I've got some Cassandra 1.2.5 instances installed on Solaris machines, and have recently tried updating from Java 1.6 to 1.7 to try and fix some performance issue's I've been running into, but after updating it is now crashing on start up with the following:
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0xffffffff7d9b236c, pid=12586, tid=7
#
# JRE version: Java(TM) SE Runtime Environment (7.0_71-b14) (build 1.7.0_71-b14)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (24.71-b01 mixed mode solaris-sparc compressed oops)
# Problematic frame:
# V [libjvm.so+0x5b236c] unsigned long CompactibleFreeListSpace::block_size(const HeapWord*)const+0x15c
#
# Core dump written. Default location: /opt/resolve/cassandra/bin/core or core.12586
#
# If you would like to submit a bug report, please visit:
# http://bugreport.sun.com/bugreport/crash.jsp
#
...
Stack: [0xffffffff76d00000,0xffffffff76e00000], sp=0xffffffff76dff240, free space=1020k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x5b236c] unsigned long CompactibleFreeListSpace::block_size(const HeapWord*)const+0x15c
V [libjvm.so+0x42a988] HeapWord*BlockOffsetArrayNonContigSpace::block_start_unsafe(const void*)const+0xf0
V [libjvm.so+0xb21964] void CardTableModRefBS::process_chunk_boundaries(Space*,DirtyCardToOopClosure*,MemRegion,MemRegion,signed char**,unsigned long,unsigned long)+0x1f4
V [libjvm.so+0xb213f4] void CardTableModRefBS::non_clean_card_iterate_parallel_work(Space*,MemRegion,OopsInGenClosure*,CardTableRS*,int)+0x3dc
V [libjvm.so+0x518564] void CardTableModRefBS::non_clean_card_iterate_possibly_parallel(Space*,MemRegion,OopsInGenClosure*,CardTableRS*)+0x64
V [libjvm.so+0x519fe0] void CardTableRS::younger_refs_in_space_iterate(Space*,OopsInGenClosure*)+0x40
V [libjvm.so+0x5f739c] void ConcurrentMarkSweepGeneration::younger_refs_iterate(OopsInGenClosure*)+0x54
V [libjvm.so+0x706f54] void GenCollectedHeap::gen_process_strong_roots(int,bool,bool,bool,SharedHeap::ScanningOption,OopsInGenClosure*,bool,OopsInGenClosure*)+0x1bc
V [libjvm.so+0xb28bd0] void ParNewGenTask::work(unsigned)+0x150
V [libjvm.so+0xcd8a0c] void GangWorker::loop()+0xa0
V [libjvm.so+0xb0786c] java_start+0x364
I found a JDK bug, http://bugs.java.com/bugdatabase/view_bug.do?bug_id=6913441, that seems to match but it's marked as a duplicate of bug id 6896647 which unless I'm reading that bug report incorrectly should have been fixed in the latest Java 1.7 build which I'm using.
I also found a similar crash report in this Open JDK bug report https://bugs.openjdk.java.net/browse/JDK-8021006, but it only mentions that it was due to memory corruption issues occurring on SQE's stt-34, stt-68, and stt-128.
Has anyone else encountered this crash, and if so are either of these answers correct or am I looking in the wrong place?