Compare commits

..
3 changed files with 427 additions and 323 deletions

No files matched your search

+2 -7
View File
@@ -188,12 +188,7 @@ export GITLAWB_NODE=http://localhost:7545
git clone gitlawb://did:key:z6Mk.../my-repo
```
For public-network use, make sure `GITLAWB_NODE` points to the node you want, over
`https://`. The helper defaults to localhost for local development, and plaintext
`http://` is allowed only to this machine: requests are signed but not encrypted, so
a cleartext hop to a remote node exposes the pack contents and the `Signature`
header. A remote `http://` node is refused, and `GITLAWB_ALLOW_INSECURE_HTTP=1`
overrides that for a trusted private network.
For public-network use, make sure `GITLAWB_NODE` points to the node you want. The helper defaults to localhost for local development.
### Full lifecycle against an iCaptcha-enforcing node
@@ -398,7 +393,7 @@ Important node settings:
| `GITLAWB_P2P_BOOTSTRAP` | Comma-separated libp2p multiaddrs. |
| `GITLAWB_BOOTSTRAP_DISABLE_SEEDS` | Disable embedded seed peers for isolated dev/test networks. |
| `GITLAWB_REQUIRE_SIGNED_PEER_WRITES` | Require signed peer announce/sync writes. Defaults to `false` during the staged rollout below. |
| `GITLAWB_ENFORCE_OWNER_PUSH` | Require the authenticated pusher to be the repo owner on `git-receive-pack`. **Defaults to `true`.** A `did:key` signature is authentication, not authorization (anyone can mint a key and sign), so with this off every signed caller may push to every repository, private ones included. Delegated and CI keys count as non-owners: a UCAN `git/push` capability is verified but not yet honored for authorization, so they cannot push while this is on. Set `false` only for a rolling upgrade; see [`docs/RUN-A-NODE.md`](docs/RUN-A-NODE.md). |
| `GITLAWB_ENFORCE_OWNER_PUSH` | Require the authenticated pusher to be the repo owner on `git-receive-pack`. **Defaults to `true`.** A `did:key` signature is authentication, not authorization — anyone can mint a key and sign — so with this off every signed caller may push to every repository, private ones included. Delegated and CI keys count as non-owners: a UCAN `git/push` capability is verified but not yet honored for authorization, so they cannot push while this is on. Set `false` only for a rolling upgrade; see [`docs/RUN-A-NODE.md`](docs/RUN-A-NODE.md). |
| `GITLAWB_AUTO_SYNC` | Enable automatic sync from known peers. |
| `GITLAWB_MAX_PACK_BYTES` | Max git pack body size for smart-HTTP routes. |
| `GITLAWB_GIT_SERVICE_TIMEOUT_SECS` | Max seconds a served git upload-pack, receive-pack, or `info/refs` advertisement may run before it is aborted (504). Default 600. Also bounds the withheld-blob classification walk (on both the upload-pack serve and receive-pack replication paths) and the push-side pin-candidate discovery (`rev-list` / `cat-file`), each reaped via process-group teardown at the deadline. On the path-scoped upload-pack path the classification walk and the pack serve share ONE deadline, so this value bounds their combined duration rather than granting each stage a full budget: a walk that consumes it leaves the serve nothing and the clone gets a 504. Serving large path-scoped repos may therefore need a higher value than they did when each stage was budgeted separately. Accepted range is 1 to 3153600000 (100 years), since the node derives deadlines from this value and a larger one cannot be represented. |
+293 -246
View File
@@ -66,10 +66,6 @@ fn main() -> Result<()> {
// v0.1: default to localhost. Override with GITLAWB_NODE env var.
let node_base =
std::env::var("GITLAWB_NODE").unwrap_or_else(|_| "http://127.0.0.1:7545".to_string());
check_transport_security(
&node_base,
std::env::var_os("GITLAWB_ALLOW_INSECURE_HTTP").is_some(),
)?;
let repo_base = format!("{}/{}/{}", node_base, short_owner, repo_name);
tracing::debug!("repo_base: {repo_base}");
@@ -82,76 +78,6 @@ fn main() -> Result<()> {
run_helper(&repo_base, keypair.as_ref())
}
/// Whether `url` is plaintext http, regardless of where it points.
fn is_http(url: &str) -> bool {
reqwest::Url::parse(url.trim())
.map(|u| u.scheme() == "http")
.unwrap_or(false)
}
/// Whether `url` would send git data off this machine in cleartext.
///
/// True only for `http://` to a non-loopback host. RFC 9421 signs the request
/// but does not encrypt it, so on a plaintext hop the pack contents and the
/// `Signature` header are both readable, and a captured signature is replayable
/// for its freshness window against any host.
