JDK-8022892 : Java VM crashes with access violation when started with -XX:+UseConcMarkSweepGC
  • Type: Bug
  • Component: hotspot
  • Sub-Component: gc
  • Affected Version: hs23,7u25
  • Priority: P3
  • Status: Resolved
  • Resolution: Duplicate
  • OS: windows_2008
  • Submitted: 2013-08-13
  • Updated: 2013-08-15
  • Resolved: 2013-08-15
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.
Other
hs23Resolved
Related Reports
Duplicate :  
Description
FULL PRODUCT VERSION :
java version  " 1.7.0_25 " 
Java(TM) SE Runtime Environment (build 1.7.0_25-b17)
Java HotSpot(TM) 64-Bit Server VM (build 23.25-b01, mixed mode)

FULL OS VERSION :
Tested on multiple platforms (redhat linux and windows)
e.g. microsoft windows server 2008 r2 enterprise
(6.1.7601 SP 1)

A DESCRIPTION OF THE PROBLEM :
When starting the Java-VM with more than 128 Gigabytes of Heapsize:
java.exe -server  -Xms140g -Xmx140g -XX:NewSize=1024m -XX:MaxNewSize=1024m -XX:+UseConcMarkSweepGC

the JAVA VM crashes with access violation dialogue (windows)
Same under linux (segmentation fault)

There are similar problems already described in the bug database:
JDK-8011190
JDK-7197906
According to the bug database these are already fixed (but it seems they are NOT)

Tested also with latest version of Java6(43) -> same problem


THE PROBLEM WAS REPRODUCIBLE WITH -Xint FLAG: Did not try

THE PROBLEM WAS REPRODUCIBLE WITH -server FLAG: Yes

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
Just start a VM using:
-server  -Xms140g -Xmx140g -XX:NewSize=1024m -XX:MaxNewSize=1024m -XX:+UseConcMarkSweepGC
as parameter

EXPECTED VERSUS ACTUAL BEHAVIOR :
Expected: Java VM starts and works properly
Actual: Java VM crashes
ERROR MESSAGES/STACK TRACES THAT OCCUR :
Error message when started on windows (contained in the error dialogue of windows):
Problem signature:
  Problem Event Name:APPCRASH
  Application Name:java.exe
  Application Version:7.0.250.17
  Application Timestamp:51c4b1fc
  Fault Module Name:msvcr100.dll
  Fault Module Version:10.0.30319.460
  Fault Module Timestamp:4db1316a
  Exception Code:c0000005
  Exception Offset:000000000003c350
  OS Version:6.1.7601.2.1.0.274.10
  Locale ID:1031
  Additional Information 1:d88d
  Additional Information 2:d88db958fddab0a6ecce64dc175d2f43
  Additional Information 3:b635
  Additional Information 4:b635a7f4a8b9426b33fc792420cfd294

Read our privacy statement online:
  http://go.microsoft.com/fwlink/?linkid=104288&clcid=0x0409

If the online privacy statement is not available, please read our privacy statement offline:
  C:\Windows\system32\en-US\erofflps.txt



REPRODUCIBILITY :
This bug can be reproduced always.

---------- BEGIN SOURCE ----------
No need for source code
---------- END SOURCE ----------

CUSTOMER SUBMITTED WORKAROUND :
It seems to work if we do not use the CMS-GC.

But according to our performance tests with our application the  " old GCs "  do lead to blocking time (and communication timeouts)
Only current alternative is G1GC which does not lead to timeouts but decreases the performance in comparison to CMS-GC

Also tried to start the VM with lower XMS value. So the VM started, but after doing some tests the access violations occured after FullGC of the system
Comments
This will not be backported as it will be part of the soon to be released 7u40. Closing as duplicate of bug JDK-7197906
15-08-2013

Sustaining is evaluating if this should be backported to hs23 or not.
15-08-2013

How much physical memory and swap space (try `free` or `swapon -s`) did the customer use? Can you provide the hs_err_*.log file when you reproduce it on (better) Linux? Thank you.
13-08-2013