BukrosSzabolcs commented on a change in pull request #3786:
URL: https://github.com/apache/hbase/pull/3786#discussion_r742685268
##########
File path:
hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/compactions/Compactor.java
##########
@@ -348,8 +353,13 @@ private InternalScanner
postCompactScannerOpen(CompactionRequestImpl request, Sc
smallestReadPoint = Math.min(fd.minSeqIdToKeep, smallestReadPoint);
cleanSeqId = true;
}
+ if (writer != null){
+ LOG.warn("Writer exists when it should not: " +
getCompactionTargets().stream()
+ .map(n -> n.toString())
+ .collect(Collectors.joining(", ", "{ ", " }")));
Review comment:
That I missed a point where writer reset should have been added or code
was changed that allowed this to happen. As far as I can tell this can not
happen now and it would be harmless even if it could happen. The worst case
scenario is that the cleaner would skip this file once because it would assume
the compaction is still ongoing. But it felt wrong not to log a warn before
overwriting the writer.
--
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]