///
/// Loopback is decided from the parsed address rather than a string match, so
/// `127.0.0.2`, `[::1]` and an IPv4-mapped `[::ffff:127.0.0.1]` are all
/// recognised as this machine. A value that does not parse, or that is not
/// http(s), is not this guard's business and returns false: it fails later with
/// its own error, and naming it a TLS problem would misdirect the reader.
fn is_insecure_remote(url: &str) -> bool {
let Ok(parsed) = reqwest::Url::parse(url.trim()) else {
return false;
};
if parsed.scheme() != "http" {
return false;
}
let Some(host) = parsed.host_str() else {
return false;
};
let host = host.trim_end_matches('.').to_ascii_lowercase();
if host == "localhost" {
return false;
}
// host_str() keeps the brackets on an IPv6 literal.
let bare = host.trim_start_matches('[').trim_end_matches(']');
if let Ok(ip) = bare.parse::<std::net::IpAddr>() {
if ip.is_loopback() {
return false;
}
// An IPv4-mapped IPv6 literal hides a v4 loopback from is_loopback().
if let std::net::IpAddr::V6(v6) = ip {
if let Some(v4) = v6.to_ipv4_mapped() {
if v4.is_loopback() {
return false;
}
}
}
}
true
}
/// Refuse a cleartext hop off this machine unless the operator has opted in.
///
/// Fail closed: the alternative is signing a request and then handing it, and
/// the pack it carries, to anyone on the path. `GITLAWB_ALLOW_INSECURE_HTTP`
/// exists for a private LAN where the operator has decided that is acceptable.
fn check_transport_security(node_base: &str, allow_insecure: bool) -> Result<()> {
if allow_insecure || !is_insecure_remote(node_base) {
return Ok(());
}
bail!(
"refusing to send git data to {node_base} over plaintext http.\n\
Requests are signed but not encrypted, so the pack contents and the \
Signature header are readable by anyone on the path, and a captured \
signature can be replayed.\n\
Use https://, or set GITLAWB_ALLOW_INSECURE_HTTP=1 to accept the risk \
(for a trusted private network only)."
)
}
// ── CLI argument handling ──────────────────────────────────────────────────────
/// How the binary was invoked, derived from its CLI arguments.
@@ -199,7 +125,6 @@ fn help_text() -> String {
ENVIRONMENT:\n\
\x20 GITLAWB_NODE Node base URL (default: http://127.0.0.1:7545)\n\
\x20 GITLAWB_KEY Identity PEM path for signed fetch/push (default: ~/.gitlawb/identity.pem)\n\
\x20 GITLAWB_ALLOW_INSECURE_HTTP Permit plaintext http:// to a non-loopback node (unset by default)\n\
\x20 GITLAWB_LOG Log filter (default: warn)\n\
\n\
FLAGS:\n\
@@ -274,8 +199,7 @@ fn handle_connect<R: Read>(
other => bail!("unsupported git service: {other}"),
}
// A loopback node must never be proxied; see build_http_client.
let client = build_http_client(!is_insecure_remote(repo_base) && is_http(repo_base))?;
let client = build_http_client()?;
// ── Phase 1: ref advertisement (GET /info/refs?service=<service>) ─────────
//
@@ -363,8 +287,10 @@ fn handle_connect<R: Read>(
// this into one POST deadlocks a multi-round fetch (#117).
//
// git-receive-pack (push):
// Git sends ref-update commands + the complete PACK blob, then closes its
// write pipe. read_to_end is safe and correct here, a single POST.
// Git sends the ref-update command list (pkt-lines ending in a flush),
// then the complete PACK blob only when a command has a non-zero new-oid.
// A delete-only push sends no pack, and git holds its write pipe open for
// the report-status response, so EOF cannot delimit the request (#369).
let post_url = format!("{}/{}", repo_base, service);
@@ -372,10 +298,7 @@ fn handle_connect<R: Read>(
return negotiate_upload_pack(&client, &post_url, service, signing_key, stdin, &mut stdout);
}
let mut request_body = Vec::new();
stdin
.read_to_end(&mut request_body)
.context("reading receive-pack request")?;
let request_body = read_receive_pack_request(stdin)?;
tracing::debug!("pack request: {} bytes from git", request_body.len());
if request_body.is_empty() {
@@ -411,22 +334,8 @@ const HTTP_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(300);
/// node named, and on a 307/308 the pack body went with them. Scope the follow to the
/// origin that issued the redirect AND to an identical request-target, which is the
/// same predicate `gl` uses.
/// `bypass_proxy` must be true whenever the node is on this machine. reqwest's
/// default client honours `HTTP_PROXY` / `ALL_PROXY` for a loopback URL too, so a
/// proxy variable pointing off-machine turns an allowed local request into a
/// cleartext hop carrying the `Signature` header, straight past
/// `check_transport_security`. Verified: with `HTTP_PROXY` set, reqwest logged
/// `proxy(...) intercepts 'http://127.0.0.1:7545/'` and dialled the proxy. A
/// local node is never legitimately reached through a proxy, so this refuses one
/// rather than trying to reimplement `NO_PROXY` parsing.
fn build_http_client(bypass_proxy: bool) -> Result<reqwest::blocking::Client> {
let builder = reqwest::blocking::Client::builder();
let builder = if bypass_proxy {
builder.no_proxy()
} else {
builder
};
Ok(builder
fn build_http_client() -> Result<reqwest::blocking::Client> {
Ok(reqwest::blocking::Client::builder()
.timeout(HTTP_TIMEOUT)
.redirect(reqwest::redirect::Policy::custom(same_origin_redirect))
.build()?)
