jiangxin369 commented on code in PR #24082:
URL: https://github.com/apache/flink/pull/24082#discussion_r1452878415
##########
tools/azure-pipelines/build-apache-repo.yml:
##########
@@ -73,6 +73,17 @@ stages:
run_end_to_end: false
container: flink-build-container
jdk: 8
+ - template: jobs-template.yml
+ parameters:
+ stage_name: ci-adaptive-scheduler
+ test_pool_definition:
+ name: Default
+ e2e_pool_definition:
+ vmImage: 'ubuntu-20.04'
+ environment: PROFILE="-Penable-adaptive-scheduler"
Review Comment:
Shall we add the `-Dflink.hadoop.version=2.10.2` in the profile to keep
consistency with other stages? We can also remove the profile from other stages
If `2.10.2` is the default value.
##########
flink-runtime/src/test/java/org/apache/flink/runtime/minicluster/MiniClusterITCase.java:
##########
@@ -149,7 +174,15 @@ private void runHandleJobsWhenNotEnoughSlots(final
JobGraph jobGraph) throws Exc
try (final MiniCluster miniCluster = new MiniCluster(cfg)) {
miniCluster.start();
- miniCluster.executeJobBlocking(jobGraph);
+ assertThatThrownBy(() -> miniCluster.executeJobBlocking(jobGraph))
+ .isInstanceOf(JobExecutionException.class)
+ .hasMessageContaining("Job execution failed")
+ .extracting(Throwable::getCause)
+ .extracting(FlinkAssertions::chainOfCauses,
FlinkAssertions.STREAM_THROWABLE)
+ .anySatisfy(
+ cause ->
+ assertThat(cause)
+
.isInstanceOf(NoResourceAvailableException.class));
Review Comment:
Would it be more explicit if we assert the TimeoutException is thrown when
running `testHandlingNotEnoughSlotsThroughTimeout` with the default scheduler?
--
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]
