[
https://issues.apache.org/jira/browse/HBASE-9338?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13756918#comment-13756918
]
Elliott Clark commented on HBASE-9338:
--------------------------------------
Still seeing this 100% of the time on out test cluster.
Running:
{code}
hbase org.apache.hadoop.hbase.test.IntegrationTestBigLinkedList --monkey
slowDeterministic Loop 5 12 25000000 IntegrationTestBigLinkedList 12
{code}
Fails every time. Usually it will get through the first few iterations but it
will fail before the 5th. Resulting in:
{code}
org.apache.hadoop.hbase.test.IntegrationTestBigLinkedList$Verify$Counts
REFERENCED=599327221
UNDEFINED=336384
UNREFERENCED=336384
{code}
{code}
org.apache.hadoop.hbase.test.IntegrationTestBigLinkedList$Verify$Counts
REFERENCED=899938580
UNDEFINED=29048
UNREFERENCED=32373
{code}
Disabling online schema change made Load and Verify pass. However there is
still something wrong. I went through the MR output and the keys listed as
undefined are really missing (going on the shell they are never returned). So
it looks like we are missing single links in the chain.
I've modified the job to produce a more usable client column. I've used that
to look at the logs for each attempt. Nothing seems too much amiss in the
logs. Can someone run the same tests with a high number of iterations (Seems
this always passes on the first loop) ?
> Test Big Linked List fails on Hadoop 2.1.0
> ------------------------------------------
>
> Key: HBASE-9338
> URL: https://issues.apache.org/jira/browse/HBASE-9338
> Project: HBase
> Issue Type: Bug
> Components: test
> Affects Versions: 0.96.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Priority: Blocker
> Fix For: 0.96.0
>
> Attachments: HBASE-9338-0.patch
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
