You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
π« Renovate PR #313 blacklisted after 7 CI failures: snap: bump version to llama.cpp b10549Β #314
After 7 CI retry attempts (3 automated watchdog retries + 4 AI-assisted retries), this Renovate PR continues to fail CI. The PR has been blacklisted and will no longer be automatically processed.
Failure Analysis
All CI failures on this PR have been infrastructure/transient in nature β not caused by the dependency bump itself. The build jobs (amd64 and arm64) have consistently passed successfully. The failures occur exclusively in the edge-tests jobs (both amd64 and arm64) during the WireGuard + LXD + spread test setup phase.
The failure annotation on the most recent run (ID 32470153884) reads:
"The hosted runner lost communication with the server. Anything in your workflow that terminates the runner process, starves it for CPU/Memory, or blocks its network access can cause this error."
This is a recurring GitHub Actions runner connectivity issue affecting the WireGuard VPN setup used for spread tests. The llama.cpp bump from b10486 β b10549 is a clean Renovate PR with no source code changes.
Last Failure Log (excerpt)
Run ID: 32470153884
Job: edge-tests (arm64, ubuntu-24.04-arm) / test β FAILED after 46m1s
Job: edge-tests (amd64, ubuntu-latest) / test β FAILED after 49m1s
Annotation: The hosted runner lost communication with the server.
Anything in your workflow that terminates the runner process,
starves it for CPU/Memory, or blocks its network access can cause this error.
Note: build (arm64) β and build (amd64) β both passed successfully.
Recommended Action
A human maintainer should:
Investigate the WireGuard/LXD spread test infrastructure β the runner keeps losing connectivity during the ~45β50 minute test window.
Consider merging this PR manually after verifying the build artifacts look correct (both build jobs passed consistently), since the failures are infra-related and not caused by the bump itself.
Consider whether the edge-tests spread suite needs a timeout adjustment or runner resource upgrade to prevent communication loss.
This issue was automatically created by Moon Skywalker (Renovate PR Maintenance Bot). Timestamp: 2026-08-21T18:25:00+00:00
Activity