[
https://issues.apache.org/jira/browse/SOLR-4802?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13652632#comment-13652632
]
Yonik Seeley commented on SOLR-4802:
------------------------------------
bq. The current work around is that i have auto-commit setup for 10000 docs
with openSearcher = false.
A lack of commits or a lack of any autocommits is not really valid with
transaction logging enabled.
This comes up enough that it seems like we should have some mechanism to
prevent unbounded tlog growth regardless of autocommit settings (such as
enabling an autocommit even if one is not configured, or some other similar
mechanism)
> Atomic Updated for large documents is very slow and at some point Server
> stops responding.
> -------------------------------------------------------------------------------------------
>
> Key: SOLR-4802
> URL: https://issues.apache.org/jira/browse/SOLR-4802
> Project: Solr
> Issue Type: Bug
> Affects Versions: 4.2.1
> Environment: Jboss 5.1 with Solr 4.2.1 and JDk 1.6.0_33u
> Reporter: Aditya
> Priority: Critical
>
> I am updating three fields in the document which are of type long and float.
> This is an atomic update. Updating around 30K doucments and i always get
> stuck after 80%.
> Update 200 docs per request in a thread and i execute 5 such thread in
> parallel.
> The current work around is that i have auto-commit setup for 10000 docs with
> openSearcher = false.
> i guess that this is related to SOLR-4589
> Some additional information
> the machine is 6core with 5GB Heap. ramBuffer=1024MB
--
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-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
