xintongsong commented on code in PR #20042:
URL: https://github.com/apache/flink/pull/20042#discussion_r903476652
##########
flink-clients/src/main/java/org/apache/flink/client/cli/CliFrontend.java:
##########
@@ -208,6 +210,16 @@ protected void runApplication(String[] args) throws
Exception {
Collections.singletonList(uri.toString()));
}
+ // In HA mode, we only support single-execute jobs at the moment.
Here, we manually
+ // generate the job id, if not configured, to keep it consistent
across failover.
+ if
(HighAvailabilityMode.isHighAvailabilityModeActivated(effectiveConfiguration)
+ && !effectiveConfiguration
+
.getOptional(PipelineOptionsInternal.PIPELINE_FIXED_JOB_ID)
+ .isPresent()) {
+ effectiveConfiguration.set(
+ PipelineOptionsInternal.PIPELINE_FIXED_JOB_ID, new
JobID().toHexString());
+ }
+
Review Comment:
I'm not sure `CliFrontend` is the best place to set this job id. It's not
the only entry point that an application can be deployed. AFAIK, Flink
Kubernetes Operator uses `ApplicationDeployer` directly. Maybe we should move
this logic to `ApplicationDeployer`.
--
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]
