JDK-6473338 : ATG application occasionally hangs during safepointing (due to monitor inflation livelock?)
  • Type: Bug
  • Component: hotspot
  • Sub-Component: runtime
  • Affected Version: 6
  • Priority: P3
  • Status: Closed
  • Resolution: Fixed
  • OS: linux,solaris_9,solaris_10
  • CPU: x86,sparc
  • Submitted: 2006-09-21
  • Updated: 2010-12-03
  • Resolved: 2007-07-11
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 6 JDK 7 Other
6u4 b01Fixed 7Fixed hs10Fixed
Related Reports
Duplicate :  
Duplicate :  
Relates :  
While I was running ATG app, I saw it hang after 1 days and 13 hours or so. I also found that the server process was hogging the system at the time of the hang. It is happening to Solaris 9 only, and I couldn't reproduce it using 2 other machines running Solaris 10 and 11. Since I couldn't keep the live process, I have restarted the test and hope I could reproduce it.

After Ramki looking at the pstack file from the gcore, it showed the vm thread was trying to safepoint the JVM while several threads appeared to be stuck looping in the monitor inflation code. We have given the heads-up to Dave, and he will look into the case.

Please see the pstack file attached.

Test related information:
hostname: jtg-blade2.sfbay.sun.com (please contact the submitter for the login)
Sun Blade 1000, 2 ways @ 900 MHZ with 2 G of RAM.

bash-2.05# cat /etc/*release*
                       Solaris 9 9/04 s9s_u7wos_09 SPARC
          Copyright 2004 Sun Microsystems, Inc.  All Rights Reserved.
                       Use is subject to license terms.
                            Assembled 29 June 2004


Flags="-server  -XX:+UseParallelOldGC"

Log and core file location: 

Please look at core.24669, which is the troubled server process that eventually hog the whole system.

 core.24669      <- The server process core file
 core.9232       <- The client process core file

EVALUATION 6518490 describes the more general problem.

EVALUATION see comments