clintropolis commented on code in PR #19625:
URL: https://github.com/apache/druid/pull/19625#discussion_r3469865443
##########
server/src/main/java/org/apache/druid/segment/metadata/AbstractSegmentMetadataCache.java:
##########
@@ -717,13 +718,42 @@ public Set<SegmentId> refreshSegments(final
Set<SegmentId> segments) throws IOEx
.add(segmentId);
}
+ // Refresh each dataSource independently so one failure (e.g. a metadata
query timeout) does not
+ // abort the cycle and starve the rest.
for (Map.Entry<String, TreeSet<SegmentId>> entry : segmentMap.entrySet()) {
- updatedSegmentIds.addAll(refreshSegmentsForDataSource(entry.getKey(),
entry.getValue()));
+ final String dataSource = entry.getKey();
+ try {
+ updatedSegmentIds.addAll(refreshSegmentsForDataSource(dataSource,
entry.getValue()));
+ }
+ catch (QueryInterruptedException e) {
+ // QueryInterruptedException also wraps ordinary query failures, not
just interruption
+ if (e.getCause() instanceof InterruptedException) {
+ Thread.currentThread().interrupt();
Review Comment:
does this not record the failure on purpose? i guess to not emit a metric
for shutdown? maybe worth a comment
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
