[
https://issues.apache.org/jira/browse/RYA-393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16195154#comment-16195154
]
ASF GitHub Bot commented on RYA-393:
------------------------------------
Github user amihalik commented on the issue:
https://github.com/apache/incubator-rya/pull/236
@ejwhite922 There's a couple issues with the batch writer and it doesn't
have to do with exceptions being caught. (1) the QueueFullChecker thread
should not be a daemon thread.... "The Java Virtual Machine exits when the only
threads running are all daemon threads" means that the JVM will exit while
flushing to mongo. (2) It doesn't seem like the secondary indexer (i.e.
geoindexer) is flushed when I close down the sail layer. Can you double check
that?
Also, it seems like the JVM takes a very long time to exit after I call
sailrepository.shutDown(). I think has to do with the BatchWriter as well.
> MongoDB DAO Batch Writer drops statements when batch contains dup key
> ---------------------------------------------------------------------
>
> Key: RYA-393
> URL: https://issues.apache.org/jira/browse/RYA-393
> Project: Rya
> Issue Type: Bug
> Reporter: Aaron Mihalik
> Assignee: Eric White
>
> The MongoDB DAO Batch Writer will write a group of statements at a single
> time. It seems like it will drop valid statements if the batch contains a
> statement with a duplicate key
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
