Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Fri, Jul 24, 2009 at 09:48:50AM +0200, Rainer Traut wrote: > Lars Ellenberg schrieb: >> On Mon, Jul 20, 2009 at 10:02:54AM +0200, Rainer Traut wrote: > ... >>> Crash happened at 06:26 while verifying vm2. >>> >>> I uploaded a screenshot of console here: >>> http://gallery.ubliga.de/drbdcrash.jpg >> >> thanks for the report. >> >> do you have any more output about what has happened before? >> netconsole? serial console? >> >> is this a self-compiled drbd, >> or some pre-build rpm? > > Hi, > > sorry no more output than the screenshot. > > It's a self-compiled rpm btw. > based on the Centos-testing drbd83 rpm with very little changes for 8.3.2. ok. but that means I cannot see where exactly it happened, because the offsets are different for different compiles. could you please do: resolve() { local s x i o; s=$1; x=${s#*+}; s=${s%+0x*}; for i in `seq -16 8 16` ; do o=$[$x + $i]; printf "l *(%s+0x%x)\n" $s $o; gdb -q -batch drbd.ko -ex "l *($s + $o)" -ex q ; done } resolve w_e_end_ov_reply+0x8b and send me the output? you probably need to do that in the directory where you build the module. or at least need to point gdb to the correct drbd.ko that still has debug symbols, and the source code should still be where it was when you compiled... > A subsequent verify run was fine though. Will report back if I > experience another crash. For any such crash, please again do the above resolve magic, with the symbol and offset printed in the "RIP" line. That gives us information where exactly something went wrong, and we then may be able to tell why just by looking at the code. Thanks. -- : Lars Ellenberg : LINBIT | Your Way to High Availability : DRBD/HA support and consulting http://www.linbit.com DRBD® and LINBIT® are registered trademarks of LINBIT, Austria. __ please don't Cc me, but send to list -- I'm subscribed