fix(manager): report a spawned run as finished, not failed_to_start

[HusseinAdeiza]

Fixes #14

Problem

The protocol distinguishes failed_to_start from finished by whether a process was created, not by which code path noticed the error:

failed_to_start Terminal. The run never reached a spawned executor: request validation, version resolution, or the spawn itself failed.

The manager did not track that fact. supervise_genvm reported every error returned by supervise_genvm_inner as fail_to_start.

That misreports a real race in run_genvm_process. The stdin select! has two arms that can fire at nearly the same instant once the child has exited:

• the stdin writer returning EPIPE (failed to write execution data to child stdin), and
• child.wait() completing normally.

Whichever loses the race returns an error, so the host is told no process ever ran even though one did. The caller cannot observe which arm won, and the reported terminal event depended on it.

The existing system test at tests/system/manager-socket/test.py only passes because it happens to lose the race in the expected direction, which is why it reads as load-bearing on that race.

Fix

Record the spawned state at the one point where it is known, immediately after proc.spawn() returns, and let that decide the terminal event.

There is no exit status to report when the failure precedes child.wait(), so the finished event carries a null exit_code. The protocol allows this: exit_code is documented as null when a run was killed before its exit code was known. The error still goes to the operator log, since finished has no field to carry it.

Failures that happen before proc.spawn() returns are unchanged: they are still failed_to_start.

Verification

The test drives report_run_failure, the function production calls for every error out of a run, so it covers the classification rather than a reimplementation of it. Neutralising the branch to the old behaviour fails it with exactly the reported symptom:

spawned run reported Event(FailedToStart { genvm_id: GenVMId(1), ...,
error: "failed to write execution data to child stdin: EPIPE" }),
expected a finished event

The test also keeps a control case (the same error with no process is still failed_to_start), so it would still fail if the branch were inverted and every failure became a finished.

• implementation lib suite: 80 passed, 0 failed
• cargo fmt --check: clean

Built against v0.6-dev, which is where this ships.

Not included

I did not touch the system test. Making the race deterministic at that level requires an execution-data body too large for the pipe buffer, which is a larger change to shared test infrastructure and reads as a separate concern from the classification fix. Flagged on the issue in case you would rather the protocol-level test also stopped depending on which arm wins.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

🗂️ Base branches to auto review (3)

• main
• v\d+.\d+
• v\d+.\d+-dev

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

• Configuration used: Repository: genlayerlabs/genvm-manager/.coderabbit.yaml
• Review profile: CHILL
• Plan: Advanced
• Run ID: f5af2ad8-86bf-4c4a-94c3-3b3c35534bf7

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

GenVM PR actions

Tick a box to run it (the box unticks itself when handled). Actions only run while the PR has the ci-safe label.

• Force run full tests
• Provision executor PRs

Commands

• /genvm-run-tests — run full tests once for the current manager snapshot
• /merge — queue the exact manager snapshot through the App-owned E2E merge train

[kp2pml30]

Thanks! Landed via the feat/manager-socket-train train (#58), reworked after review with you credited as co-author; it reaches v0.6-dev when #58 merges