danny0405 commented on code in PR #18790:
URL: https://github.com/apache/hudi/pull/18790#discussion_r3300893988
##########
hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/sink/partitioner/BucketAssignFunction.java:
##########
@@ -145,6 +145,26 @@ public void open(Configuration parameters) throws
Exception {
this.recordProcessor = initRecordProcessor();
this.metrics = new
FlinkBucketAssignMetrics(getRuntimeContext().getMetricGroup());
this.metrics.registerMetrics();
+ initRliShardAssignMetric();
+ }
+
+ /**
+ * Computes and registers the number of RLI shards assigned to this task
when global RLI is active.
+ * Each task owns the file groups whose index satisfies {@code fgIndex %
numPartitions == taskIndex}.
+ */
+ private void initRliShardAssignMetric() {
+ if (!OptionsResolver.isGlobalRecordLevelIndex(conf)) {
+ return;
+ }
+ try {
+ int numFileGroups =
GlobalRecordIndexPartitioner.fetchNumFileGroupsForRecordIndexPartition(conf);
Review Comment:
The JVM lifecycle cache is risky if the mdt shards changed, would incur
errors, let's remove the cache since it is a one-shot action.
--
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]