@@ -625,6 +534,108 @@ fn read_upload_pack_round<R: Read>(stdin: &mut R) -> Result<(Vec<u8>, RoundEnd)>
}
}
/// Read one git-receive-pack request: optional `shallow` lines, then EITHER a
/// command list (pkt-lines terminated by a flush) or a push-cert block
/// (pkt-lines terminated by a "push-cert-end" line), then the negotiated
/// push-options block, then the pack ONLY when a command carries a non-zero
/// new-oid.
///
/// A delete-only push sends no pack, and git holds its write pipe open waiting
/// for the report-status response, so a plain read_to_end deadlocks the two
/// (#369). A signed push is the other trap: when the client sends a push-cert
/// it sends NO command list at all (the commands live inside the cert), so the
/// request starts with a "push-cert" line and no flush ever arrives. The
/// negotiated capability list is on the push-cert line, or on the first
/// command line after its NUL when there is no cert. EOF at any point also
/// ends the read, covering git versions that close the pipe early.
fn read_receive_pack_request<R: Read>(stdin: &mut R) -> Result<Vec<u8>> {
/// Read one pkt-line into `body` (length prefix + data), returning its data
/// or `None` on flush or EOF. Malformed lengths bail rather than collapse
/// into a synthesized flush (#192 F5).
fn read_pkt<R: Read>(stdin: &mut R, body: &mut Vec<u8>) -> Result<Option<Vec<u8>>> {
let mut len_bytes = [0u8; 4];
match stdin.read_exact(&mut len_bytes) {
Ok(()) => {}
Err(e) if e.kind() == io::ErrorKind::UnexpectedEof => return Ok(None),
Err(e) => return Err(e.into()),
}
let Ok(len_hex) = std::str::from_utf8(&len_bytes) else {
bail!("malformed pkt-line length prefix: non-UTF-8 bytes {len_bytes:02x?}");
};
let Ok(pkt_len) = usize::from_str_radix(len_hex, 16) else {
bail!("malformed pkt-line length prefix: non-hex {len_hex:?}");
};
body.extend_from_slice(&len_bytes);
if pkt_len == 0 {
return Ok(None);
}
if pkt_len < 4 {
bail!("invalid pkt-line length: {pkt_len}");
}
let mut data = vec![0u8; pkt_len - 4];
stdin
.read_exact(&mut data)
.context("reading pkt-line data")?;
body.extend_from_slice(&data);
Ok(Some(data))
}
/// "<40-hex old> <40-hex new> <refname>" — the command shape, whether it
/// appears in a command list or inside a push-cert.
fn is_command(data: &[u8]) -> bool {
data.len() >= 82
&& data[..40].iter().all(|b| b.is_ascii_hexdigit())
&& data[40] == b' '
&& data[41..81].iter().all(|b| b.is_ascii_hexdigit())
&& data[81] == b' '
}
let mut body = Vec::new();
let mut caps = String::new();
let mut pack_expected = false;
let mut in_cert = false;
loop {
let Some(data) = read_pkt(stdin, &mut body)? else {
// Flush ends a command list; EOF ends anything. A cert has no
// legal flush inside it, so the same break covers malformed
// input without hanging.
break;
};
if !in_cert && data.starts_with(b"push-cert") {
in_cert = true;
if let Some(nul) = data.iter().position(|&b| b == 0) {
caps = String::from_utf8_lossy(&data[nul + 1..]).into_owned();
}
continue;
}
if in_cert && data == b"push-cert-end\n" {
break;
}
if is_command(&data) {
if caps.is_empty() {
if let Some(nul) = data.iter().position(|&b| b == 0) {
caps = String::from_utf8_lossy(&data[nul + 1..]).into_owned();
}
}
if data[41..81].iter().any(|&b| b != b'0') {
pack_expected = true;
}
}
}
// When both are negotiated the push-options block follows the cert; when
// only options are negotiated it follows the command flush. Either way it
// is pkt-lines to a flush.
if caps.split(' ').any(|c| c.starts_with("push-options")) {
while read_pkt(stdin, &mut body)?.is_some() {}
}
if pack_expected {
stdin
.read_to_end(&mut body)
.context("reading receive-pack pack")?;
}
Ok(body)
}
/// POST one self-contained stateless-RPC request body and stream the response to
/// `stdout`. Signs the request when `signing_key` is present, so every round of a
/// private-repo fetch carries the caller's signature (the node visibility-gates
@@ -919,148 +930,6 @@ fn resolve_key_path() -> std::path::PathBuf {
// ── Tests ─────────────────────────────────────────────────────────────────────
#[cfg(test)]
mod insecure_transport_tests {
use super::*;
/// Loopback in every form the transport can legitimately be pointed at.
