Am 01.01.2007 um 12:31 schrieb Chris:
Is it possible to make it auto run debugger, then auto run backtrace
then dump the output to disk so people who have no local access can
get backtraces?
No, but you can enable crashdumps and run the debugger on the dump
afterwards. Add
dumpdev="AUTO"
to /etc/rc.conf, then run /etc/rc.d/dumpon start. When the next
panic occurs, the kernel will write the dump to the configured swap
space, and the system will save the dump to /var/crash on rebooting.
See
http://www.freebsd.org/doc/en_US.ISO8859-1/books/developers-handbook/
kerneldebug-gdb.html
on how to get a backtrace from that dump.
If you're short on space in /var, and/or you have a large amount of
RAM, you can set the sysctl debug.minidump to 1 to have the kernel
only dump relevant portions of memory, instead of everything. This,
however, is experimental in -stable, afaik.
Stefan
--
Stefan Bethke <[EMAIL PROTECTED]> Fon +49 170 346 0140
_______________________________________________
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
