hutiefang76 commented on PR #12943: URL: https://github.com/apache/seatunnel/pull/12943#issuecomment-6341578489
I hit the null-log case in the existing single-master recovery test on current dev plus my Protobuf patch: [JDK 17 job](https://github.com/hutiefang76/seatunnel/actions/runs/39772122616/job/119806030508), source `91d1078a`. `KubernetesApplicationRecoveryIT` fails at the initial marker wait (`livePodLogs(...).contains(...)`, line 65), before the checkpoint and worker-failure steps. The suite reports 11 tests, 1 error. The recovery test and shared helper are identical to dev `4dcfa28c`. Your `retainedPodLogs` null-to-empty change in `a5230b0` already covers that path. Would you be open to extracting that small test-helper fix ahead of the HA changes? The same run also loses diagnostic output when `recordApplicationDiagnostics` calls `logs.getBytes(...)` on null. That call is still unguarded in this PR, as is the new `Files.writeString(..., previous)` path. It would help to handle missing logs there too, so one unavailable container log does not stop diagnostics for the remaining Pods. I have not run this PR or verified the recovery stages; this is evidence from the existing single-master suite. -- 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]
