_entry_for_test(&peer.ip().to_string(), &[Duration::from_secs(99)])
+            .await;
+        let router = ip_limited_router(IpRateLimiter {
+            limiter,
+            trust: TrustedProxy::None,
+        });
+        let resp = resp_from(&router, peer).await;
+        assert_eq!(resp.status(), StatusCode::TOO_MANY_REQUESTS);
+        let retry = retry_after_secs(&resp);
+        assert!(retry >= 1, "retry-after must clamp to >= 1, got {retry}");
+        assert!(
+            retry <= 2,
+            "with 99s of a 100s window elapsed, the remaining delay is ~1s, got {retry}"
+        );
+    }
 }

crates/gitlawb-node/src/server.rs

@@ -14,7 +14,7 @@ use tracing::Level;
 
 use crate::api::{
     agents, arweave, bounties, certs, changelog, events, ipfs, issues, labels, peers, profiles,
-    protect, pulls, register, replicas, repos, resolve, stars, status, tasks, visibility, webhooks,
+    protect, pulls, register, replicas, repos, resolve, stars, tasks, visibility, webhooks,
 };
 use crate::auth;
 use crate::rate_limit;
@@ -55,18 +55,50 @@ fn add_auth_layers(router: Router<AppState>, state: AppState) -> Router<AppState
         .layer(middleware::from_fn(auth::require_signature))
 }
 
+/// Attach the shared per-IP write brake to an authenticated write-route group,
+/// outermost so a flood is rejected before signature verification. Every
+/// authenticated non-creation write group MUST call this or it silently ships
+/// unbraked with no compile error. Deliberately a separate call, NOT folded into
+/// `add_auth_layers`: the creation routes use a different bucket
+/// (`create_ip_rate_limiter`) and the graphql group is not built through
+/// `add_auth_layers`, so a helper keyed on auth would attach the wrong limiter.
+trait WriteBraked {
+    fn write_braked(self, limiter: &rate_limit::IpRateLimiter) -> Self;
+}
+
+impl WriteBraked for Router<AppState> {
+    fn write_braked(self, limiter: &rate_limit::IpRateLimiter) -> Self {
+        self.layer(middleware::from_fn(rate_limit::rate_limit_by_ip))
+            .layer(axum::Extension(limiter.clone()))
+    }
+}
+
 pub fn build_router(state: AppState) -> Router {
     // ── GraphQL routes ─────────────────────────────────────────────────────
     let schema = state.graphql_schema.as_ref().clone();
+
+    // Per-IP write brake, shared across every authenticated non-creation write
+    // group (see `AppState::write_rate_limiter`). Built once here so each group
+    // clones the same bucket; attached outermost on each group so a flood is
+    // rejected before auth runs. Separate bucket from the creation brake.
+    let write_ip_limiter = rate_limit::IpRateLimiter {
+        limiter: state.write_rate_limiter.clone(),
+        trust: state.push_limiter_trust,
+    };
+
     let graphql_routes = Router::new()
         .route("/graphql", get(graphql_playground).post(graphql_handler))
         // Attach the verified DID to /graphql when a signature is present. The
         // layer covers only routes added before it, so /graphql/ws (added after,
-        // read-only subscriptions) stays open.
+        // read-only subscriptions) stays open. The per-IP write brake covers the
+        // same /graphql (KTD-5: an HTTP-layer brake cannot tell a query POST from
+        // a mutation POST, so it counts every /graphql request — acceptable, both
+        // cost work); /graphql/ws subscriptions stay unbraked.
         .layer(middleware::from_fn(auth::optional_signature))
+        .write_braked(&write_ip_limiter)
         .route_service("/graphql/ws", GraphQLSubscription::new(schema));
 
-    // ── Task routes (write — require HTTP Signature) ───────────────────────
+    // ── Task routes (write — require HTTP Signature + per-IP write brake) ──
     let task_write_routes = add_auth_layers(
         Router::new()
             .route("/api/v1/tasks", post(tasks::create_task))
@@ -74,7 +106,8 @@ pub fn build_router(state: AppState) -> Router {
             .route("/api/v1/tasks/{id}/complete", post(tasks::complete_task))
             .route("/api/v1/tasks/{id}/fail", post(tasks::fail_task)),
         state.clone(),
-    );
+    )
+    .write_braked(&write_ip_limiter);
 
     // ── Task routes (read — open) ──────────────────────────────────────────
     let task_read_routes = Router::new()
@@ -111,7 +144,13 @@ pub fn build_router(state: AppState) -> Router {
     .layer(middleware::from_fn(rate_limit::rate_limit_by_ip))
     .layer(axum::Extension(create_ip_limiter));
 
