[HBASE-5114] TestMetaReaderEditor can timeout during test execution

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.94.0
• Component/s: test
• Hadoop Flags: Reviewed
• Assignee: Nicolas Liochon
• Reporter: Nicolas Liochon
• Created: 01/Jan/12 16:46
• Updated: 12/Jun/22 20:02
• Resolved: 02/Jan/12 05:22

Description

Likely because it belongs to the too long list of flaky tests. A test should not fail randomly. Here is the list, from 2974 to 2598, with a timeouted TestMetaReaderEditor
https://builds.apache.org/job/HBase-TRUNK/2597/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2588/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2585/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2584/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2582/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2577/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/HBase-TRUNK/2574/testReport/org.apache.hadoop.hbase.catalog/

It happens on hadoop-qa as well:
https://builds.apache.org/job/PreCommit-HBASE-Build/640/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/PreCommit-HBASE-Build/638/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/PreCommit-HBASE-Build/630/testReport/org.apache.hadoop.hbase.catalog/
https://builds.apache.org/job/PreCommit-HBASE-Build/622/testReport/org.apache.hadoop.hbase.catalog/

Looking at the source code, main issues are:

• in testRetrying, some threads are started, but will not be stopped if an assertion fails. As a consequence, the process will not stop
• still in testRetrying, it's not possible to use @Test(timeout=180000), because if this timeout is reached, the thread won't be stopped.

This does not explain why the test fails sometimes, but explain why we have a timeout instead of a clean failure.

I also fixed some stuff that seems unrelated, and I cannot reproduce the error anymore locally after 10 tries. So I will try the patch on hadoop-qa.

The patch includes as well the removal of TestAdmin#testHundredsOfTable because this test takes 2 minutes to execute; but proves nothing on success as there is no check on ressources used.

Issue Links

• relates to: HBASE-5064 utilize surefire tests parallelization
• relates to: HBASE-4602 Make the suite run in at least half the time