XComp commented on a change in pull request #15170:
URL: https://github.com/apache/flink/pull/15170#discussion_r594307989
##########
File path:
flink-tests/src/test/java/org/apache/flink/test/recovery/ProcessFailureCancelingITCase.java
##########
@@ -221,15 +210,11 @@ public Long map(Long value) throws Exception {
final Collection<JobID> jobIds = waitForRunningJobs(clusterClient,
timeout);
assertThat(jobIds, hasSize(1));
- final JobID jobId = jobIds.iterator().next();
// kill the TaskManager after the job started to run
taskManagerProcess.destroy();
taskManagerProcess = null;
- // try to cancel the job
- clusterClient.cancel(jobId).get();
-
// we should see a failure within reasonable time (10s is the ask
timeout).
// since the CI environment is often slow, we conservatively give
it up to 2 minutes,
// to fail, which is much lower than the failure time given by the
heartbeats ( > 2000s)
Review comment:
Ah, that one I missed. But then, it would be `1000ms` instead of `>
2000s`...
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