-    // ── Write routes — require HTTP Signature (no rate limit) ─────────────
+    // ── Write routes — require HTTP Signature + per-IP write brake ────────
+    // Same per-IP flood defense as the creation routes, on its own `write`
+    // bucket (see `AppState::write_rate_limiter`). Per-DID is deliberately not
+    // paired (a DID farm never trips it; busy legitimate agents would false-
+    // positive). The bundled visibility GET is a read that also draws from the
+    // write bucket — harmless, since the bucket is sized far above any legit
+    // per-IP read rate.
     let write_routes = add_auth_layers(
         Router::new()
             .route(
@@ -181,65 +220,8 @@ pub fn build_router(state: AppState) -> Router {
                 axum::routing::delete(agents::deregister_agent),
             ),
         state.clone(),
-    );
-
-    // ── Status claim writes — signature, plus the per-DID throttle AND the
-    // per-IP flood brake `creation_routes` carries. Not in `write_routes`, which
-    // has neither: every call here appends a row, so an unthrottled writer grows
-    // the claim log on demand and the three per-repo caps would be the only bound.
-    let status_write_routes = add_auth_layers(
-        Router::new()
-            .route(
-                "/api/v1/repos/{owner}/{repo}/statuses/{sha}",
-                post(status::create_status),
-            )
-            .layer(middleware::from_fn(rate_limit::rate_limit_by_did))
-            .layer(axum::Extension(state.rate_limiter.clone())),
-        state.clone(),
     )
-    // The route's advertised body bound, applied at the transport. It sits
-    // OUTSIDE the auth layers because `require_signature` collects and hashes
-    // the whole body before any handler runs: the handler's own check measures
-    // an allocation that has already happened, so without this layer the 8 KiB
-    // figure bounded what gets stored and not what gets buffered, and
-    // concurrent requests multiplied exactly the cost it was written to
-    // prevent. It sits INSIDE the rate limiters so an oversized request is
-    // still counted by the per-IP brake before it is refused.
-    //
-    // A declared Content-Length over the limit is answered 413 without reading
-    // a byte; a body that declares nothing (or understates) is truncated at the
-    // limit and fails where it is read, which `require_signature` reports as an
-    // unreadable body.
-    .layer(RequestBodyLimitLayer::new(status::MAX_REQUEST_BODY_BYTES))
-    // The one route that persists the signed request body, so the one route
-    // whose signature material carries it. Applied OUTSIDE `add_auth_layers`
-    // (outermost = runs first) because `require_signature` reads this marker to
-    // decide whether to carry the body; a layer added inside the auth pair runs
-    // after the middleware and is never seen. `create_status` refuses an absent
-    // body rather than storing an empty column, so getting this order wrong
-    // fails loudly instead of quietly writing unverifiable claims.
-    .layer(axum::Extension(auth::PersistsSignedBody))
-    .layer(middleware::from_fn(rate_limit::rate_limit_by_ip))
-    .layer(axum::Extension(rate_limit::IpRateLimiter {
-        limiter: state.create_ip_rate_limiter.clone(),
-        trust: state.push_limiter_trust,
-    }));
-
-    // ── Status claim reads — open, behind `optional_signature` so the handler
-    // sees the caller when one signs and still answers a public repo for an
-    // anonymous reader. Without the layer the handler's optional auth extension
-    // is None for EVERY caller, including the owner, and every private repo
-    // reads as not-found.
-    let status_read_routes = Router::new()
-        .route(
-            "/api/v1/repos/{owner}/{repo}/commits/{sha}/status",
-            get(status::commit_status),
-        )
-        .route(
-            "/api/v1/repos/{owner}/{repo}/pulls/{number}/status",
-            get(status::pull_request_status),
-        )
-        .layer(middleware::from_fn(auth::optional_signature));
+    .write_braked(&write_ip_limiter);
 
     // Body limit is raised to GITLAWB_MAX_PACK_BYTES (default 2 GB) for git
     // routes only — all other API routes keep axum's default 2 MB cap.
@@ -305,7 +287,8 @@ pub fn build_router(state: AppState) -> Router {
                 post(bounties::dispute_bounty),
             ),
         state.clone(),
-    );
+    )
+    .write_braked(&write_ip_limiter);
 
     // ── Bounty routes (read — open) ──────────────────────────────────────
     let bounty_read_routes = Router::new()
@@ -326,9 +309,10 @@ pub fn build_router(state: AppState) -> Router {
     let profile_write_routes = add_auth_layers(
         Router::new().route("/api/v1/profile", axum::routing::put(profiles::set_profile)),
         state.clone(),
-    );
+    )
+    .write_braked(&write_ip_limiter);
 
-    // ── Issue routes (write — require HTTP Signature, no rate limit) ─────
+    // ── Issue routes (write — require HTTP Signature + per-IP write brake) ─
     let issue_write_routes = add_auth_layers(
         Router::new()
             .route(
@@ -340,7 +324,8 @@ pub fn build_router(state: AppState) -> Router {
                 post(issues::create_issue_comment),
             ),
         state.clone(),
-    );
+    )
+    .write_braked(&write_ip_limiter);
 
