Any difference in video driver? I've heard of some video drivers causing the JVM to crash, but I think that's usually ATI drivers on Windows.
Both systems are using NVIDIA GeForce 4 MX video boards. One system is an Athlon CPU, the other (the one that doesn't work right) is a dual Opteron system installed with the i386 distribution, not the x86_64. Both are up to date on patches, and both are running 2.6.8-1.521 kernels (the dual Opteron is running the -smp kernel).
Is there any way you can get appletviewer to run the applet, rather than using Mozilla? That would at least separate Mozilla and the JVM.
It's something in java:
draco> /usr/local/j2re1.4.2_05/bin/java -version Segmentation fault
Here are the last few lines of an strace of the above command:
mkdir("/tmp/hsperfdata_iglesias", 0755) = -1 EEXIST (File exists) lstat64("/tmp/hsperfdata_iglesias", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 open("/tmp/hsperfdata_iglesias/1067", O_RDWR|O_CREAT|O_TRUNC, 0600) = 3 ftruncate(3, 16384) = 0 mmap2(NULL, 16384, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0xf6ffc000 close(3) = 0 mmap2(0xfee0e000, 12288, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0xfee0e000 mprotect(0xfee0e000, 12288, PROT_NONE) = 0 mmap2(0xfee04000, 40960, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0xfee04000 sigaltstack({ss_sp=0xfee04000, ss_flags=0, ss_size=40960}, NULL) = 0 rt_sigaction(SIGHUP, NULL, {SIG_DFL}, 8) = 0 rt_sigaction(SIGINT, NULL, {SIG_DFL}, 8) = 0 rt_sigaction(SIGTERM, NULL, {SIG_DFL}, 8) = 0 rt_sigprocmask(SIG_UNBLOCK, [HUP INT QUIT USR2 TERM CHLD], NULL, 8) = 0 --- SIGSEGV (Segmentation fault) @ 0 (0) --- --- SIGSEGV (Segmentation fault) @ 0 (0) --- +++ killed by SIGSEGV +++
and here's a stack trace from gdb on a core file:
#0 0x00664514 in JVM_handle_linux_signal () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #1 0x00663728 in signalHandler () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #2 <signal handler called> #3 0x0080eb44 in code_template.15742 () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #4 0x00665257 in os::Linux::init_thread_fpu_state () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #5 0x00661655 in os::create_attached_thread () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #6 0x0066422c in os::create_main_thread () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #7 0x006ab7c5 in Threads::create_vm () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #8 0x005d21d8 in JNI_CreateJavaVM () from /usr/local/j2re1.4.2_05/lib/i386/client/libjvm.so #9 0x0804aea3 in ?? () #10 0xfefff814 in ?? () #11 0xfefff810 in ?? () #12 0xfeffd780 in ?? () #13 0x0805813c in ?? () #14 0x00000000 in ?? ()
That doesn't look good. It certainly looks like a JVM problem with your Opteron system.
There must be some version of the JVM which runs on Opteron - Sun's new Java Station is Opteron after all, and they offer dual cpu with RedHat x86 as an option.
I'm not really sure what to suggest. You might try asking in the comp.lang.java.help newsgroup if anyone there has had a similar problem and knows the cause. I've not seen any posts there specific to Opteron, and if there was a general problem with the JVM and Opteron I'm sure there would have been some posts.
-- Nigel Wade, System Administrator, Space Plasma Physics Group, University of Leicester, Leicester, LE1 7RH, UK E-mail : nmw@xxxxxxxxxxxx Phone : +44 (0)116 2523548, Fax : +44 (0)116 2523555
