JDK-5030149 : RAS - Tiger reliability testsuite(SwingBasher) throws OutOfMemoryError
  • Type: Bug
  • Component: client-libs
  • Sub-Component: javax.swing
  • Affected Version: 5.0
  • Priority: P2
  • Status: Closed
  • Resolution: Cannot Reproduce
  • OS: solaris_9
  • CPU: sparc
  • Submitted: 2004-04-12
  • Updated: 2004-05-12
  • Resolved: 2004-05-12
Related Reports
Relates :  
Description
SwingBasher is modified to accomodate new tiger features for Swing. Now when the modified swing basher is executed it throws OutOfMemoryError on Solaris9 and Solaris10 after 28th loop when run continuously for Reliability testing.

Location for Swingbasher: 
--------------------------
Find attached with this bug.

Steps to reproduce:
--------------------
1. Set JAVA_HOME to b43/b44
2. Set SB_TIMELIMIT to 4320 (72 Hours i.e.)
3. sh runnit.sh


//Error Log
Total Memory :66912256 , Free Memory :15970384
Started Loop #25 at Fri Apr 02 01:15:22 IST 2004 with Metal
Total Memory :66912256 , Free Memory :6101832

Total Memory :66912256 , Free Memory :5626392
Started Loop #26 at Fri Apr 02 01:56:35 IST 2004 with CDE/Motif
Total Memory :66912256 , Free Memory :7272456

Total Memory :66912256 , Free Memory :6737584
Started Loop #27 at Fri Apr 02 02:42:07 IST 2004 with GTK look and feel
Total Memory :66912256 , Free Memory :8742968

Total Memory :66912256 , Free Memory :7800280
Started Loop #28 at Fri Apr 02 03:29:57 IST 2004 with Synth look and feel

Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
gnumake: *** [runloop] Error 1
//

Find complete results attached with this bug.


Previously SwingBasher tests developed for Merlin features executed against tiger beta1 build and it threw OutOfMemoryError (refer Bug id 4984042). Now the bug is fixed and the merlin reliability testsuite without any tiger features is working fine from build b43. But now the new tiger reliability testsuite is throwing OutOfMemoryError in solaris9 and solaris10. 

###@###.### 2004-04-12

Comments
EVALUATION Commit to Tiger ###@###.### 2004-04-14 I was able to reproduce this with b46 on a SunBlade 1000 running Solaris 9, though it did make it through 160+ iterations. I'm going to rerun with the Swing nightly build from 4/17, which has the 4836639 fix in it. I expect that it will pass. ###@###.### 2004-04-19 Confirmed with the submitter that this bug is no longer reproducible with b49. ###@###.### 2004-05-12
19-04-2004