     // ── Peer discovery routes ─────────────────────────────────────────────
     // Peer writes accept signatures when present and can require them after a
@@ -431,10 +416,6 @@ pub fn build_router(state: AppState) -> Router {
             "/api/v1/repos/{owner}/{repo}/events",
             get(events::list_repo_events),
         )
-        .route(
-            "/api/v1/repos/{owner}/{repo}/push-events",
-            get(events::list_repo_push_events),
-        )
         .route("/api/v1/agents", get(agents::list_agents))
         .route("/api/v1/agents/{did}", get(agents::show_agent))
         .route("/api/v1/agents/{did}/trust", get(agents::get_trust))
@@ -531,8 +512,6 @@ pub fn build_router(state: AppState) -> Router {
         .merge(profile_write_routes)
         .merge(creation_routes)
         .merge(write_routes)
-        .merge(status_write_routes)
-        .merge(status_read_routes)
         .merge(git_write_routes)
         .merge(git_read_routes)
         .merge(issue_write_routes)

crates/gitlawb-node/src/state.rs

@@ -62,6 +62,16 @@ pub struct AppState {
     /// resolved client IP is what actually stops a single-source flood (same
     /// rationale as `push_rate_limiter`). Keyed by `push_limiter_trust`.
     pub create_ip_rate_limiter: RateLimiter,
+    /// Per-client-IP rate limiter for the authenticated write surface that is
+    /// not repo/agent creation: issue/PR comments, labels, stars, merges,
+    /// protect/unprotect, replicas, visibility, tasks, bounties, profile, and
+    /// the GraphQL `MutationRoot`. Separate bucket from `create_ip_rate_limiter`
+    /// (its own budget, per the `sync_trigger`/`peer_write` precedent) so a
+    /// write flood cannot drain the creation budget and vice versa. Per-DID is
+    /// deliberately NOT paired here: a DID farm never trips it, and busy
+    /// legitimate agents would false-positive. `GITLAWB_WRITE_RATE_LIMIT`
+    /// overrides the default, 0 disables. Keyed by `push_limiter_trust`.
+    pub write_rate_limiter: RateLimiter,
     /// Per-client-IP rate limiter for git-receive-pack. Per-DID limits cannot
     /// brake a push flood from a DID farm (one throwaway DID per repo), so the
     /// push path throttles on the resolved client IP instead.
@@ -118,4 +128,48 @@ impl AppState {
     pub fn is_shutting_down(&self) -> bool {
         *self.shutdown_tx.borrow()
     }
+
+    /// Reclaim expired entries from EVERY rate limiter. Single source of truth for
+    /// the periodic cleanup loop (main.rs), co-located with the limiter fields so a
+    /// newly-added limiter is cleaned here rather than silently omitted from a
+    /// hand-maintained list in main() (H2: `write_rate_limiter` was the omitted
+    /// one). Every limiter field above MUST be swept here; the guard test
+    /// `cleanup_rate_limiters_reaps_the_write_limiter` fails if it is not.
+    pub(crate) async fn cleanup_rate_limiters(&self) {
+        self.rate_limiter.cleanup().await;
+        self.create_ip_rate_limiter.cleanup().await;
+        self.write_rate_limiter.cleanup().await;
+        self.push_rate_limiter.cleanup().await;
+        self.sync_trigger_rate_limiter.cleanup().await;
+        self.peer_write_rate_limiter.cleanup().await;
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    /// H2 guard: the write-surface limiter must be reaped by the shared cleanup
+    /// routine. Seeds a reclaimable entry into `write_rate_limiter` and asserts
+    /// `cleanup_rate_limiters` removes it. Goes RED if `write_rate_limiter` is
+    /// dropped from `cleanup_rate_limiters` — the exact H2 omission, now guarded.
+    #[tokio::test]
+    async fn cleanup_rate_limiters_reaps_the_write_limiter() {
+        let state = crate::test_support::test_state_lazy();
+        state
+            .write_rate_limiter
+            .insert_reclaimable_for_test("some-ip")
+            .await;
+        assert_eq!(
+            state.write_rate_limiter.tracked_keys().await,
+            1,
+            "precondition: the reclaimable entry is tracked"
+        );
+
+        state.cleanup_rate_limiters().await;
+
+        assert_eq!(
+            state.write_rate_limiter.tracked_keys().await,
+            0,
+            "write_rate_limiter must be reaped by cleanup_rate_limiters (H2)"
+        );
+    }
 }

