Community
Participate
Working Groups
Build Identifier: 1.0.1 I tried to run ParseHeapDump.sh against a heapdump and get a java.lang.NoClassDefFoundError on com.ibm.dtfj.image.CorruptDataException Full console output is here: http://pastebin.com/aMYntUga Version info from MAT about box: Version: 1.0.1.201008091353 Build id: 1.0.1 Reproducible: Always Steps to Reproduce: 1. run with openjdk or sunjdk on Linux x64 2. ./ParseHeapDump.sh somePath/someDump.hprof 3. Boom
This message also appears with the Windows version when running ParseHeapDump without DTFJ installed. I think it's harmless. The problem is that the org.eclipse.mat.dtfj doesn't require the IBM DTFJ plugins as the idea is that it can use them later once they are installed. However, when this plugin is stopped it calls some code which presumes the availability of the IBM DTFJ com.ibm.java.dtfj
I have modified InitDTFJ so that does not call com.ibm.dtfj.image.CorruptDataException unless there is an com.ibm.dtfj.api.ImageFactory extension point.