NOTICE! This is a static HTML version of a legacy Fiji BugZilla bug.

The Fiji project now uses GitHub Issues for issue tracking.

Please file all new issues there.

Bug 173 - JVM crash with command finder
JVM crash with command finder
Status: RESOLVED WORKSFORME
Product: Fiji
Classification: Unclassified
Component: Plugins
unspecified
PC Linux
: P4 normal
Assigned To: ImageJ Bugs Mailing List
Depends on:
Blocks:
 
Reported: 2010-05-27 04:50 CDT by Albert Cardona
Modified: 2011-03-23 15:55 CDT
1 user (show)

See Also:

Description Albert Cardona 2010-05-27 04:50:12 CDT
Run fiji like:

$ ./fiji -agentlib:jdwp=transport=dt_socket,address=8010,server=y,suspend=n -Xincgc


Then type 'l', type "expression", push return to execute the Image Expression Launcher.

What happens is that the JVM crashes. It doesn't happen always, but sometimes.

I saw this error in the past and I thought I had worked around it by ensuring that some commands in the ij.plugin.CommandFinder class are launched within the proper event dispatch thread. Those changes seemed to work, but obviously didn't, and the problem now resurfaces.


Listening for transport dt_socket at address: 8010
The program '<unknown>' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadWindow (invalid Window parameter)'.
  (Details: serial 959 error_code 3 request_code 20 minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the --sync command line
   option to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)

Information about your version of Java - this information is useful for the Fiji developers:
  os.arch => amd64
  os.name => Linux
  os.version => 2.6.31-19-generic
  java.version => 1.6.0_17
  java.vendor => Sun Microsystems Inc.
  java.runtime.name => Java(TM) SE Runtime Environment
  java.runtime.version => 1.6.0_17-b04
  java.vm.name => Java HotSpot(TM) 64-Bit Server VM
  java.vm.version => 14.3-b01
  java.vm.vendor => Sun Microsystems Inc.
  java.vm.info => mixed mode
  java.awt.graphicsenv => sun.awt.X11GraphicsEnvironment
  java.specification.name => Java Platform API Specification
  java.specification.version => 1.6
  sun.cpu.endian => little
  sun.desktop => gnome
  file.separator => /
Comment 1 Johannes Schindelin 2011-03-23 15:17:50 CDT
Whoa. Does this still happen with 1.6u24?
Comment 2 Albert Cardona 2011-03-23 15:55:17 CDT
(In reply to comment #1)
> Whoa. Does this still happen with 1.6u24?

So far I have not run into the issue. But Mark mentioned that now Swing is even more evil than before: that all operations on all Swing classes have to run under the event dispatch thread. That is, no building a GUI on a regular thread and then calling .pack() and setVisible(true) in an invokeLater. Despite this, Fiji seems to hold well--I haven't heard any complains, and here the whole institute uses Fiji on a daily basis.