[
https://issues.apache.org/jira/browse/HBASE-5569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13232341#comment-13232341
]
Lars Hofhansl commented on HBASE-5569:
--------------------------------------
I spent a lot of more time looking at this. I thought it might be due to the
flushes being executed in parallel by multiple threads, but synchronizing this
part made the failure more likely!
Doing this and increasing the frequency of flushes reproduces the problem
multiple times on every test run now, which it good.
But... My initial hunch was correct. When I enable KEEP_DELETED_CELLS on the
store the problem goes away!
Hence this definitely has to do with collection of deletes and delete markers.
> Do not collect deleted KVs when they are still in use by a scanner.
> -------------------------------------------------------------------
>
> Key: HBASE-5569
> URL: https://issues.apache.org/jira/browse/HBASE-5569
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Fix For: 0.94.0, 0.96.0
>
> Attachments: 5569-v2.txt, 5569.txt,
> TestAtomicOperation-output.trunk_120313.rar
>
>
> I noticed this because TestAtomicOperation.testMultiRowMutationMultiThreads
> fails rarely.
> The solution is similar to HBASE-2856, where expired KVs are not collected
> when in use by a scanner.
> ---
> What I pieced together so far is that it is the *scanning* side that has
> problems sometimes.
> Every time I see a assertion failure in the log I see this before:
> {quote}
> 2012-03-12 21:48:49,523 DEBUG [Thread-211] regionserver.StoreScanner(499):
> Storescanner.peek() is changed where before =
> rowB/colfamily11:qual1/75366/Put/vlen=6,and after =
> rowB/colfamily11:qual1/75203/DeleteColumn/vlen=0
> {quote}
> The order of if the Put and Delete is sometimes reversed.
> The test threads should always see exactly one KV, if the "before" was the
> Put the thread see 0 KVs, if the "before" was the Delete the threads see 2
> KVs.
> This debug message comes from StoreScanner to checkReseek. It seems we still
> some consistency issue with scanning sometimes :(
--
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
