You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@lucene.apache.org by "Joel Bernstein (JIRA)" <ji...@apache.org> on 2015/04/06 15:50:12 UTC

[jira] [Created] (SOLR-7352) Synchronize CloudSolrStream's EOF Tuple Map

Joel Bernstein created SOLR-7352:
------------------------------------

             Summary: Synchronize CloudSolrStream's EOF Tuple Map
                 Key: SOLR-7352
                 URL: https://issues.apache.org/jira/browse/SOLR-7352
             Project: Solr
          Issue Type: Bug
            Reporter: Joel Bernstein


CloudSolrStream currently collects the EOF Tuples from the underlying SolrStreams in an un-synchronized Map. 

This ticket synchronizes this Map because of the following race condition:

The underlying SolrStream's are opened in parallel. If the first Tuple returned by two or more SolrStream's is an EOF Tuple, then a race condition will occur as the EOF Tuples are added to the Map.

The race condition only occurs if EOF tuples are returned during the *opening* of the SolrStreams (zero results returned). This is because the *reading* of SolrStreams is performed in a single thread, so if the initial Tuple returned is not an EOF Tuple no race condition is possible.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

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