/// Plaintext to this machine is the documented local-alpha default and must
/// keep working, so a false positive here breaks every stock install.
#[test]
fn loopback_http_is_allowed() {
for url in [
"http://127.0.0.1:7545",
"http://localhost:7545",
"http://[::1]:7545",
"http://127.0.0.2:7545",
"http://[::ffff:127.0.0.1]:7545",
"http://LocalHost:7545",
] {
assert!(
!is_insecure_remote(url),
"{url} is this machine; plaintext to it must stay allowed"
);
}
}
/// The case the guard exists for: cleartext git data leaving the machine.
#[test]
fn remote_http_is_refused() {
for url in [
"http://node.example.com:7545",
"http://10.0.0.36:7777",
"http://192.168.1.10",
"http://[2001:db8::1]:7545",
"http://8.8.8.8",
] {
assert!(
is_insecure_remote(url),
"{url} sends git data off-machine in cleartext and must be refused"
);
}
}
/// TLS is always fine, loopback or not, so the guard keys on the scheme and
/// not merely on the host being remote.
#[test]
fn https_is_always_allowed() {
for url in [
"https://node.gitlawb.com",
"https://10.0.0.36:7777",
"https://127.0.0.1:7545",
] {
assert!(!is_insecure_remote(url), "{url} is TLS and must be allowed");
}
}
/// Spellings that defeat a naive loopback check. `Url` normalizes the scheme
/// and the host at parse time (lowercasing, and the WHATWG numeric forms), so
/// these must all still read as this machine.
#[test]
fn loopback_spellings_are_normalized_not_string_matched() {
for url in [
"HTTP://127.0.0.1:7545",
"Http://LOCALHOST:7545",
"http://localhost.:7545",
"http://user:[email protected]:7545",
"http://2130706433:7545",
"http://0x7f000001:7545",
"http://[::ffff:7f00:1]:7545",
] {
assert!(
!is_insecure_remote(url),
"{url} resolves to this machine; plaintext to it must stay allowed"
);
}
}
/// The mirror of the above: an uppercase scheme must not become an escape
/// from the guard, and a remote host in any spelling is still remote.
#[test]
fn uppercase_scheme_does_not_escape_the_guard() {
for url in [
"HTTP://node.example.com:7545",
"Http://8.8.8.8",
"http://NODE.EXAMPLE.COM",
"http://user:[email protected]",
] {
assert!(
is_insecure_remote(url),
"{url} is a remote cleartext hop and must be refused"
);
}
}
/// The gate itself, both directions, including the opt-in escape hatch.
#[test]
fn gate_refuses_remote_plaintext_unless_opted_in() {
let err = check_transport_security("http://node.example.com:7545", false)
.expect_err("remote plaintext must be refused by default");
let msg = err.to_string();
assert!(
msg.contains("plaintext http"),
"the refusal must say why: {msg}"
);
assert!(
msg.contains("GITLAWB_ALLOW_INSECURE_HTTP"),
"the refusal must name the escape hatch: {msg}"
);
check_transport_security("http://node.example.com:7545", true)
.expect("the opt-in must permit the same URL");
check_transport_security("http://127.0.0.1:7545", false)
.expect("the local default must keep working with no opt-in");
check_transport_security("https://node.gitlawb.com", false)
.expect("TLS must need no opt-in");
}
/// The documented knob must appear in --help. A gate the operator cannot
/// discover reads as a broken transport rather than a deliberate refusal.
#[test]
fn help_documents_the_opt_in() {
assert!(help_text().contains("GITLAWB_ALLOW_INSECURE_HTTP"));
}
/// An unparseable or non-http value is not classified as insecure here: it
/// fails later with its own error, and reporting it as a TLS problem would
/// send the reader after the wrong thing.
#[test]
fn unparseable_or_other_scheme_is_not_this_guards_problem() {
for url in [
"",
" ",
"not a url",
"ftp://node.example.com",
"file:///tmp/x",
] {
assert!(
!is_insecure_remote(url),
"{url:?} is not a cleartext-http-to-remote case"
);
}
}
}
#[cfg(test)]
mod tests {
use super::*;
@@ -1161,7 +1030,7 @@ mod tests {
#[test]
fn signed_requests_do_not_follow_a_redirect_off_the_node_origin() {
let kp = Keypair::generate();
let client = build_http_client(false).unwrap();
let client = build_http_client().unwrap();
let mut elsewhere = mockito::Server::new();
let never = elsewhere
@@ -1263,7 +1132,7 @@ mod tests {
#[test]
fn a_same_origin_path_changing_redirect_is_refused() {
let kp = Keypair::generate();
let client = build_http_client(false).unwrap();
let client = build_http_client().unwrap();
let mut node = mockito::Server::new();
let bounce = node
@@ -1314,7 +1183,7 @@ mod tests {
#[test]
fn an_identical_target_redirect_is_still_followed_up_to_the_chain_bound() {
let kp = Keypair::generate();
let client = build_http_client(false).unwrap();
let client = build_http_client().unwrap();
let mut node = mockito::Server::new();
let loop_route = node
@@ -1475,7 +1344,7 @@ mod tests {
fn a_rewritten_target_never_receives_the_signature() {
let kp = Keypair::generate();
let expected_did = kp.did().to_string();
let client = build_http_client(false).unwrap();
let client = build_http_client().unwrap();
let slot = std::sync::Arc::new(std::sync::Mutex::new(None::<Verdict>));
let mut node = mockito::Server::new();
@@ -1537,7 +1406,7 @@ mod tests {
fn a_direct_signed_advertisement_verifies_under_the_node_verifier() {
let kp = Keypair::generate();
let expected_did = kp.did().to_string();
let client = build_http_client(false).unwrap();
let client = build_http_client().unwrap();
let slot = std::sync::Arc::new(std::sync::Mutex::new(None::<Verdict>));
let mut node = mockito::Server::new();
@@ -2249,6 +2118,184 @@ mod tests {
assert!(help.ends_with('\n'));
}
// ── #369 delete-only push deadlock ───────────────────────────────────────
/// A delete-only receive-pack request ends at the command flush; git then
/// holds the write pipe open for the report and sends no pack. A reader that
/// blocks after the command list reproduces the hang that a Cursor (EOF)
/// hides. The read must return at the flush, inside the timeout.
#[test]
fn delete_only_receive_pack_request_returns_at_command_flush() {
let old = "a".repeat(40);
let zero = "0".repeat(40);
let mut seed = Vec::new();
seed.extend_from_slice(&pkt(&format!("{old} {zero} refs/heads/gone\n")));
seed.extend_from_slice(b"0000"); // command-list flush; git then blocks
let (tx, rx) = std::sync::mpsc::channel::<()>();
let mut reader = BlockAfterSeed {
seed: io::Cursor::new(seed),
gate: rx,
};
let (done_tx, done_rx) = std::sync::mpsc::channel::<Vec<u8>>();
let handle = std::thread::spawn(move || {
let _ = done_tx.send(read_receive_pack_request(&mut reader).unwrap());
});
let got = done_rx.recv_timeout(std::time::Duration::from_secs(2));
let _ = tx.send(()); // unblock the reader so the thread can exit
let _ = handle.join();
let body = got.expect(
"read_receive_pack_request blocked past the command flush waiting for \
a pack a delete-only push never sends (#369)",
);
assert!(
body.ends_with(b"0000"),
"the request body keeps the command list and its flush terminator"
);
}
/// A push carrying objects still reads the pack through to EOF, and the
/// command flush sits between the commands and the pack in the body.
#[test]
fn receive_pack_request_with_pack_reads_through_eof() {
let old = "a".repeat(40);
let new = "b".repeat(40);
let mut stream = Vec::new();
stream.extend_from_slice(&pkt(&format!(
"{old} {new} refs/heads/main\0 report-status-v2\n"
)));
stream.extend_from_slice(b"0000");
stream.extend_from_slice(b"PACK\x00\x00\x00\x02fakepack");
let mut stdin = io::Cursor::new(stream);
let body = read_receive_pack_request(&mut stdin).unwrap();
let pos = body.windows(4).position(|w| w == b"0000").unwrap();
assert_eq!(
&body[pos + 4..],
b"PACK\x00\x00\x00\x02fakepack",
"a non-zero new-oid means the pack follows the flush"
);
}
/// A signed delete-only push sends NO command list: the request starts
/// with the "push-cert" line, the commands live inside the cert, and the
/// cert ends with a "push-cert-end" pkt-line (not a flush). The reader
/// must return after push-cert-end without waiting on the pipe.
#[test]
fn signed_delete_only_push_returns_after_push_cert_end() {
let old = "a".repeat(40);
let zero = "0".repeat(40);
let mut seed = Vec::new();
seed.extend_from_slice(&pkt("push-cert\0 report-status-v2 push-cert\n"));
seed.extend_from_slice(&pkt("certificate version 0.1\n"));
seed.extend_from_slice(&pkt("pusher did:key:z6Mk\n"));
seed.extend_from_slice(&pkt("\n")); // empty line ends the headers
seed.extend_from_slice(&pkt(&format!("{old} {zero} refs/heads/gone\n")));
seed.extend_from_slice(&pkt("sig-line\n"));
seed.extend_from_slice(&pkt("push-cert-end\n"));
// git now blocks for the report; the pipe stays open.
