[infinispan-issues] [JBoss JIRA] (ISPN-3311) Unmarshalling error switching from 5.2 to 5.3

Sanne Grinovero (JIRA) jira-events at lists.jboss.org
Fri Jul 19 07:38:26 EDT 2013


    [ https://issues.jboss.org/browse/ISPN-3311?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12790968#comment-12790968 ] 

Sanne Grinovero commented on ISPN-3311:
---------------------------------------

Hi Galder,
the {{lucene-v3}} vs. {{lucene}} split in Maven is only a source-level split. The assembled jar contains compiled classes from both v3 and v4 modules: we had to use a Maven hack to support both versions of Lucene and still be backwards compatible in terms of module definition. Nothing too magic, we just merge the two built modules and then generate the OSGi stuff as last step.

If this just affects the entries stored via the Lucene Directory then it's definitely less critical: indexes can be rebuilt from the data; if it's a {{LifecycleCallbacks}} issue I guess that could even be compensated for by registering them explicitly?
                
> Unmarshalling error switching from 5.2 to 5.3
> ---------------------------------------------
>
>                 Key: ISPN-3311
>                 URL: https://issues.jboss.org/browse/ISPN-3311
>             Project: Infinispan
>          Issue Type: Bug
>    Affects Versions: 5.3.0.Final
>            Reporter: Galder Zamarreño
>            Assignee: Sanne Grinovero
>             Fix For: 6.0.0.Alpha2, 6.0.0.Final
>
>
> Error:
> {code}Caused by: org.infinispan.CacheException: Type of data read is
> unknown. Id=-2147482348 is not amongst known reader indexes.
> at org.infinispan.marshall.jboss.ExternalizerTable.readObject(ExternalizerTable.java:222){code}
> Full stacktrace: https://gist.github.com/galderz/770679223e02f5b211f8

--
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



More information about the infinispan-issues mailing list