[
https://issues.apache.org/jira/browse/HBASE-13531?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14522970#comment-14522970
]
Jingcheng Du commented on HBASE-13531:
--------------------------------------
Hi Jon.
I ran this TestAcidGuarantees for 20 times, and didn't encounter any issues,
and ran the IT with useMob=true, and didn't encounter the issue either.
However, I have some thoughts on this issue.
Now HBase has a new flush policy, each region flush might only flush part of
stores. For all cells (not only for mob), the following case is possible?
# One put of row0 has two cells row0,cf0:q0,ts0=value0 and
row0,cf1:q0,ts0=value0 into memstore.
# Scanner is opened.
# Now the memstore is flushed, only cf1 is flushed.
# Now the cell row0:cf1:q0:ts0 is in store file, row0:cf0:q0:ts0 is still in
memstore.
# Another put of row0 has two cells row0:cf0:q0:ts0=value1 and
row0:cf1:q0:ts0=value1 into memstore.
# Now what can scanner see? The row0:cf0:q0:ts0=value0 had been replaced by
row0:cf0:q0:ts0=value1 in memstore, but the scanner can not read it because of
the readPt, but row0:cf1:q0:ts0=value0 is still visible. The ACID is violated?
Is this possible?
> After 4/18/15 merge, flakey failures of
> TestAcidGuarantees#testMobScanAtomicity
> -------------------------------------------------------------------------------
>
> Key: HBASE-13531
> URL: https://issues.apache.org/jira/browse/HBASE-13531
> Project: HBase
> Issue Type: Sub-task
> Components: regionserver, Scanners
> Affects Versions: hbase-11339
> Reporter: Jonathan Hsieh
> Fix For: hbase-11339
>
>
> After the merge of master from 4/18/15 with hbase-11339 branch, we encounter
> some atomicity violations. We want to fix before calling merge to trunk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
