]
Galder Zamarreño updated ISPN-2723:
-----------------------------------
Status: Pull Request Sent (was: Coding In Progress)
Git Pull Request:
NPE using cache loader preload with Lucene directory
----------------------------------------------------
Key: ISPN-2723
URL:
https://issues.jboss.org/browse/ISPN-2723
Project: Infinispan
Issue Type: Bug
Components: Distributed Cache, Loaders and Stores
Affects Versions: 5.2.0.CR1
Reporter: Christopher Wong
Assignee: Galder Zamarreño
Fix For: 5.2.0.Final
Attachments: infinispan.cfg.xml, infinispan.log
I am seeing an NPE that looks a lot like ISPN-1470, except this is happening in version
5.2.0.CR1 of Infinispan. I have configured Infinispan's Lucene directory provider for
use in Hibernate Search. The Hibernate SessionFactory is configured with a JTA transaction
manager. Starting with no index works fine, but if I shut down Tomcat (with shutdown.sh)
and restart, a huge pile of exceptions occur, starting with an NPE. The cache
configuration in infinispan.cfg.xml looks like the following. I will attach a log file
excerpt with a sampling of the exceptions being logged. This only happens with distributed
mode. Replicated mode is fine. I have seen this happen with both the Jdbm and file cache
store.
<namedCache
name="LuceneIndexesData">
<clustering
mode="dist">
<stateTransfer fetchInMemoryState="true"/>
<sync
replTimeout="50000" />
<l1 enabled="false" />
</clustering>
<loaders shared="true" preload="true">
<loader class="org.infinispan.loaders.file.FileCacheStore"
fetchPersistentState="false" ignoreModifications="false"
purgeOnStartup="false">
<properties>
<property name="location"
value="/some/path/.index/data" />
</properties>
</loader>
</loaders>
</namedCache>
--
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: