lxbme commented on PR #13078: URL: https://github.com/apache/apisix/pull/13078#issuecomment-4125078446
Hi @Baoyuantop, I've checked the CI logs in detail. The failure is **not related to my changes**, it occurs in `test_prometheus_stream.sh`, not in `test_standalone.sh`. ### Failure details The failing assertion (from the [job log](https://github.com/apache/apisix/actions/runs/23145486176/job/68498654676)): - grep 'apisix_stream_connection_total{route="1"} 1' - echo 'failed: prometheus can'''t work in stream subsystem' - exit 1 In stream-only mode (`proxy_mode: stream`, `enable_admin: false`), the prometheus metrics output does not contain `apisix_stream_connection_total`, causing the grep to fail. Since the CI runner executes `t/cli/test_*.sh` in alphabetical order, `test_prometheus_stream.sh` fails before `test_standalone.sh` ever runs — so my code was never reached. ### Same failure exists on master The most recent [CLI Test run on master](https://github.com/apache/apisix/actions/runs/23284595409) (`chore(deps): bump actions/cache #13106`) shows the exact same pattern: - `linux_apisix_current_luarocks` → **failure** - `linux_apisix_current_luarocks_in_customed_nginx` → **success** This confirms it is a pre-existing issue unrelated to this PR. Could you please re-run the failed job to see if it passes? The `test_prometheus_stream.sh` failure appears to be intermittent on master (roughly 2 out of 5 recent master runs fail with the same error). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
