Skip to content

fix(tests): copy the runner, not /bin/cat, as the POSIX blocked image - #1921

Merged
DeusData merged 1 commit into
DeusData:mainfrom
OhOkThisIsFine:fix/daemon-runtime-uutils-image-donor
Aug 31, 2026
Merged

fix(tests): copy the runner, not /bin/cat, as the POSIX blocked image#1921
DeusData merged 1 commit into
DeusData:mainfrom
OhOkThisIsFine:fix/daemon-runtime-uutils-image-donor

Conversation

@OhOkThisIsFine

@OhOkThisIsFine OhOkThisIsFine commented Aug 29, 2026

Copy link
Copy Markdown
Contributor

Problem

daemon_runtime_process_fingerprint_never_hashes_replacement_path (POSIX variant) copies /bin/cat to a temp file named image and executes the copy. On Ubuntu 25.10+/26.04 — and any system with uutils/Rust coreutils — /bin/cat is a multi-call binary. A copy invoked under the name image prints coreutils: unknown program 'image' and exits 1, so the test fails at ASSERT(setup).

Reproduced on WSL Ubuntu 26.04 (uutils coreutils 0.8.0):

$ cp /bin/cat "$d/image" && echo hi | "$d/image"
coreutils: unknown program 'image'
exit=1

Fix

Mirror the Windows variant of the same test, which already copies the test runner and parks it in a __cbm_runtime_image_holder argv mode:

  • tests/test_main.c — the __cbm_runtime_image_holder mode gains a POSIX branch: block reading stdin until EOF, exactly the behaviour the cat donor provided.
  • tests/test_daemon_runtime.c — runtime_test_spawn_blocked_executable passes the holder argument; its target must now be a copy of the runner (it has exactly one caller). The POSIX fixture copies the runner via runtime_test_copy_self_image instead of /bin/cat.
  • The /bin/echo replacement donor is gone too. It was only hashed, never executed — and on a uutils system cat and echo can be the same multi-call bytes, which would break the fixture's original != replacement precondition. The replacement is now a small data file, mirroring the Windows variant.

No macOS signing change is needed: the image copy stays byte-identical to the runner, and the identical-copy fixture already proves such a copy executes.

Verification

scripts/test.sh --suites daemon_runtime BUILD_DIR=build/wsl on WSL Ubuntu 26.04 (uutils coreutils 0.8.0): 44 passed / 0 failed, including daemon_runtime_process_fingerprint_never_hashes_replacement_path.

🤖 Generated with Claude Code

@OhOkThisIsFine @claude
fix(tests): copy the runner, not /bin/cat, as the POSIX blocked image
68b8809 
daemon_runtime_process_fingerprint_never_hashes_replacement_path (POSIX
variant) copied /bin/cat to a temp file named "image" and executed the
copy. On Ubuntu 25.10+/26.04 and any other uutils system, /bin/cat is a
multi-call binary: a copy invoked under the name "image" prints
"coreutils: unknown program 'image'" and exits 1, so the fixture died at
ASSERT(setup). Reproduced on WSL Ubuntu 26.04 (uutils coreutils 0.8.0).

The Windows variant of the same test already avoids this class of donor:
it copies the test runner itself and parks it in a
__cbm_runtime_image_holder argv mode. The POSIX variant now does the
same. The holder mode gains a POSIX branch that blocks reading stdin
until EOF — exactly the behaviour the cat donor provided — and
runtime_test_spawn_blocked_executable passes the holder argument, so its
target must now be a copy of this runner (it has exactly one caller).

The /bin/echo replacement donor is gone too: it was only ever hashed,
never executed, and on a uutils system cat and echo can be the same
multi-call bytes, which would have broken the fixture's
original != replacement precondition. The replacement is now a small
data file, mirroring the Windows variant.

No signing change is needed on macOS: the image copy stays
byte-identical to the runner, and the identical-copy fixture already
proves such a copy executes.

daemon_runtime suite on WSL Ubuntu 26.04: 44 passed / 0 failed.

Co-Authored-By: Claude Fable 5 <[email protected]>
Signed-off-by: OhOkThisIsFine <[email protected]>
@github-actions

github-actions Bot commented Aug 29, 2026

Copy link
Copy Markdown

Thanks for opening this — it has been seen, and it is queued.

This note is automated, but it is not a brush-off: it exists so you know where your PR stands instead of having to guess from silence.

Current review status: working through a backlog. 0.9.1-rc.1 is out, so the release freeze that held reviews is over — but it left a large queue of open pull requests behind it, and we are reading through them oldest-first. The background is in discussion #1144.

What that means for this PR, concretely:

  • It will not be closed for inactivity. No stale bot touches pull requests here.
  • It may still sit a while before a human reads it. That is on us, not on you.
  • Older PRs are read first, so a recent one is not being skipped — it is behind a queue.

Things that will genuinely speed it up whenever review does happen:

  • Keep it rebased on main — the tree is moving quickly right now, and a conflicting branch cannot be reviewed as the diff you intended.
  • Get CI green, or say which failures you believe are pre-existing.
  • Keep the change to one claim. Bundled features and refactors get split before they get merged, which costs you a round trip.
  • Every commit needs a sign-off (git commit -s) — CI enforces DCO.

If this fixes a bug, a reproduction we can run is worth more than a description of the symptom.

Thanks for contributing, and sorry in advance for the wait.

@DeusData
DeusData merged commit 0a8a1f2 into DeusData:main Aug 31, 2026
33 checks passed
@DeusData

DeusData commented Aug 31, 2026

Copy link
Copy Markdown
Owner

Merged as 0a8a1f29.

This is a fix for a failure most of the project has not seen yet, which is the useful kind. /bin/cat being a multi-call binary under uutils means the copy-and-rename trick stops working the moment a distro switches coreutils implementations — so this test was going to start failing across the board on its own schedule, with a message (coreutils: unknown program 'image') that points nowhere near the actual cause. Catching it on Ubuntu 26.04 before it arrives everywhere else is worth more than the diff size suggests.

Three things done right:

You mirrored the Windows variant instead of inventing a mechanism. The __cbm_runtime_image_holder argv mode already existed and already solved exactly this problem on the other platform. Extending it with a POSIX branch that blocks on stdin — the one behaviour the cat donor was actually providing — leaves one pattern in the tree rather than two.

You found the second instance of the same root cause. Removing the /bin/echo replacement donor because on a uutils system cat and echo can be the same multi-call bytes, which would silently break the fixture's original != replacement precondition, is the part that would have been missed. That donor was only ever hashed, never executed, so nothing pointed at it — you got there from the mechanism rather than from a failure.

You addressed the platform question before it was asked. Noting that the image copy stays byte-identical to the runner, and that the identical-copy fixture already proves such a copy executes, closed the macOS-signing concern in one sentence.

Reproduced with the exact failing command and output, fixed, and verified 44/0 on the affected platform. Thank you — a clean first contribution.

@DeusData DeusData mentioned this pull request Aug 31, 2026
Merged
@DeusData DeusData mentioned this pull request Sep 25, 2026
Closed
2 tasks done
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@DeusData DeusData Awaiting requested review from DeusData DeusData is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@OhOkThisIsFine @DeusData