JDK-4296009 : jdk1.3.P JavaMonitor not loading on win2000RC2
  • Type: Bug
  • Component: client-libs
  • Sub-Component: javax.sound
  • Affected Version: 1.3.0
  • Priority: P1
  • Status: Closed
  • Resolution: Duplicate
  • OS: windows_nt
  • CPU: x86
  • Submitted: 1999-12-01
  • Updated: 2000-05-16
  • Resolved: 2000-05-16
Related Reports
Duplicate :  
Description
Accessibility Tool-JavaMonitor not able to load while testing Java2D demo, get a HotSpot Virtual Machine Error, Exception_Access_Violation

Java2demo quits.

steps to reproduce:
on win2000rc2, install the latest accessibility tool-JavaMonitor
launch java2demo

result:
demo quits.

cpu tested on: jfc-swinging
*********************************************

zelony@eng 1999-12-17
tried with build R, not able to reproduce.

Comments
EVALUATION Tried with build-p classic and it fails, according to Moe. peter.kessler@Eng 1999-12-03 I ran the tests suggested by Peter Kestler (see comments). All the tests failed in the same way. This is neither a hotspot nor an accessibility bug. The Java2demo fails both with and without accessibility and both with and without hotspot. lynn.monsanto@Eng 1999-12-09 ================ As per this email snippert it doesn't seem to be necessary to run JavaMonitor to show the problem which is probably just as well ... > Roger, > > I reassigned bug 4296009 to the "demo" subcategory. Peter Kessler, Moe > and I looked at this bug and determined that the Java2Demo always fails with > the same exception whether or not JavaMonitor loading is requested. The > demo also always fails whether or not it is run under hotspot or classic java. > The problem appears to be in the Java2Demo on Win2000 RC2. > > Lynn I have win2000 RC1 (not RC2) on my PC. When I start up SwingSet2 from JDK 1.3 "P" it starts fine, When I start up Java2Demo it gets a HotSpot exception, but it looks to be a JavaSound problem. Since the original submitter didn't provide any kind of stack trace I have to assume its the same although as well as I see .. marking bug incomplete as submitter ought to add that information. G:\jdk13\demo\jfc\Java2D>\jdk13\bin\java -jar Java2Demo.jar javax.sound.midi.InvalidMidiDataException: Failed to load sequence at com.sun.media.sound.MixerSequencer.setSequence(MixerSequencer.java:28 7) at com.sun.media.sound.JavaSoundAudioClip.openAsSequence(JavaSoundAudioC lip.java:452) at com.sun.media.sound.JavaSoundAudioClip.<init>(JavaSoundAudioClip.java :128) at java.lang.reflect.Constructor.newInstance(Native Method) at sun.applet.AppletAudioClip.createAppletAudioClip(AppletAudioClip.java :183) at sun.applet.AppletAudioClip.<init>(AppletAudioClip.java:67) at java.applet.Applet.newAudioClip(Applet.java:237) at demos.Mix.JavaMedia$JavaSound.<init>(JavaMedia.java:306) at demos.Mix.JavaMedia.<init>(JavaMedia.java:124) at java.lang.Class.newInstance0(Native Method) at java.lang.Class.newInstance(Class.java:241) at DemoPanel.<init>(DemoPanel.java:56) at DemoGroup.<init>(DemoGroup.java:85) at Java2Demo.<init>(Java2Demo.java:125) at Java2Demo.main(Java2Demo.java:466) ... ... ... # # HotSpot Virtual Machine Error, EXCEPTION_ACCESS_VIOLATION # # Error ID: 4F533F57494E13120E43505002BE # abnormal program termination (I've added the full exception as an attachment). BTW I have a SoundBlaster Live in my PC and who knows if the drivers are fully certified yet .. although I have has very similar problems on the past in NT 4.0 with JavaSound. phil.race@eng 1999-12-09 ======================== I got quite different results than all the above. On Windows 2000, RC2 (the engineer above needs to upgrade) I get: 1) Java2Demo by itself runs fine with P. 2) Java2Demo with Accessibility and classic run fine with P. 3) Java2Demo with Accessibility and hotspot crashes. I was unable to coax a Dr. Watson log out of this machine. There is no more information to give. Perhaps the engineers would like to see this machine? nancy.schorr@eng 1999-12-13 The bug is very much reproducible with classic. It just takes a few tries to get the classic to crash. The crash happens at 77% when loading demo.mix. The stack trace may or may not be available. And, regardless on your windows version <NT 4.0 SP5 and 2000rc2/rc1, did not try 98> it crashes. This is not 2000 problem. The crash is random, and not reproducible with java_g. However, build-r seems fix the problem. try /net/mizu/re/jdk1.3/binary/win32/latest/bin/java and the bug does not happen. Note, that this was build-r of jdk and build-p of hotspot. I will assign the bug to the javasound guy. I do think the bug is fixed. ###@###.### 1999-12-14 ----------- The crash of Java2Demo seen here is the result of bug 4295662. We had code that allocated memory from the global heap in Windows, and did not free the memory when it should have been freed. The implications of this bug are wide -- anything needing to access global memory could be affected. Depending on the configuration of the machine, this bug may be easy or difficult to reproduce. After talking with Moe and Robert Kim about this bug, I am marking it fixed and keeping it open until it is verified when R gets promoted. I do, however, believe that this is a duplicate of 4295662. ###@###.### 1999-12-14
14-12-1999