crates/gitlawb-node/src/test_support.rs

@@ -77,6 +77,7 @@ fn build_state(db: Arc<crate::db::Db>, pool: PgPool) -> AppState {
         repo_store: crate::git::repo_store::RepoStore::for_testing(PathBuf::from("/tmp"), pool),
         rate_limiter: RateLimiter::new(100, Duration::from_secs(60)),
         create_ip_rate_limiter: RateLimiter::new(1000, Duration::from_secs(3600)),
+        write_rate_limiter: RateLimiter::new(1000, Duration::from_secs(3600)),
         push_rate_limiter: RateLimiter::new(600, Duration::from_secs(3600)),
         push_limiter_trust: crate::rate_limit::TrustedProxy::None,
         sync_trigger_rate_limiter: RateLimiter::new(60, Duration::from_secs(3600)),
@@ -109,153 +110,6 @@ pub(crate) fn signed_request_as(did: &str, method: Method, uri: &str, body: Body
         .expect("request builder")
 }
 
-/// Slice a region out of a module's own source (obtained with `include_str!`)
-/// and drop full-line `//` comments from it.
-///
-/// Several guards in `api/` assert that a piece of PRODUCTION code does or does
-/// not contain something: that the receive-pack handler really calls the
-/// push-event recorder, that the status module never acquires the repository.
-/// Each guard carried its own copy of this parse, and three copies of a parser
-/// is three places for the slicing to drift apart while every guard still reads
-/// green.
-///
-/// `start` and `end` are literal anchors. The region runs from the first
-/// occurrence of `start` (inclusive; the top of the file when `None`) to the
-/// next occurrence of `end` (exclusive; end of file when `None`).
-///
-/// Returns `None` when an anchor is missing, rather than silently falling back
-/// to a larger or empty region. Each caller supplies its own message for that
-/// case and keeps its own assertion that the region really covers what it claims
-/// to, which is what stops a bug in this one helper from turning every guard
-/// that uses it into a vacuous pass.
-///
-/// Comment lines are stripped so that a doc comment naming the very call a guard
-/// searches for cannot stand in for the call itself.
-pub(crate) fn scrape_source_region(
-    src: &str,
-    start: Option<&str>,
-    end: Option<&str>,
-) -> Option<String> {
-    let from = match start {
-        Some(anchor) => src.find(anchor)?,
-        None => 0,
-    };
-    let rest = &src[from..];
-    let to = match end {
-        // Search past the first CHARACTER, so an `end` anchor that also matches
-        // at the very start of the region cannot collapse it to nothing. A fixed
-        // one-byte skip does the same job on ASCII and lands mid-character on a
-        // multibyte first character, where the slice fails and the helper answers
-        // `None`, which is indistinguishable, to every caller, from a missing
-        // anchor.
-        Some(anchor) => {
-            let skip = rest.chars().next().map_or(0, char::len_utf8);
-            rest.get(skip..)?.find(anchor)? + skip
-        }
-        None => rest.len(),
-    };
-    Some(
-        rest[..to]
-            .lines()
-            .filter(|l| !l.trim_start().starts_with("//"))
-            .collect::<Vec<_>>()
-            .join("\n"),
-    )
-}
-
-#[cfg(test)]
-mod scrape_source_region_tests {
-    use super::scrape_source_region;
-
-    const SRC: &str = "prelude line\nfn target() {\n    // calls_the_thing()\n    calls_the_thing();\n}\nfn later() {\n    calls_the_thing();\n}\n";
-
-    #[test]
-    fn stops_at_the_end_anchor_so_a_later_function_cannot_satisfy_a_guard() {
-        let body = scrape_source_region(SRC, Some("fn target()"), Some("\n}")).unwrap();
-        assert!(body.contains("calls_the_thing();"));
-        assert!(
-            !body.contains("fn later()"),
-            "the region must stop at the first column-zero brace, got: {body}"
-        );
-    }
-
-    #[test]
-    fn strips_full_line_comments_so_a_doc_comment_cannot_stand_in_for_code() {
-        let commented = "fn target() {\n    // calls_the_thing();\n}\n";
-        let body = scrape_source_region(commented, Some("fn target()"), Some("\n}")).unwrap();
-        assert!(
-            !body.contains("calls_the_thing"),
-            "a commented-out call must not survive the scrape, got: {body}"
-        );
-    }
-
-    #[test]
-    fn no_start_anchor_reads_from_the_top_of_the_file() {
-        let body = scrape_source_region(SRC, None, Some("fn later()")).unwrap();
-        assert!(body.starts_with("prelude line"));
-        assert!(!body.contains("fn later()"));
-    }
-
-    #[test]
-    fn no_end_anchor_reads_to_the_end_of_the_file() {
-        let body = scrape_source_region(SRC, Some("fn later()"), None).unwrap();
-        assert!(body.trim_end().ends_with('}'));
-    }
-
-    /// A missing anchor is None, never a silently wider or empty region: that is
-    /// what lets each caller keep its own "the thing I meant to scan is gone"
-    /// message instead of asserting against whatever the fallback produced.
-    #[test]
-    fn a_missing_anchor_is_none_rather_than_a_fallback_region() {
-        assert!(scrape_source_region(SRC, Some("fn absent()"), Some("\n}")).is_none());
-        assert!(scrape_source_region(SRC, Some("fn target()"), Some("~absent~")).is_none());
-    }
-
-    /// A region whose first character is multibyte is scraped, not reported
-    /// missing.
-    ///
-    /// The end-anchor search skips the region's first CHARACTER. Skipping a
-    /// fixed byte instead lands mid-character here, the slice fails, and the
-    /// helper answers `None`, which every caller reads as "the anchor is gone"
-    /// and reports as a missing region. A guard that says "not found"
-    /// for input that is present is worse than one that fails loudly, and the
-    /// five callers sharing this helper all inherit it.
-    #[test]
-    fn a_multibyte_first_character_does_not_read_as_a_missing_anchor() {
-        let src = "é prelude\nfn target() {\n    body();\n}\nfn later() {}\n";
-
-        let body = scrape_source_region(src, None, Some("\n}"))
-            .expect("a region starting on a multibyte character must be scraped");
-        assert!(
-            body.starts_with("é prelude") && body.contains("body();"),
-            "the region must cover the multibyte start through the end anchor, got: {body}"
-        );
-        assert!(
-            !body.contains("fn later()"),
-            "the region must still stop at the end anchor, got: {body}"
-        );
-
-        // The same through a multibyte START anchor, so the skip is measured
-        // from the region rather than from the top of the file.
-        let body = scrape_source_region(src, Some("é"), Some("\n}"))
-            .expect("a multibyte start anchor must be scraped");
-        assert!(body.starts_with('é'), "got: {body}");
-    }
-
-    /// The end anchor is searched past the region's first character, so a
-    /// region whose own start also matches the end anchor is not collapsed to
-    /// nothing.
-    #[test]
-    fn an_end_anchor_matching_at_the_start_does_not_collapse_the_region() {
-        let src = "\n}\nfn target() {\n    body();\n}\n";
-        let body = scrape_source_region(src, None, Some("\n}")).unwrap();
-        assert!(
-            body.contains("fn target()") && body.contains("body();"),
-            "got: {body}"
-        );
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use super::*;

