lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yonik Seeley (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (SOLR-3815) add hash range to shard
Date Thu, 13 Sep 2012 03:49:07 GMT

    [ https://issues.apache.org/jira/browse/SOLR-3815?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13454608#comment-13454608
] 

Yonik Seeley edited comment on SOLR-3815 at 9/13/12 2:49 PM:
-------------------------------------------------------------

Folks, while working to add the "replicas" level to shards (to make room for other properties),
I noticed that the Overseer.updateSlice() method changed the existing ClusterState (which
is advertised as being immutable).  I re-wrote the method to be much shorter, and immutable
with respect to the existing ClusterState, and started getting a test failure.

I eventually tried just adding back the part of the code that erroneously modified the existing
ClusterState, and the test passed again (see the nocommit block in Overseer).

Any idea what's going on?

edit: the test that failed was LeaderElectionIntegrationTest.  Not sure if it caused other
failures.

edit: in Overseer.run() we have "ClusterState clusterState = reader.getClusterState();" and
that is the state that is accidentally being modified (that accidentally makes things work).
 I assume this is OK, as the reader is supposed to update it's state via zookeeper - which
means there is perhaps something wrong with reader.updateClusterState(true)?
                
      was (Author: yseeley@gmail.com):
    Folks, while working to add the "replicas" level to shards (to make room for other properties),
I noticed that the Overseer.updateSlice() method changed the existing ClusterState (which
is advertised as being immutable).  I re-wrote the method to be much shorter, and immutable
with respect to the existing ClusterState, and started getting a test failure.

I eventually tried just adding back the part of the code that erroneously modified the existing
ClusterState, and the test passed again (see the nocommit block in Overseer).

Any idea what's going on?

edit: the test that failed was LeaderElectionIntegrationTest.  Not sure if it caused other
failures.
                  
> add hash range to shard
> -----------------------
>
>                 Key: SOLR-3815
>                 URL: https://issues.apache.org/jira/browse/SOLR-3815
>             Project: Solr
>          Issue Type: Sub-task
>            Reporter: Yonik Seeley
>         Attachments: SOLR-3815_addrange.patch, SOLR-3815_clusterState_immutable.patch,
SOLR-3815.patch
>
>


--
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