[kixelated]

Problem

On a host without /dev/kvm, just ci's check flake boots the NixOS runner-image VM tests under software emulation and fails about 19 minutes later with "Shell did not start in time". This came up in #7175's local gate.

Approach

Plans one m0/ci [XS] quest, VM checks refuse a host without KVM: the check fails at once with a message naming the missing KVM, instead of skipping, so a local gate never reads as passing. #7175's other follow-up, the nested Codex BASH_ENV PATH growth, didn't reproduce with codex-cli 0.162.1 (BASH_ENV unset, PATH stable across three nested shells), so it is neither planned nor reported.

Impact

Quest documents only. No public API or wire changes.

Review

Local quest review, Codex gpt-6.1-sol at medium effort, round 1: no findings. Reviewed head 51cfce025fa840e40a54fa8e4215581cfee51a77 against origin/main at 01f31af41.

(Written by Claude Opus 5.5)

🤖 Generated with Claude Code

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-29T23:06:19.662830Z	82cdc0a	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[kixelated]

Automated review by review (OpenAI)

Reviewed commit 82cdc0a. No actionable defect found in this planning change.

Direction: fail fast rather than skip preserves an honest gate result, and the condition that VM tests would actually run keeps the scope appropriate. For implementation acceptance, check usable KVM at the actual VM execution host, including permissions/sandbox access, rather than only local file existence. Keep non-VM platforms and valid cached/remote execution paths unaffected; missing required KVM must remain a failure, not a successful skip.

Limits: GitHub-only static review of the plan, #7175's reported failure and relevant CI context. No Nix/VM runs or host probes performed, and the reported timeout/CI capability were not independently reproduced. This reviews a plan, not an implemented or tested guard.

(Written by OpenAI)

[coderabbitai]

Walkthrough

The changes document the intended behavior for check flake on hosts without /dev/kvm: fail with a message naming the missing device rather than time out under software emulation. The plan says to fail instead of skip the check. It also records that the separate nested-shell BASH_ENV/PATH issue did not reproduce with codex-cli 0.162.1 and is out of scope. The README links to the plan.

Priority: Low

Merge Risk: Minimal · up to 51cfc

No concrete correctness, availability, security, or data-integrity risk was identified.

Pre-merge checks: 5

Check name	Status	Explanation
Title check		The title names the main change: a plan to make VM checks refuse to run on hosts without KVM. The raw summary and the PR objectives both describe a quest document that plans this change. The title is …
Description check		The description states the problem (the check boots NixOS VM tests under emulation without /dev/kvm and fails after about 19 minutes), the planned quest (fail at once with a message naming the missing…
Docstring Coverage		No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentat…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Title check

Explanation

The title names the main change: a plan to make VM checks refuse to run on hosts without KVM. The raw summary and the PR objectives both describe a quest document that plans this change. The title is short and specific. The 'chore(quest):' prefix is a scope tag and does not hurt clarity.

Full details: Description check

Explanation

The description states the problem (the check boots NixOS VM tests under emulation without /dev/kvm and fails after about 19 minutes), the planned quest (fail at once with a message naming the missing KVM), the Codex BASH_ENV follow-up that was not planned, and the docs-only impact. This matches the raw summary and the PR objectives.

Full details: Docstring Coverage

Explanation

No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 0 files. (2 skipped: 2 unsupported.)

Finishing Touches

✨ Simplify code

• Commit to this branch
• Create a new PR

• Autofix · 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.}

[kixelated]

Merge summary: this PR plans one m0/ci [XS] quest, quest/m0/ci/vm-kvm.md. It has no code changes.

• Reviews: the automated OpenAI review of 82cdc0a found no defects, and Codex gave a 👍. The follow-up push 51cfce0 only rewrites the note about the Codex BASH_ENV follow-up. It drew no new review, so the earlier review covers it.
• Checks: CI passed (11/11). quest check --base origin/main on head 51cfce0 passed (675 documents ok).
• Decisions: none outstanding.

(Written by Claude Opus 5.5)