[
https://issues.apache.org/jira/browse/CASSANDRA-12730?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15647254#comment-15647254
]
Benjamin Roth commented on CASSANDRA-12730:
-------------------------------------------
Thanks for your response.
My proposal was driven by the assumption that the session prepare causes most
of these flushes, but after reading some more code I also recognized that there
are a lot more occasions where flushes may occur - e.g. initiating a validation
compaction.
So probably I will first have to log flushes with traces to see where all these
flushes come from. Unfortunately I am currentyl on Web Summit with poor
connectivity, so I dont know when I will be able to set up logging / graphing
for switches and being able to produce an appropriate case.
The info you asked for:
- 7 Nodes
- 1 DC
- VNodes (256 tokens per node)
- Repair is managed by reaper with parallel, full, subrange (around 3500 ranges)
- RF 3
But I also got a question:
If a repair triggers a stream, isn't it going through the regular write path
like any other mutation (just to take care of index updates, MVs, ...)? To me
it also looks like that when looking at StreamReceiveTask. If so, many small
streams should not create many small SSTables if a flush is not forced by other
circumstances.
Thanks for your help!
> Thousands of empty SSTables created during repair - TMOF death
> --------------------------------------------------------------
>
> Key: CASSANDRA-12730
> URL: https://issues.apache.org/jira/browse/CASSANDRA-12730
> Project: Cassandra
> Issue Type: Bug
> Components: Local Write-Read Paths
> Reporter: Benjamin Roth
> Priority: Critical
>
> Last night I ran a repair on a keyspace with 7 tables and 4 MVs each
> containing a few hundret million records. After a few hours a node died
> because of "too many open files".
> Normally one would just raise the limit, but: We already set this to 100k.
> The problem was that the repair created roughly over 100k SSTables for a
> certain MV. The strange thing is that these SSTables had almost no data (like
> 53bytes, 90bytes, ...). Some of them (<5%) had a few 100 KB, very few (<1%
> had normal sizes like >= few MB). I could understand, that SSTables queue up
> as they are flushed and not compacted in time but then they should have at
> least a few MB (depending on config and avail mem), right?
> Of course then the node runs out of FDs and I guess it is not a good idea to
> raise the limit even higher as I expect that this would just create even more
> empty SSTables before dying at last.
> Only 1 CF (MV) was affected. All other CFs (also MVs) behave sanely. Empty
> SSTables have been created equally over time. 100-150 every minute. Among the
> empty SSTables there are also Tables that look normal like having few MBs.
> I didn't see any errors or exceptions in the logs until TMOF occured. Just
> tons of streams due to the repair (which I actually run over cs-reaper as
> subrange, full repairs).
> After having restarted that node (and no more repair running), the number of
> SSTables went down again as they are compacted away slowly.
> According to [~zznate] this issue may relate to CASSANDRA-10342 +
> CASSANDRA-8641
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
