clang-tidy flake #546
Comments
|
So personally, I'm observing about 30% of these clang-tidy CI jobs to fail. One case of a timeout observed, where clang-tidy seemed to hang, and a few other cases where clang-tidy wrote a diff, but there was no trace of an actual complaint in the logs. I'll ask around on Slack if this is a known issue for envoy proxy in general, as I suspect that this crept in via an Envoy update, or an envoy build image update (which we usually update at the same time). |
|
The last merge to master also ran into this, the logs ended with: |
|
The latest merge on master (61996df) needed 4 tries before succeeding the clang-tidy CI check. |
|
In #566 this seems to happen more or less structurally. I can't spot an actual problem, and neither does clang_tidy report an actual problem. But somehow the CI task keeps exiting with an exit code that indicates a problem. |
At some point clang-tidy runs broke in CI, and this couldn't be reproduced locally. The after a while local runs would exit with a failure seemingly mid- run. This PR addresses both these things. - Update the CircleCI SHA of the envoy build docker image. - Sync the script we use to run Envoy's build docker image. - Make clang-tidy runs use libstdc++ to avoid standard library headers from producing errors. - Sync run_clang_tidy.sh - Update MAINTAINERS.md to mention we should keep more things in sync. Fixes #546 Signed-off-by: Otto van der Schaaf <[email protected]>
In #533 I observed a first occurrency of a non-reproducible clang-tidy flake.
Since this CI task has proven to be quite stable over time, it's worth tracking further occurrences of this here.
If this turns out problematic, the first potential suspect I would investigate is the last Envoy update, which at this time is 8fcf416
The text was updated successfully, but these errors were encountered: