[ 
https://issues.apache.org/jira/browse/SOLR-4878?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Shawn Heisey updated SOLR-4878:
-------------------------------

    Attachment: SOLR-4878.patch

First patch doesn't work.  The dataDir field in SolrIndexConfig is null when 
the infostream config is processed.  This is because the dataDir field only 
refers to a dataDir element in the solrconfig.xml.  That means that the patch 
(as written) would only be useful if dataDir is defined in solrconfig.xml, but 
in a multicore world, that is rare.

This new patch uses a different approach.  There's one detail that I don't like 
- because the only path information I found to be available was indexDir, I 
used ".." to construct the final path.  It seems like a bad idea to clutter the 
actual index directory with the infostream file, but that would make the code 
cleaner.
                
> Change default directory for infostream from CWD to dataDir
> -----------------------------------------------------------
>
>                 Key: SOLR-4878
>                 URL: https://issues.apache.org/jira/browse/SOLR-4878
>             Project: Solr
>          Issue Type: Bug
>    Affects Versions: 4.3
>            Reporter: Shawn Heisey
>            Assignee: Shawn Heisey
>             Fix For: 5.0, 4.4
>
>         Attachments: SOLR-4878.patch, SOLR-4878.patch
>
>
> The default directory for the infoStream file is CWD.  In a multicore system 
> where all the cores share similar configs, the output from all cores is 
> likely to end up in the same file.  Although this is sometimes the desired 
> outcome, it seems less than ideal.  If you've got cores that literally share 
> the same config file, or you're using SolrCloud where more than one core on 
> the system uses the same config set, you won't have the option of putting 
> different files in different configs.
> If the default directory were dataDir rather than CWD, each core would get 
> its own infostream file.  You could still get the original behavior by 
> specifying an absolute path.

--
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: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to