[
https://issues.apache.org/jira/browse/HBASE-19658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16344941#comment-16344941
]
Anastasia Braginsky commented on HBASE-19658:
---------------------------------------------
Hey [~ram_krish],
If the sizes are deterministic, that could be correct the explanation that you
are giving. But this is not what do we see in debugging. According to sizes
that we see, the single cell is not big enough to be flushed in memory. At
least it shouldn't according to our debugging and calculations. So from out
point of view the test flow is deterministic and shouldn't be flaking.
{quote}By the time the runnable actually runs the 3rd cell also gets added in
some cases and hence you get issues in the size accounting.
{quote}
The question is exactly what are those "some cases"? According to what we see
it should go like this: first cell is added, nothing happens. Then second cell
is added and flush in memory happens. Only then the in-memory threshold is
passed. Indeed there is in-memory-flush on separate thread while third cell is
added. However, third cell shouldn't cause in-memory flush. So at the end two
cells in pipeline and one in active, no flakiness.
So from here is my question, how come to you see the third cell is causing in
memory flush? What is the cell size that you see and what is the
in-memory-flush-threshold that you see?
> Fix and reenable
> TestCompactingToCellFlatMapMemStore#testFlatteningToJumboCellChunkMap
> --------------------------------------------------------------------------------------
>
> Key: HBASE-19658
> URL: https://issues.apache.org/jira/browse/HBASE-19658
> Project: HBase
> Issue Type: Bug
> Components: test
> Affects Versions: 2.0.0-beta-1
> Reporter: stack
> Assignee: Anastasia Braginsky
> Priority: Major
> Fix For: 2.0.0-beta-2
>
> Attachments: HBASE-19658-V01.patch, HBASE-19658-V02.patch,
> HBASE-19658-V03.patch, HBASE-19658-V04.patch, HBASE-19658-V05.patch,
> HBASE-19658.0007.patch, HBASE-19658.006.patch, HBASE-19658.05.patch,
> org.apache.hadoop.hbase.regionserver.TestCompactingToCellFlatMapMemStore-output.txt
>
>
> testFlatteningToJumboCellChunkMap was disabled so could commit HBASE-19282 on
> branch-2. This test is failing reliably. Assigned to [~anastas]. This issue
> is about fixing the failing test and reenabling it in time for beta-2. Thanks
> A.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