let (tx, rx) = std::sync::mpsc::channel::<()>();
let mut reader = BlockAfterSeed {
seed: io::Cursor::new(seed),
gate: rx,
};
let (done_tx, done_rx) = std::sync::mpsc::channel::<Vec<u8>>();
let handle = std::thread::spawn(move || {
let _ = done_tx.send(read_receive_pack_request(&mut reader).unwrap());
});
let got = done_rx.recv_timeout(std::time::Duration::from_secs(2));
let _ = tx.send(());
let _ = handle.join();
let body = got.expect(
"read_receive_pack_request blocked past push-cert-end on a signed \
delete-only push (#369)",
);
assert!(
body.windows(6).any(|w| w == b"pusher"),
"the cert block is part of the request body"
);
}
/// A signed pack-bearing push: cert with the commands inside, then the
/// pack. Everything must land in the body.
#[test]
fn signed_pack_push_keeps_cert_and_pack() {
let old = "a".repeat(40);
let new = "b".repeat(40);
let mut stream = Vec::new();
stream.extend_from_slice(&pkt("push-cert\0 report-status-v2 push-cert\n"));
stream.extend_from_slice(&pkt("certificate version 0.1\n"));
stream.extend_from_slice(&pkt("\n"));
stream.extend_from_slice(&pkt(&format!("{old} {new} refs/heads/main\n")));
stream.extend_from_slice(&pkt("push-cert-end\n"));
stream.extend_from_slice(b"PACK\x00\x00\x00\x02fakepack");
let mut stdin = io::Cursor::new(stream);
let body = read_receive_pack_request(&mut stdin).unwrap();
assert!(
body.windows(5).any(|w| w == b"PACK\x00"),
"the pack follows the cert into the body"
);
}
/// A negotiated push-options block is pkt-lines ending in its own flush.
/// With no cert it follows the command flush; with a cert it follows
/// push-cert-end.
#[test]
fn receive_pack_request_consumes_a_push_options_block() {
let old = "a".repeat(40);
let new = "b".repeat(40);
let mut stream = Vec::new();
stream.extend_from_slice(&pkt(&format!(
"{old} {new} refs/heads/main\0 report-status-v2 push-options\n"
)));
stream.extend_from_slice(b"0000"); // command flush
stream.extend_from_slice(&pkt("ci.skip\n"));
stream.extend_from_slice(b"0000"); // options flush
stream.extend_from_slice(b"PACK\x00\x00\x00\x02fakepack");
let mut stdin = io::Cursor::new(stream);
let body = read_receive_pack_request(&mut stdin).unwrap();
assert!(body.windows(7).any(|w| w == b"ci.skip"), "options survived");
assert!(
body.ends_with(b"PACK\x00\x00\x00\x02fakepack"),
"the pack lands after the options block"
);
}
/// Cert and options together: the cert comes first and the options block
/// sits between push-cert-end and the pack.
#[test]
fn signed_push_with_options_reads_cert_then_options_then_pack() {
let old = "a".repeat(40);
let new = "b".repeat(40);
let mut stream = Vec::new();
stream.extend_from_slice(&pkt(
"push-cert\0 report-status-v2 push-cert push-options\n",
));
stream.extend_from_slice(&pkt("certificate version 0.1\n"));
stream.extend_from_slice(&pkt("\n"));
stream.extend_from_slice(&pkt(&format!("{old} {new} refs/heads/main\n")));
stream.extend_from_slice(&pkt("push-cert-end\n"));
stream.extend_from_slice(&pkt("ci.skip\n"));
stream.extend_from_slice(b"0000"); // options flush
stream.extend_from_slice(b"PACK\x00\x00\x00\x02fakepack");
let mut stdin = io::Cursor::new(stream);
let body = read_receive_pack_request(&mut stdin).unwrap();
assert!(body.windows(7).any(|w| w == b"ci.skip"), "options survived");
assert!(
body.ends_with(b"PACK\x00\x00\x00\x02fakepack"),
"the pack lands after the options block"
);
}
// ── #117 multi-round fetch negotiation ───────────────────────────────────
/// Encode a git pkt-line: 4-byte hex length (incl. the 4 bytes) + data.
+132 -70
View File
@@ -95,6 +95,36 @@ fn upload_pack(repo: &Path, advertise: bool, input: &[u8]) -> Vec<u8> {
out.stdout
}
/// Run `git receive-pack --stateless-rpc [--advertise-refs] <repo>`, optionally
/// feeding `input` on stdin. Returns raw stdout (the wire bytes). The receive
/// half of `upload_pack`, used so the shim can serve a real push.
fn receive_pack(repo: &Path, advertise: bool, input: &[u8]) -> Vec<u8> {
let mut cmd = Command::new("git");
cmd.arg("receive-pack").arg("--stateless-rpc");
if advertise {
cmd.arg("--advertise-refs");
}
cmd.arg(repo)
.stdin(Stdio::piped())
.stdout(Stdio::piped())
.stderr(Stdio::piped());
let mut child = cmd.spawn().expect("spawn git receive-pack");
let mut stdin = child.stdin.take().unwrap();
let input = input.to_vec();
let writer = std::thread::spawn(move || {
let _ = stdin.write_all(&input);
// drop closes stdin
});
let out = child.wait_with_output().expect("wait receive-pack");
writer.join().ok();
assert!(
out.status.success(),
"git receive-pack failed: {}",
String::from_utf8_lossy(&out.stderr)
);
out.stdout
}
#[derive(Clone, Copy)]
enum ShimMode {
/// Faithful v0 serving: pipe each POST to `git upload-pack --stateless-rpc`.
