zentol commented on code in PR #22251:
URL: https://github.com/apache/flink/pull/22251#discussion_r1147690649
##########
flink-runtime/src/main/java/org/apache/flink/runtime/jobmanager/DefaultJobGraphStore.java:
##########
@@ -241,6 +243,22 @@ public void putJobGraph(JobGraph jobGraph) throws
Exception {
LOG.info("Added {} to {}.", jobGraph, jobGraphStateHandleStore);
}
+ @Override
+ public void putJobResourceRequirements(
+ JobID jobId, JobResourceRequirements jobResourceRequirements)
throws Exception {
+ synchronized (lock) {
+ @Nullable final JobGraph jobGraph = recoverJobGraph(jobId);
+ if (jobGraph == null) {
+ throw new FileNotFoundException(
+ String.format(
+ "JobGraph for job [%s] was not found in
JobGraphStore and is needed for attaching JobResourceRequirements.",
+ jobId));
+ }
+ JobResourceRequirements.writeToJobGraph(jobGraph,
jobResourceRequirements);
+ putJobGraph(jobGraph);
Review Comment:
That can indeed happen. It's unlikely to be a _problem_ though, because even
in the worst case (i.e., a JM is writing the requirements while loosing
leadership, and another one gains leadership and starts reading job graphs
before that write is complete, which is already unlikely) you we might "forget"
the last set requirements after a failover.
--
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]
