ferenc-csaky commented on code in PR #27539:
URL: https://github.com/apache/flink/pull/27539#discussion_r3007706626
##########
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/adaptive/Executing.java:
##########
@@ -95,18 +98,50 @@ class Executing extends StateWithExecutionGraph
this.rescaleOnFailedCheckpointCount = rescaleOnFailedCheckpointCount;
this.failedCheckpointCountdown = null;
+ recordRescaleForJobIntoExecuting(logger, context);
+
deploy();
// check if new resources have come available in the meantime
context.runIfState(
this,
() -> {
- stateTransitionManager.onChange();
+ stateTransitionManager.onChange(true);
stateTransitionManager.onTrigger();
},
Duration.ZERO);
}
+ private void recordRescaleForJobIntoExecuting(Logger logger, Context
context) {
+ // As described at
https://github.com/apache/flink/pull/27539#discussion_r3006364815.
Review Comment:
I'd rather not reference the PR comment here, but instead make this comment
contain the whole context, which is almost already true, maybe:
```
// When "AdaptiveScheduler -> stopWithSavepoint" fails, the job is not
stopped, but
// instead transitions back to the "Executing" state. In this case, there is
no
// parallelism change, hence we ignore it for rescale.
```
--
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]
