zhuzhurk commented on a change in pull request #9675: [FLINK-13953] [runtime]
Facilitate enabling new scheduler in MiniCluster Tests
URL: https://github.com/apache/flink/pull/9675#discussion_r325292832
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/jobmaster/JobMaster.java
##########
@@ -810,15 +812,24 @@ private void startHeartbeatServices() {
log);
}
- private void assignScheduler(
+ private void reassignScheduler(
SchedulerNG newScheduler,
JobManagerJobMetricGroup newJobManagerJobMetricGroup) {
validateRunsInMainThread();
checkState(schedulerNG.requestJobStatus().isTerminalState());
checkState(jobManagerJobMetricGroup == null);
- schedulerNG = newScheduler;
- jobManagerJobMetricGroup = newJobManagerJobMetricGroup;
+ assignScheduler(newScheduler, newJobManagerJobMetricGroup);
+ }
+
+ private void assignScheduler(
+ SchedulerNG scheduler,
+ JobManagerJobMetricGroup jobManagerJobMetricGroup) {
+
+ this.schedulerNG = scheduler;
+ this.jobManagerJobMetricGroup = jobManagerJobMetricGroup;
+
+ log.info("Scheduler {} is used for job {}.", schedulerNG,
jobGraph.getJobID());
Review comment:
Agreed that we can infer from the cluster config log for the scheduler used.
But it is not very straight forward, especially when it is not configured and
the default config is used.
As we are already logging the failover strategy for each job, I think
logging the scheduler for each job is acceptable.
To avoid logging it for multiple times, we can just log it the JobMaster
constructor.
WDYT?
----------------------------------------------------------------
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:
us...@infra.apache.org
With regards,
Apache Git Services
