[
https://issues.apache.org/jira/browse/HBASE-5708?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13247095#comment-13247095
]
Phabricator commented on HBASE-5708:
------------------------------------
mbautin has commented on the revision "[jira] [HBASE-5708] [89-fb] Improving
robustness of map-reduce-related and other unit tests".
INLINE COMMENTS
src/main/java/org/apache/hadoop/hbase/master/OldLogsCleaner.java:98-100 Saw a
NullPointerException on the next line.
src/main/java/org/apache/hadoop/hbase/master/SplitLogManager.java:984-987 Saw
an infinite loop with a SESSIONEXPIRED here in TestSplitLogManager.
src/main/java/org/apache/hadoop/hbase/master/SplitLogManager.java:1070-1073
Saw an infinite loop with a SESSIONEXPIRED here in TestSplitLogManager.
src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java:1591
Previously, we would proceed to process the open region message, even though we
have re-queued it for later.
src/test/java/org/apache/hadoop/hbase/HBaseTestCase.java:52 A lot of stuff in
this deprecated class should eventually be delegated to HBaseTestingUtility.
src/test/java/org/apache/hadoop/hbase/HBaseTestingUtility.java:138-139
Sometimes we just want to reuse the mini-map-reduce-cluster-starting
capabilities of HBaseTestingUtility, but provide a NameNode URI ourselves.
src/test/java/org/apache/hadoop/hbase/regionserver/wal/TestHLog.java:364
Observed failures to start namenode's trash cleaning thread when it failed to
connect as a client to the newly started namenode. Adding a delay fixed the
problem in some of those cases.
src/test/java/org/apache/hadoop/hbase/util/TestMiniClusterLoadSequential.java:80
The load balancer would unassign some regions in a load-test unit test,
resulting in a failure, which is totally ridiculous.
src/test/resources/hbase-site.xml:145 A lot of tests used to fail with ZK
session expiration errors. I have set ZK session timeout to 1 sec in a few
places where it seemed necessary.
src/test/resources/hbase-site.xml:155 We did not have enough retries in some
unit tests. This will hopefully add some robustness.
REVISION DETAIL
https://reviews.facebook.net/D2601
> [89-fb] Make MiniMapRedCluster directory a subdirectory of target/test
> ----------------------------------------------------------------------
>
> Key: HBASE-5708
> URL: https://issues.apache.org/jira/browse/HBASE-5708
> Project: HBase
> Issue Type: Bug
> Reporter: Mikhail Bautin
> Priority: Minor
> Attachments: D2601.1.patch, D2601.2.patch, D2601.3.patch
>
>
> Some map-reduce-based tests are failing when executed concurrently in 89-fb
> because mini-map-reduce cluster uses /tmp/hadoop-<username> for temporary
> data.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