@@ -218,7 +248,21 @@ fn handle_conn(stream: TcpStream, repo: &Path, mode: ShimMode, posts: &AtomicUsi
reader.read_exact(&mut body).ok();
}
let (content_type, payload) = if method == "GET" && target.contains("/info/refs") {
let (content_type, payload) = if method == "GET" && target.contains("service=git-receive-pack")
{
// Receive-pack advertisement, same wrapper shape as upload-pack.
let adv = receive_pack(repo, true, b"");
let mut wrapped = pkt(b"# service=git-receive-pack\n");
wrapped.extend_from_slice(b"0000");
wrapped.extend_from_slice(&adv);
("application/x-git-receive-pack-advertisement", wrapped)
} else if method == "POST" && target.ends_with("/git-receive-pack") {
posts.fetch_add(1, Ordering::SeqCst);
(
"application/x-git-receive-pack-result",
receive_pack(repo, false, &body),
)
} else if method == "GET" && target.contains("/info/refs") {
// v0 advertisement, wrapped exactly as the node's info_refs does.
let adv = upload_pack(repo, true, b"");
let mut wrapped = pkt(b"# service=git-upload-pack\n");
@@ -267,15 +311,6 @@ fn write_response(mut stream: TcpStream, status: &str, content_type: &str, body:
/// Run `git fetch` in `clone` through the helper, with a hard timeout so a
/// regression to the deadlock fails fast instead of hanging the suite.
fn fetch_with_helper(clone: &Path, node_url: &str) -> (bool, std::process::Output) {
fetch_with_helper_env(clone, node_url, &[])
}
/// [`fetch_with_helper`] with extra environment for the helper process.
fn fetch_with_helper_env(
clone: &Path,
node_url: &str,
extra_env: &[(&str, &str)],
) -> (bool, std::process::Output) {
let helper_bin = PathBuf::from(env!("CARGO_BIN_EXE_git-remote-gitlawb"));
let helper_dir = helper_bin.parent().unwrap().to_path_buf();
let path_env = match std::env::var_os("PATH") {
@@ -298,14 +333,35 @@ fn fetch_with_helper_env(
// machine/CI with git-l10n installed and LANG set to a translated locale.
.env("LC_ALL", "C")
.env("GITLAWB_NODE", node_url)
.env("GITLAWB_KEY", "/nonexistent-key-for-anon-fetch")
// The helper inherits this process's environment, so an operator or CI
// that exports the insecure-HTTP override would decide the transport
// policy for every test below. Clear it and let each test opt in.
.env_remove("GITLAWB_ALLOW_INSECURE_HTTP");
for (k, v) in extra_env {
cmd.env(k, v);
}
.env("GITLAWB_KEY", "/nonexistent-key-for-anon-fetch");
run_bounded(cmd, Duration::from_secs(30))
}
/// Run `git push` in `local` through the helper, same wiring and hard timeout as
/// [`fetch_with_helper`]. `args` carries the push spec, e.g.
/// `["--delete", "doomed"]`.
fn push_with_helper(local: &Path, node_url: &str, args: &[&str]) -> (bool, std::process::Output) {
let helper_bin = PathBuf::from(env!("CARGO_BIN_EXE_git-remote-gitlawb"));
let helper_dir = helper_bin.parent().unwrap().to_path_buf();
let path_env = match std::env::var_os("PATH") {
Some(p) => {
let mut dirs = vec![helper_dir.clone()];
dirs.extend(std::env::split_paths(&p));
std::env::join_paths(dirs).unwrap()
}
None => helper_dir.clone().into_os_string(),
};
let mut cmd = Command::new("git");
cmd.arg("-C")
.arg(local)
.arg("push")
.arg("origin")
.args(args)
.env("PATH", path_env)
.env("LC_ALL", "C")
.env("GITLAWB_NODE", node_url)
.env("GITLAWB_KEY", "/nonexistent-key-for-anon-push");
run_bounded(cmd, Duration::from_secs(30))
}
@@ -1259,70 +1315,76 @@ fn run_bounded_bounds_join_when_leader_exits_leaving_a_pipe_holder() {
);
}
/// The plaintext-transport gate, proven at its CALL SITE rather than in
/// isolation. `main.rs`'s unit tests cover `is_insecure_remote` and
/// `check_transport_security` as functions; neither notices if the call in
/// `main()` is deleted, which is the regression that would silently restore the
/// cleartext hop. Only driving the built binary binds the wiring.
