Re: -current panic w/linux netscape
> :I finally saw my first -current panic in more than 5 months (not a bad > :track record). > : > :I have a panic that I can duplicate with a 24 hour old "make world" > :and a 4 hour old -current kernel. If I run the linux netscape (installed > :from ports less than a week ago), the kernel panics in copystr(). > :Minimal gdb output follows. I'll have to generate a kernel with > :debug info for more information. > : > :The linux netscape worked fine with my old kernel from 9/16. > :I'm willing to try and track this down, but I hope someone > :will pop up and have an idea what is wrong. > : > :-Mike I must have caught two badly timed cvsups in a row. After doing another cvsup and recompiling/installing a kernel and modules (which I had done before), everything is working fine. I usually ignore any glitches after a fresh "cvsup/make world", except in this case it took me 3 cvsup's to get a fully-functional system. I'm still not sure exactly what was out of sync here, but after examing the crash dump with full symbols, it really looked like something wasn't in sync (i386/trap.c:syscall() had some very odd values it was trying to use). I know how to debug kernels, I just haven't had to do so for such a long time, so I wasn't prepared :-) And doing it this time, just reminds me how lacking gdb is in some respects compared to other kernel debugging tools I've used in the past in some areas. In other areas it is much better. Thanks for taking the time to respond. -Mike > Compile your kernel with debugging. In the kernel configuration file > in /usr/src/sys/i386/conf/YOURCONFIGFILE, add: > > makeoptions DEBUG="-g" > > Then config the kernel and rebuild. > > When you make install, the non-debug version will be installed and the > debug version will be sitting in the compile directory as 'kernel.debug'. > > Then reboot, then cause the crash again and get another core (be sure > to clear out space from /var/crash if necessary). > > Now bring the system up again, and use this for your gdb: > > cd /var/crash > gdb -k /usr/src/sys/compile/YOURCONFIGNAME/kernel.debug vmcore.XX > > And do the 'back' again to get the stack backtrace. You should see a > whole lot more information. > > -Matt -- Mike Pritchard [EMAIL PROTECTED] or [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: -current panic w/linux netscape
Mike Pritchard wrote: > I have a panic that I can duplicate with a 24 hour old "make world" > and a 4 hour old -current kernel. If you use the linux module, make sure it's in sync with the kernel. If that doesn't help, either follow Matt's advice or use ktrace/truss. -- Marcel Moolenaarmailto:[EMAIL PROTECTED] SCC Internetworking & Databases http://www.scc.nl/ The FreeBSD projectmailto:[EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: -current panic w/linux netscape
:I finally saw my first -current panic in more than 5 months (not a bad :track record). : :I have a panic that I can duplicate with a 24 hour old "make world" :and a 4 hour old -current kernel. If I run the linux netscape (installed :from ports less than a week ago), the kernel panics in copystr(). :Minimal gdb output follows. I'll have to generate a kernel with :debug info for more information. : :The linux netscape worked fine with my old kernel from 9/16. :I'm willing to try and track this down, but I hope someone :will pop up and have an idea what is wrong. : :-Mike Compile your kernel with debugging. In the kernel configuration file in /usr/src/sys/i386/conf/YOURCONFIGFILE, add: makeoptions DEBUG="-g" Then config the kernel and rebuild. When you make install, the non-debug version will be installed and the debug version will be sitting in the compile directory as 'kernel.debug'. Then reboot, then cause the crash again and get another core (be sure to clear out space from /var/crash if necessary). Now bring the system up again, and use this for your gdb: cd /var/crash gdb -k /usr/src/sys/compile/YOURCONFIGNAME/kernel.debug vmcore.XX And do the 'back' again to get the stack backtrace. You should see a whole lot more information. -Matt To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
-current panic w/linux netscape
I finally saw my first -current panic in more than 5 months (not a bad track record). I have a panic that I can duplicate with a 24 hour old "make world" and a 4 hour old -current kernel. If I run the linux netscape (installed from ports less than a week ago), the kernel panics in copystr(). Minimal gdb output follows. I'll have to generate a kernel with debug info for more information. The linux netscape worked fine with my old kernel from 9/16. I'm willing to try and track this down, but I hope someone will pop up and have an idea what is wrong. -Mike Script started on Sun Sep 26 05:47:00 1999 mpp 2# gdb -k /usr/src/sys/compile/MPP/kernel /var/crash/vmcore.0 GNU gdb 4.18 (no debugging symbols found)... IdlePTD 3264512 initial pcb at 29f1c0 panicstr: from debugger panic messages: --- Fatal trap 12: page fault while in kernel mode fault virtual address = 0x7 fault code = supervisor write, page not present instruction pointer = 0x8:0xc0222d50 stack pointer = 0x10:0xcb0bdc80 frame pointer = 0x10:0xcb0bdce8 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = 385 (communicator-4.6) interrupt mask = none panic: from debugger panic: from debugger dumping to dev #wd/0x50001, offset 131072 dump 191 ... 1 --- #0 0xc013a644 in boot () (kgdb) bt #0 0xc013a644 in boot () #1 0xc013a9e9 in panic () #2 0xc011f435 in db_panic () #3 0xc011f3d5 in db_command () #4 0xc011f49a in db_command_loop () #5 0xc012151f in db_trap () #6 0xc0215210 in kdb_trap () #7 0xc02240ac in trap_fatal () #8 0xc0223d85 in trap_pfault () #9 0xc02239c7 in trap () #10 0xc0222d50 in copystr () #11 0xc0dc7209 in ?? () #12 0xc0dc5fbd in ?? () #13 0xc0224316 in syscall () #14 0xc0215b06 in Xint0x80_syscall () #15 0x28abf564 in ?? () -- Mike Pritchard [EMAIL PROTECTED] or [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message