crates/gl/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "gl"
 description = "The gitlawb CLI — identity management, node control, MCP server"
-version = "0.7.1" # x-release-please-version
+version = "0.6.0" # x-release-please-version
 edition.workspace = true
 rust-version.workspace = true
 license.workspace = true

crates/icaptcha-client/src/lib.rs

@@ -550,174 +550,4 @@ mod tests {
         );
         assert!(key_trusted, "key stays trusted with operator origin");
     }
-
-    #[test]
-    fn new_withholds_key_from_untrusted_origin_but_attaches_for_operator() {
-        let guard = ICAPTCHA_ENV_LOCK.lock().unwrap();
-        let prev_key = std::env::var_os("GITLAWB_ICAPTCHA_API_KEY");
-        let prev_op = std::env::var_os("GITLAWB_ICAPTCHA_URL");
-        std::env::set_var("GITLAWB_ICAPTCHA_API_KEY", "secret-bearer");
-
-        // Arm 1: No operator configured, node advertises an attacker origin.
-        // The key has no trusted destination; the advert is rejected.
-        std::env::remove_var("GITLAWB_ICAPTCHA_URL");
-        let untrusted =
-            IcaptchaCfg::new("did:key:zTEST", Some("https://evil.example".into()), None);
-
-        // Arm 1b: No operator configured, no URL advertised. The fallback goes
-        // to the default public origin, and with no operator to trust, the key
-        // must not ride.
-        let no_operator_no_advert = IcaptchaCfg::new("did:key:zTEST", None, None);
-
-        // Arm 2: Operator origin configured and advertised — key attached.
-        std::env::set_var("GITLAWB_ICAPTCHA_URL", "https://icap.mynode.example");
-        let operator_match = IcaptchaCfg::new(
-            "did:key:zTEST",
-            Some("https://icap.mynode.example/v1".into()),
-            None,
-        );
-
-        // Arm 3: Operator configured, default host advertised. The default host
-        // is allowlisted and selected, but its origin differs from the
-        // operator's, so key must not ride.
-        let default_advert = IcaptchaCfg::new(
-            "did:key:zTEST",
-            Some("https://icaptcha.gitlawb.com/v2".into()),
-            None,
-        );
-
-        // Arm 4: Operator configured, no URL advertised (the node sent only
-        // x-icaptcha-level, not x-icaptcha-url). resolve_solver_url falls back
-        // to the operator origin and marks the key trusted. This is the common
-        // API-key deployment path — a regression that additionally required
-        // url.is_some() for key attachment would leave all current tests green
-        // but make these fallback challenge/answer requests omit Authorization
-        // and fail against the protected operator service.
-        let operator_fallback = IcaptchaCfg::new("did:key:zTEST", None, Some(3));
-
-        // Arm 5: Operator configured, node advertises an attacker origin.
-        // The advert is rejected and the solver falls back to the operator's
-        // own origin, which keeps the key trusted — this is the exfiltration
-        // scenario the test name describes (withhold key from *untrusted*
-        // origin, not from the operator's fallback).
-        let operator_hostile_advert =
-            IcaptchaCfg::new("did:key:zTEST", Some("https://evil.example".into()), None);
-
-        // Arm 6: Operator configured, same host but different port.
-        // origin_key includes the port, so :8443 must not match the
-        // operator origin at the default :443 — the advert is rejected
-        // and the solver falls back to the operator origin.
-        let operator_diff_port = IcaptchaCfg::new(
-            "did:key:zTEST",
-            Some("https://icap.mynode.example:8443/evil".into()),
-            None,
-        );
-
-        // Restore env and release the lock BEFORE asserting, so a failing
-        // assertion can't poison the shared lock or cascade into a sibling.
-        match prev_key {
-            Some(v) => std::env::set_var("GITLAWB_ICAPTCHA_API_KEY", v),
-            None => std::env::remove_var("GITLAWB_ICAPTCHA_API_KEY"),
-        }
-        match prev_op {
