lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Muir (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (LUCENE-4484) NRTCachingDir can't handle large files
Date Tue, 16 Oct 2012 13:25:04 GMT

    [ https://issues.apache.org/jira/browse/LUCENE-4484?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13476981#comment-13476981
] 

Robert Muir commented on LUCENE-4484:
-------------------------------------

I know it does: I think a much safer general solution to keep e.g. file counts low would be
to just match the lucene defaults:
FSDirectory.open and CFS enabled.

I tend to agree with Mike on "NRTCachingDirectory can really be especially for the NRT use
case"
because otherwise I think its going to be ugly to make it work well for all use-cases... and

even then not OOM'ing doesnt necessarily mean working well. If its always overflowing its
cache
and having to uncache files because its not really an NRT use case that doesn't seem great.

But i don't disagree with trying to make it more general either, I do just think that this
should be done in NRTCachingDir
itself and not hacked into indexwriter (flushing when stored files get too large is illogical
outside of hacking around this)

                
> NRTCachingDir can't handle large files
> --------------------------------------
>
>                 Key: LUCENE-4484
>                 URL: https://issues.apache.org/jira/browse/LUCENE-4484
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Michael McCandless
>
> I dug into this OOME, which easily repros for me on rev 1398268:
> {noformat}
> ant test  -Dtestcase=Test4GBStoredFields -Dtests.method=test -Dtests.seed=2D89DD229CD304F5
-Dtests.multiplier=3 -Dtests.nightly=true -Dtests.slow=true -Dtests.linedocsfile=/home/hudson/lucene-data/enwiki.random.lines.txt
-Dtests.locale=ru -Dtests.timezone=Asia/Vladivostok -Dtests.file.encoding=UTF-8 -Dtests.verbose=true
> {noformat}
> The problem is the test got NRTCachingDir ... which cannot handle large files because
it decides up front (when createOutput is called) whether the file will be in RAMDir vs wrapped
dir ... so if that file turns out to be immense (which this test does since stored fields
files can grow arbitrarily huge w/o any flush happening) then it takes unbounded RAM.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message