[
https://issues.apache.org/jira/browse/HBASE-17081?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15777945#comment-15777945
]
Anastasia Braginsky commented on HBASE-17081:
---------------------------------------------
Hey!
We found the problem with TestAsyncGetMultiThread. This was a rare concurrent
scenario where the segments were hidden for a very very short moment. When
requested to flush to disk the segments are first removed from the pipeline to
a local variable, then added to the snapshot. So there is a short moment when
CompactingMemStore's segments are not seen to the scanners/getters. We created
HBASE-17373 to resolve this issue and also to add 3 variants of
TestAsyncGetMultiThread test for NONE/BASIC/EAGER MemStores.
Anyway this problem wasn't introduced by this JIRA and TestAsyncGetMultiThread
is not failing now when after HBASE-17333 the default was set to be NONE.
Please consider committing the latest patch once again. I think the re-base is
yet not necessary and the last patch can still be committed. Please contact me
if re-base is still needed.
> Flush the entire CompactingMemStore content to disk
> ---------------------------------------------------
>
> Key: HBASE-17081
> URL: https://issues.apache.org/jira/browse/HBASE-17081
> Project: HBase
> Issue Type: Sub-task
> Reporter: Anastasia Braginsky
> Assignee: Anastasia Braginsky
> Attachments: HBASE-15787_8.patch, HBASE-17081-V01.patch,
> HBASE-17081-V02.patch, HBASE-17081-V03.patch, HBASE-17081-V04.patch,
> HBASE-17081-V05.patch, HBASE-17081-V06.patch, HBASE-17081-V06.patch,
> HBASE-17081-V07.patch, HBaseMeetupDecember2016-V02.pptx,
> Pipelinememstore_fortrunk_3.patch
>
>
> Part of CompactingMemStore's memory is held by an active segment, and another
> part is divided between immutable segments in the compacting pipeline. Upon
> flush-to-disk request we want to flush all of it to disk, in contrast to
> flushing only tail of the compacting pipeline.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
