[
https://issues.apache.org/jira/browse/HBASE-12030?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14146157#comment-14146157
]
Anoop Sam John commented on HBASE-12030:
----------------------------------------
Thanks a lot Jingcheng for the test.
2 comments on the test
{code}
+ for (int i = 0; i < compactionThreshold; i++) {
+ Put p = new Put(Bytes.toBytes(i));
+ p.setDurability(Durability.SKIP_WAL);
+ p.add(COLUMN_FAMILY, Bytes.toBytes("colX"), dummyData);
+ loader.put(p);
+ loader.flushcache();
+ }
+ assertEquals("Before compaction: store files", compactionThreshold,
+ countStoreFiles());
+ assertEquals("Before compaction: mob file count", compactionThreshold,
+ countMobFiles());
+ Delete delete = new Delete(deleteRow);
+ delete.deleteFamily(COLUMN_FAMILY);
+ region.delete(delete);
+ loader.flushcache();
+ assertEquals("Before compaction: store files", compactionThreshold + 1,
+ countStoreFiles());
+ region.compactStores(true);
+ assertEquals("Before compaction: store files", 1, countStoreFiles());
{code}
We flush compactionThreshold with puts. This itself can trigger an auto
compaction (before the delete going in) So the assert
assertEquals(compactionThreshold + 1, countStoreFiles()) can get failed some
times. Better we add compactionThreshold -1 times put and one time delete and
then go with compaction.
assertEquals("Before compaction: store files", 1, countStoreFiles()); -> The
msg to be After compaction
> Wrong compaction report and assert when MOB compaction switches to minor
> ------------------------------------------------------------------------
>
> Key: HBASE-12030
> URL: https://issues.apache.org/jira/browse/HBASE-12030
> Project: HBase
> Issue Type: Bug
> Components: Compaction, regionserver
> Affects Versions: hbase-11339
> Reporter: Matteo Bertozzi
> Priority: Critical
> Fix For: hbase-11339
>
> Attachments: HBASE-12030-V2.diff, HBASE-12030.patch
>
>
> when zookeeper is down during a major compaction or a sweep tool run in
> progress, we switch to a minor.
> {code}
> try {
> zk = MobZookeeper.newInstance(this.conf, compactionName);
> } catch (KeeperException e) {
> LOG.error("Cannot connect to the zookeeper, ready to perform the minor
> compaction instead", e);
> // change the major compaction into a minor one
> compaction.getRequest().setIsMajor(false);
> return super.compact(compaction);
> }
> {code}
> but the "request start" (HRegion.reportCompactionRequestStart) is "major" and
> increments the major-compactions counter
> while the "request end" (HRegion.reportCompactionRequestEnd) is "minor" and
> decrements the minor-compactions counter
> triggering the assert newValue >= 0, since we are decrementing the wrong
> counter
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
