SIGSEGV Crash on Solaris in JVM

Question

Somewhere in the middle of a program I have, I get a Solaris (5.10) coredump crash in a JVM (1.5) call to strlen. When captured with dbx, it displays a stack thusly for the Motif toolkit:
=>[1] strlen(0x1a, 0x10000000, 0x1a, 0xfffffffffffffffa, 0x2, 0x0), at 0xff2b0c54
[2] MergeStatus(0x16b1e88, 0xfeb0f940, 0x14f3520, 0xc, 0xfefbe000, 0x0), at 0xdb2393d8
[3] UpdateStatus(0x16b1e88, 0x0, 0x14f3520, 0xc, 0x0, 0x1b3b970), at 0xdb238968
[4] RepaintStatus(0x14f3520, 0x14f3520, 0xd903f7f0, 0x16b1e88, 0x18002fa, 0x16b1e88), at 0xdb2387c0
[5] 0xd9c1825c(0x1, 0x6ae7c, 0x6af14, 0xd9c18230, 0x0, 0xd9c86a90), at 0xd9c1825b
[6] 0xd9c17a34(0x5950cc, 0x549718, 0x0, 0x8, 0xd9c82b14, 0xd9c87948), at 0xd9c17a33
[7] Java_sun_awt_motif_MToolkit_run(0x5950cc, 0xfea06ba0, 0x6b2d8, 0xfea869a8, 0xd9c82b14, 0xff00a630), at 0xd9c17874
[8] 0xf900c280(0x595010, 0xd903fa6c, 0xd903f9f8, 0xffffff80, 0x80000000, 0x1404), at 0xf900c27f
[9] 0xf900c224(0xf04599d8, 0xdb82c000, 0x0, 0x4, 0xfefbe000, 0xd903fa10), at 0xf900c223
[10] 0xf9005c2c(0x0, 0x35d4c, 0x0, 0xf9016210, 0x7bac, 0xd903fa88), at 0xf9005c2b
[11] 0xf9000218(0xd903fb70, 0xd903fd70, 0xa, 0xdb82d230, 0xf900a840, 0xd903fc88), at 0xf9000217
[12] JavaCalls::call_helper(0x1, 0x595010, 0xd903fc80, 0xd903fb80, 0x4, 0xd903fd70), at 0xfe99a808
[13] JavaCalls::call_virtual(0x7a9c, 0x595010, 0x4d90a8, 0x4d90b4, 0x4d90b0, 0xf046c8b0), at 0xfeac319c
[14] thread_entry(0xdb82f280, 0x595010, 0x4d9494, 0xff024640, 0xff024710, 0xff024130), at 0xfeae2700
[15] JavaThread::run(0x595010, 0x0, 0xff00c790, 0x6c00, 0x7dd4, 0x7c00), at 0xfeade29c
[16] _start(0x595010, 0x22cc, 0xfefbe000, 0x0, 0x4f24, 0x4c00), at 0xfee64240
Or, like so for the X11 toolkit:
=>[1] strlen(0x1d, 0x7, 0x1d, 0x4001c5, 0x1, 0x0), at 0xff2b0c58
[2] MergeStatus(0x18edc08, 0xb6, 0x3452c8, 0xf90147c8, 0xc68, 0x0), at 0xd64b93d8
[3] UpdateStatus(0x18edc08, 0x0, 0x3452c8, 0xf9015cc0, 0x18ae, 0x18f12a0), at 0xd64b8968
[4] RepaintStatus(0x3452c8, 0x3452c8, 0x57cad0, 0x18edc08, 0x4001c5, 0x18edc08), at 0xd64b87c0
[5] Java_sun_awt_X11_XlibWrapper_XFilterEvent(0x59494c, 0xd95bf8d8, 0x0, 0x57cad0, 0x0, 0x0), at 0xf8c3b4a8
[6] 0xf900c280(0x30a9, 0xd95bf95c, 0xd95bf8d8, 0xffffff80, 0xdf8da9f8, 0x0), at 0xf900c27f
[7] 0xf900c224(0x0, 0xb8, 0x0, 0x10, 0x2e2, 0xd95bf8f0), at 0xf900c223
[8] 0xf90057a8(0xf045cf98, 0xff0163d4, 0xdc69e5e8, 0xf90160b4, 0x8800, 0xd95bf990), at 0xf90057a7
[9] 0xf9005764(0xf045cf98, 0xdb82c000, 0x0, 0xf9015d08, 0xfefbe000, 0xd95bfa10), at 0xf9005763
[10] 0xf9005c2c(0x0, 0x35d4c, 0x0, 0xf9016210, 0x7bac, 0xd95bfa88), at 0xf9005c2b
[11] 0xf9000218(0xd95bfb70, 0xd95bfd70, 0xa, 0xdb82d230, 0xf900a840, 0xd95bfc88), at 0xf9000217
[12] JavaCalls::call_helper(0x1, 0x594890, 0xd95bfc80, 0xd95bfb80, 0x4, 0xd95bfd70), at 0xfe99a808
[13] JavaCalls::call_virtual(0x7a9c, 0x594890, 0x23ca38, 0x23ca44, 0x23ca40, 0xf0465b00), at 0xfeac319c
[14] thread_entry(0xdb82f280, 0x594890, 0x23ce24, 0xff024640, 0xff024710, 0xff024130), at 0xfeae2700
[15] JavaThread::run(0x594890, 0x0, 0xff00c790, 0x6c00, 0x7dd4, 0x7c00), at 0xfeade29c
[16] _start(0x594890, 0x6107, 0xfefbe000, 0x0, 0x4f24, 0x4c00), at 0xfee64240
Does anyone know or have any experience with a similar problem????

Answer 1

Crazy thought here. I noticed in your stacktrace a call to Java_sun_awt_motif_MToolkit_run. Is your solaris server set up with a Motif/X-Windows server? Or is it a 'headless' installation? Many java AWT and Graphics libraries rely on an underlying native windowing system.