| Filename | Latest commit message | Latest commit date |
|---|---|---|
Felipe M.
4d1af76510
All checks were successful
CI / goreleaser-lint (pull_request) Successful in 10s
CI / format (pull_request) Successful in 1m30s
CI / lint (pull_request) Successful in 4m43s
CI / test (pull_request) Successful in 6m59s
CI / build (pull_request) Successful in 6m25s
CI / e2e (pull_request) Successful in 25m10s
Retrying dockerd was a coin flip. containerd needs a little over ten seconds to boot in the runner's container and dockerd waits exactly ten, so every attempt loses that race when the host is busy: run 19 burned all five and the job waited three minutes for a daemon that never arrived. Starting containerd first and handing dockerd the socket removes the timeout from the picture. Measured on the runner: ready in 30 seconds, no retries. |
||
| .. | ||
| workflows | ||