-            Some(v) => std::env::set_var("GITLAWB_ICAPTCHA_URL", v),
-            None => std::env::remove_var("GITLAWB_ICAPTCHA_URL"),
-        }
-        drop(guard);
-
-        let mut failures: Vec<String> = Vec::new();
-
-        if untrusted.api_key.is_some() {
-            failures.push(format!(
-                "arm 1 (no operator, attacker advert): expected api_key=None, got {:?}",
-                untrusted.api_key
-            ));
-        }
-        if untrusted.url != DEFAULT_URL {
-            failures.push(format!(
-                "arm 1 (no operator, attacker advert): expected url={DEFAULT_URL}, got {}",
-                untrusted.url
-            ));
-        }
-
-        if no_operator_no_advert.api_key.is_some() {
-            failures.push(format!(
-                "arm 1b (no operator, no advert): expected api_key=None, got {:?}",
-                no_operator_no_advert.api_key
-            ));
-        }
-        if no_operator_no_advert.url != DEFAULT_URL {
-            failures.push(format!(
-                "arm 1b (no operator, no advert): expected url={DEFAULT_URL}, got {}",
-                no_operator_no_advert.url
-            ));
-        }
-
-        if operator_match.api_key.as_deref() != Some("secret-bearer") {
-            failures.push(format!(
-                "arm 2 (operator match): expected api_key=Some(\"secret-bearer\"), got {:?}",
-                operator_match.api_key
-            ));
-        }
-        if operator_match.url != "https://icap.mynode.example/v1" {
-            failures.push(format!(
-                "arm 2 (operator match): expected url=https://icap.mynode.example/v1, got {}",
-                operator_match.url
-            ));
-        }
-
-        if default_advert.api_key.is_some() {
-            failures.push(format!(
-                "arm 3 (operator + default advert): expected api_key=None, got {:?}",
-                default_advert.api_key
-            ));
-        }
-        if default_advert.url != "https://icaptcha.gitlawb.com/v2" {
-            failures.push(format!(
-                "arm 3 (operator + default advert): expected url=https://icaptcha.gitlawb.com/v2, got {}",
-                default_advert.url
-            ));
-        }
-
-        if operator_fallback.api_key.as_deref() != Some("secret-bearer") {
-            failures.push(format!(
-                "arm 4 (operator fallback, no advert): expected api_key=Some(\"secret-bearer\"), got {:?}",
-                operator_fallback.api_key
-            ));
-        }
-        if operator_fallback.url != "https://icap.mynode.example" {
-            failures.push(format!(
-                "arm 4 (operator fallback, no advert): expected url=https://icap.mynode.example, got {}",
-                operator_fallback.url
-            ));
-        }
-
-        if operator_hostile_advert.api_key.as_deref() != Some("secret-bearer") {
-            failures.push(format!(
-                "arm 5 (operator + hostile advert): expected api_key=Some(\"secret-bearer\"), got {:?}",
-                operator_hostile_advert.api_key
-            ));
-        }
-        if operator_hostile_advert.url != "https://icap.mynode.example" {
-            failures.push(format!(
-                "arm 5 (operator + hostile advert): expected url=https://icap.mynode.example, got {}",
-                operator_hostile_advert.url
-            ));
-        }
-
-        if operator_diff_port.api_key.as_deref() != Some("secret-bearer") {
-            failures.push(format!(
-                "arm 6 (operator + different port advert): expected api_key=Some(\"secret-bearer\"), got {:?}",
-                operator_diff_port.api_key
-            ));
-        }
-        if operator_diff_port.url != "https://icap.mynode.example" {
-            failures.push(format!(
-                "arm 6 (operator + different port advert): expected url=https://icap.mynode.example, got {}",
-                operator_diff_port.url
-            ));
-        }
-
-        assert!(failures.is_empty(), "\n{}", failures.join("\n"));
-    }
 }