///
/// `192.0.2.1` is TEST-NET-1 (RFC 5737): non-loopback, reserved for
/// documentation, and never routable, so the gate is what stops this and no
/// connection is attempted even if it regressed.
/// #369 end-to-end: a real `git push --delete` through the helper completes.
/// The receive-pack request for a delete-only push is commands + flush with NO
/// pack, and git holds its write pipe open for the report — so a helper that
/// reads stdin to EOF deadlocks against it. The shim pipes the POST to a real
/// `git receive-pack --stateless-rpc`, so the branch actually has to be deleted
/// server-side for the assertions to pass.
#[test]
fn real_git_fetch_refuses_a_remote_plaintext_node() {
// A shallow fixture is enough: the gate fires before any negotiation.
let repos = build_divergent_repos(1);
let (ok, out) = fetch_with_helper(&repos.clone, "http://192.0.2.1:7545");
let stderr = String::from_utf8_lossy(&out.stderr);
fn real_git_push_delete_completes() {
let dir = tempfile::tempdir().unwrap();
let server = dir.path().join("server.git");
let local = dir.path().join("local");
assert!(
!ok || !out.status.success(),
"the fetch must not succeed: {stderr}"
);
assert!(
stderr.contains("plaintext http"),
"the helper must refuse the cleartext hop and say so; stderr was: {stderr}"
);
assert!(
stderr.contains("GITLAWB_ALLOW_INSECURE_HTTP"),
"the refusal must name the escape hatch; stderr was: {stderr}"
);
}
/// A proxy variable must not divert a LOOPBACK node off this machine.
///
/// reqwest's default client honours `HTTP_PROXY` for a loopback URL, so before
/// the fix a proxy pointing off-machine carried the signed plaintext request to
/// it, past the transport guard, which only ever inspected the URL. Observed
/// directly: `proxy(http://...:9999/) intercepts 'http://127.0.0.1:7545/'`.
///
/// The proxy here is a black hole on TEST-NET-1, so if the bypass regresses this
/// fetch cannot succeed: it either hangs to the harness timeout or fails to
/// connect. Success is only possible when the request went straight to the shim.
#[test]
fn real_git_fetch_ignores_a_proxy_for_a_loopback_node() {
let repos = build_divergent_repos(1);
let (server, clone) = (repos.server.clone(), repos.clone.clone());
let shim = start_shim(server.clone(), ShimMode::Normal);
let (completed, out) = fetch_with_helper_env(
&clone,
&shim.base_url,
// A bare repo holding the doomed branch, plus a live one.
git(
dir.path(),
&[
("HTTP_PROXY", "http://192.0.2.1:9999"),
("http_proxy", "http://192.0.2.1:9999"),
("ALL_PROXY", "http://192.0.2.1:9999"),
"init",
"--bare",
server.to_str().unwrap(),
"--initial-branch=main",
],
);
let work = dir.path().join("work");
git(
dir.path(),
&["init", work.to_str().unwrap(), "--initial-branch=main"],
);
git(&work, &["config", "user.email", "t@t"]);
git(&work, &["config", "user.name", "t"]);
std::fs::write(work.join("f"), "x").unwrap();
git(&work, &["add", "f"]);
git(&work, &["commit", "-m", "c"]);
git(&work, &["branch", "doomed"]);
git(&work, &["push", server.to_str().unwrap(), "main", "doomed"]);
// The pusher: a local repo whose origin is the gitlawb remote pointing at
// the shim. Nothing needs to be fetched first; git takes the old-oid from
// the receive-pack advertisement.
git(
dir.path(),
&["init", local.to_str().unwrap(), "--initial-branch=main"],
);
git(
&local,
&["remote", "add", "origin", "gitlawb://did:key:z6MkTest/repo"],
);
let shim = start_shim(server.clone(), ShimMode::Normal);
let (completed, out) = push_with_helper(&local, &shim.base_url, &["--delete", "doomed"]);
let posts = shim.posts.load(Ordering::SeqCst);
assert!(
completed,
"the fetch did not finish; a proxied loopback request would stall here. stderr:\n{}",
"git push --delete did not complete within the timeout (deadlock signature). stderr:\n{}",
String::from_utf8_lossy(&out.stderr)
);
assert!(
out.status.success(),
"a loopback fetch must ignore the proxy variables. stderr:\n{}",
"git push --delete failed. stderr:\n{}",
String::from_utf8_lossy(&out.stderr)
);
assert_eq!(posts, 1, "a delete-only push is exactly one POST");
let gone = Command::new("git")
.arg("-C")
.arg(&server)
.args(["rev-parse", "--verify", "--quiet", "refs/heads/doomed"])
.output()
.unwrap();
assert!(
!gone.status.success(),
"refs/heads/doomed must be deleted in the server repo"
);
}