JDK-6324874 : vtest crashed with ParallelOldGC after running 1 day and 6 hours
  • Type: Bug
  • Component: hotspot
  • Sub-Component: gc
  • Affected Version: 6
  • Priority: P4
  • Status: Closed
  • Resolution: Duplicate
  • OS: linux_suse_sles_9
  • CPU: x86
  • Submitted: 2005-09-16
  • Updated: 2010-04-02
  • Resolved: 2005-12-20
Related Reports
Relates :  
Description
vtest crashed with ParallelOldGC after running 1 day and 6 hours.

Test machine: jtg-amd3.sfbay
# cat /etc/*release*
SuSE Linux 9.0 - 64
VERSION = 9.0

# /usr/j2se/bin/java -client -version
java version "1.6.0-ea"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.6.0-ea-b52)
Java HotSpot(TM) 64-Bit Server VM (build 20050908154901.jcoomes.gc_merge-debug, mixed mode)

JVM flags used: -client  -XX:CompileThreshold=100 -XX:+UseParallelGC -XX:+UseParallelOldGC

jtg-amd3:/bt/VolanoTestrun.16553.-client # pwd
/bt/VolanoTestrun.16553.-client
jtg-amd3:/bt/VolanoTestrun.16553.-client # tail run*out
Unexpected Error
------------------------------------------------------------------------------
Internal Error at typeArrayOop.hpp:28, pid=19323, tid=1080375648

Do you want to debug the problem?

To debug, run 'gdb /proc/19323/exe 19323'; then switch to thread 1080375648

Stack Trace:
-------------
(gdb) thread 373
[Switching to thread 373 (Thread 1080375648 (LWP 19328))]#0
0x0000002a95911e75 in __nanosleep_nocancel () from /lib64/tls/libc.so.6
(gdb) where
#0  0x0000002a95911e75 in __nanosleep_nocancel () from /lib64/tls/libc.so.6
#1  0x0000002a95911ca7 in sleep () from /lib64/tls/libc.so.6
#2  0x0000002a9619b08f in os::message_box ()
from /usr/j2se/jre/lib/amd64/server/libjvm.so
#3  0x0000002a96376ae6 in VMError::show_message_box ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#4  0x0000002a96376568 in VMError::report_and_die ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#5  0x0000002a95dc993e in report_assertion_failure ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#6  0x0000002a95dbb71a in constantPoolKlass::oop_update_pointers ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#7  0x0000002a96210a30 in oopDesc::update_contents ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#8  0x0000002a9620b829 in PSParallelCompact::update_not_updated ()
from /usr/j2se/jre/lib/amd64/server/libjvm.so
#9  0x0000002a96206cee in PSParallelCompact::invoke_no_policy ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#10 0x0000002a962135bf in PSScavenge::invoke ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#11 0x0000002a961b094d in ParallelScavengeHeap::failed_mem_allocate ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#12 0x0000002a96377bf2 in VM_ParallelGCFailedAllocation::doit ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#13 0x0000002a9639e057 in VM_Operation::evaluate ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#14 0x0000002a9639d116 in VMThread::evaluate_operation ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#15 0x0000002a9639d543 in VMThread::loop ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#16 0x0000002a9639cd70 in VMThread::run ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#17 0x0000002a9619bf94 in java_start ()
 from /usr/j2se/jre/lib/amd64/server/libjvm.so
#18 0x0000002a956738e2 in start_thread () from /lib64/tls/libpthread.so.0

Comments
EVALUATION The above note should refer to 6334748.
20-12-2005

EVALUATION There are aspects of 6334738 that could cause chunks above the top of a space to be processed in some way (possible copied to the wrong place or processed when they should not be). This situations occur when the top of the space coincide with the end of the last chunk in the space. These would be very intermittent and could cause the vm to crash. I'm closing this as a duplicate of 6334738. If a failure is observed again after the fix for 6334738 is promoted, a new bug should be filed.
20-12-2005

EVALUATION The failure is occurring during the part of the collection when the objects which cross a chunk boundary and which have not been updated are updated.
25-10-2005