infra/fly/gitlawb-node-2.fly.toml

@@ -1,65 +0,0 @@
-# fly.toml app configuration file generated for gitlawb-node-2 on 2026-07-22T09:47:18+08:00
-#
-# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
-#
-
-app = 'gitlawb-node-2'
-primary_region = 'sjc'
-
-[build]
-  dockerfile = '../../Dockerfile'
-
-[env]
-  GITLAWB_AUTO_SYNC = 'true'
-  GITLAWB_BOOTSTRAP_PEERS = 'https://node.gitlawb.com,https://node3.gitlawb.com'
-  GITLAWB_HOST = '0.0.0.0'
-  GITLAWB_KEY = '/data/keys/identity.pem'
-  GITLAWB_MAX_PACK_BYTES = '524288000'
-  GITLAWB_P2P_PORT = '7546'
-  GITLAWB_PORT = '7545'
-  GITLAWB_PUBLIC_URL = 'https://node2.gitlawb.com'
-  GITLAWB_REPOS_DIR = '/data/repos'
-  ICAPTCHA_MODE = 'enforce'
-  ICAPTCHA_PUBKEY = 'xjyPNqIbvc9U-kwXW6u9mDqRJ7E2UUMOaJdUWhpEXq8'
-  ICAPTCHA_REQUIRED_LEVEL = '3'
-
-[[mounts]]
-  source = 'gitlawb_data'
-  destination = '/data'
-
-[http_service]
-  internal_port = 7545
-  force_https = true
-  auto_stop_machines = 'off'
-  auto_start_machines = true
-  min_machines_running = 1
-
-  [http_service.concurrency]
-    type = 'connections'
-    hard_limit = 500
-    soft_limit = 400
-
-  [http_service.http_options]
-    idle_timeout = 120
-
-  [[http_service.checks]]
-    interval = '15s'
-    timeout = '5s'
-    grace_period = '30s'
-    method = 'GET'
-    path = '/ready'
-
-[[services]]
-  protocol = 'udp'
-  internal_port = 7546
-
-  [[services.ports]]
-    port = 7546
-
-[[vm]]
-  size = 'shared-cpu-1x'
-  memory = '1024'
-
-[[metrics]]
-  port = 9091
-  path = '/metrics'

infra/fly/gitlawb-node-3.fly.toml

@@ -1,65 +0,0 @@
-# fly.toml app configuration file generated for gitlawb-node-3 on 2026-07-22T09:47:19+08:00
-#
-# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
-#
-
-app = 'gitlawb-node-3'
-primary_region = 'nrt'
-
-[build]
-  dockerfile = '../../Dockerfile'
-
-[env]
-  GITLAWB_AUTO_SYNC = 'true'
-  GITLAWB_BOOTSTRAP_PEERS = 'https://node.gitlawb.com,https://node2.gitlawb.com'
-  GITLAWB_HOST = '0.0.0.0'
-  GITLAWB_KEY = '/data/keys/identity.pem'
-  GITLAWB_MAX_PACK_BYTES = '524288000'
-  GITLAWB_P2P_PORT = '7546'
-  GITLAWB_PORT = '7545'
-  GITLAWB_PUBLIC_URL = 'https://node3.gitlawb.com'
-  GITLAWB_REPOS_DIR = '/data/repos'
-  ICAPTCHA_MODE = 'enforce'
-  ICAPTCHA_PUBKEY = 'xjyPNqIbvc9U-kwXW6u9mDqRJ7E2UUMOaJdUWhpEXq8'
-  ICAPTCHA_REQUIRED_LEVEL = '3'
-
-[[mounts]]
-  source = 'gitlawb_data'
-  destination = '/data'
-
-[http_service]
-  internal_port = 7545
-  force_https = true
-  auto_stop_machines = 'off'
-  auto_start_machines = true
-  min_machines_running = 1
-
-  [http_service.concurrency]
-    type = 'connections'
-    hard_limit = 500
-    soft_limit = 400
-
-  [http_service.http_options]
-    idle_timeout = 120
-
-  [[http_service.checks]]
-    interval = '15s'
-    timeout = '5s'
-    grace_period = '30s'
-    method = 'GET'
-    path = '/ready'
-
-[[services]]
-  protocol = 'udp'
-  internal_port = 7546
-
-  [[services.ports]]
-    port = 7546
-
-[[vm]]
-  size = 'shared-cpu-1x'
-  memory = '1024'
-
-[[metrics]]
-  port = 9091
-  path = '/metrics'

