Description
Summary
The gosec (insecure patterns) job in the Go Security Analysis workflow has failed at least six times on 2026-10-31 with the same annotation and no code cause:
The self-hosted runner lost communication with the server. Verify the machine is running and has a healthy network connection. Anything in your workflow that terminates the runner process, starves it for CPU/Memory, or blocks its network access can cause this error.
In every case the job logs are BlobNotFound (the runner died before uploading), the sibling jobs in the same workflow run (govulncheck, golangci-lint, action pins resolve, NOTICE matches the module graph) completed green, and the same PR head passed gosec on a later or earlier attempt where one exists.
Evidence
| Ref | Run | Job |
|---|---|---|
v5 @ de6127e |
— | runner lost (bead gh-v5-gosec) |
| #16014 attempt 1 | 41198755355 | 124530335184 |
| #16014 attempt 2 | 41200489328 | 124538397043 |
| #16007 | 41197691718 | 124527124413 |
| #16031 | 41200277080 | — |
| #16027 | 41201935213 | 124542607459 |
Meanwhile v5 runs 41199008359 → 41202519218 (six consecutive) are green, so this is not a code regression on the default branch — it is the gosec job specifically losing its self-hosted runner, repeatedly.
Why this needs a signal
gosec (insecure patterns)is a required status check, so a runner loss blocks the merge of an otherwise-green PR (🌱 chore(rfcnotify): add tests for card.go helper functions #16014 is stuck on it right now with every other check passing).- Agents cannot
gh run rerun(out of tier), so a PR hit by this has no path back to green other than a new push. - Only gosec is affected among the four jobs in the workflow, which points at that job's resource profile (gosec on the full
./...tree is memory-heavy) rather than at the runner pool in general.
Asks
- Capture the runner-side evidence (OOM / eviction / network) for one of the job ids above so the cause is known rather than guessed.
- Either give the gosec job a
retry/rerun-failed-jobs guard for the "runner lost communication" annotation, or move it to a runner class with more memory, or scopegosecper package shard like the tests. - Until then, treat
gosec (insecure patterns)failures carrying this annotation as infra, not as a code verdict.
Filed from the scanner lane during CI repair of #16014.
— hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88
Activity
hivecommons-hive commentedon Oct 31, 2026
🔒 Claimed by @Karibhou (contributor on hive hosted-available-oke-11-placeholder-r05x). The claim lasts 30m0s without activity and renews while the work continues. Reply
/unclaimto release, or/claimto take it over if you outrank the holder.hivecommons-hive commentedon Oct 31, 2026
🔓 Claim by @Karibhou released (lease revoked).
🐛 fix(ci): bound gosec concurrency so Scan stops losing its runner
🐛 fix(ci): regenerate integration-candidate.yml for the gosec concurr…
🐛 fix(ci): bound gosec concurrency so Scan stops losing its runner (#…
github-actions commentedon Nov 2, 2026
@hivecommons-hive[bot] this issue was closed by @kubestellar-prow[bot] as completed.
If this was closed by a merged PR, the fix only takes effect once it ships in a tagged release and your deployment updates to it — check your running version against the fix before concluding the fix failed.
If you are still seeing the problem, comment
/reopenand it will be reopened for you — you do not need write access. Please add what you are still observing (running version, logs, steps) so it is not closed the same way again.