scripts/build-bins.sh

@@ -46,7 +46,7 @@ build_darwin_arm64() {
   echo "==> Building darwin-arm64 (native cargo)..."
   cd "$ROOT"
   export PATH="$HOME/.cargo/bin:$PATH"
-  cargo build --release --locked -p gl -p git-remote-gitlawb
+  cargo build --release -p gl -p git-remote-gitlawb
   cp target/release/gl "$OUT/gl-darwin-arm64"
   cp target/release/git-remote-gitlawb "$OUT/git-remote-gitlawb-darwin-arm64"
   chmod +x "$OUT/gl-darwin-arm64" "$OUT/git-remote-gitlawb-darwin-arm64"

scripts/check-gitlawb-core-deps.sh

@@ -1,75 +0,0 @@
-#!/usr/bin/env bash
-#
-# Dependency-purity gate for the `gitlawb-core` crate.
-#
-# gitlawb-core is embedded by every consumer (the gl CLI, git-remote-gitlawb,
-# the node daemon), so it must stay lean. This script recomputes gitlawb-core's
-# NORMAL (non-dev, non-build) transitive dependency set and fails if it contains
-# any crate not present in ci/gitlawb-core-allowed-deps.txt.
-#
-# Hard-fail direction: a crate present now but NOT allowlisted. That is the case
-# the gate exists to catch (core silently gaining a heavy dependency).
-# Informational only: an allowlisted crate no longer present (stale entry) — a
-# legitimate dependency removal should not red CI, so it is reported, not failed.
-#
-# Runnable from anywhere in the repo.
-set -euo pipefail
-
-ROOT="$(git rev-parse --show-toplevel)"
-ALLOW="$ROOT/ci/gitlawb-core-allowed-deps.txt"
-
-if [ ! -f "$ALLOW" ]; then
-  echo "ERROR: allowlist not found at $ALLOW" >&2
-  exit 1
-fi
-
-# `cargo tree` resolves WITHOUT --locked on purpose, unlike the blocking cargo
-# jobs in pr-checks. This gate answers one question, whether gitlawb-core's
-# dependency closure gained something unallowlisted, and a stale Cargo.lock
-# would red it for a reason that has nothing to do with that closure. Lock
-# staleness is already caught by the --locked cargo jobs, so it does not need a
-# second gate here.
-#
-# Resolving can refresh Cargo.lock as a side effect, so snapshot and restore
-# it: this check must never leave the working tree dirty when run locally.
-lock_backup="$(mktemp)"
-cp "$ROOT/Cargo.lock" "$lock_backup"
-restore_lock() { cp "$lock_backup" "$ROOT/Cargo.lock"; rm -f "$lock_backup"; }
-trap restore_lock EXIT
-
-# Current normal-dependency closure of gitlawb-core, one crate name per line.
-# Must match the regen command documented in the allowlist header exactly.
-current="$(
-  cargo tree -p gitlawb-core --edges normal --prefix none --manifest-path "$ROOT/Cargo.toml" \
-    | sed -E 's/ v[0-9].*$//' \
-    | grep -v '^gitlawb-core$' \
-    | sort -u
-)"
-
-# Allowlist with comments and blank lines stripped.
-allowed="$(grep -vE '^[[:space:]]*(#|$)' "$ALLOW" | sort -u)"
-
-# comm needs sorted input; both sides are sorted above.
-offenders="$(comm -23 <(printf '%s\n' "$current") <(printf '%s\n' "$allowed"))"
-stale="$(comm -13 <(printf '%s\n' "$current") <(printf '%s\n' "$allowed"))"
-
-if [ -n "$stale" ]; then
-  echo "NOTE: allowlisted crates no longer in gitlawb-core's dependency tree"
-  echo "      (safe to prune from ci/gitlawb-core-allowed-deps.txt):"
-  printf '  %s\n' $stale
-  echo
-fi
-
-if [ -n "$offenders" ]; then
-  {
-    echo "ERROR: gitlawb-core gained dependencies not on the allowlist:"
-    printf '  %s\n' $offenders
-    echo
-    echo "gitlawb-core must stay embeddable and lean. If a new dependency is"
-    echo "intentional, add it to ci/gitlawb-core-allowed-deps.txt (regen command"
-    echo "is in that file's header). Otherwise, drop the dependency."
-  } >&2
-  exit 1
-fi
-
-echo "gitlawb-core dependency purity: OK ($(printf '%s\n' "$current" | wc -l | tr -d ' ') normal deps, all allowlisted)."