Jeff Ross <jross <at> openvistas.net> writes: > > Jeff Ross wrote: > > Followup message below: > > > OpenBSD 4.4.-beta i386 (dmesg at the bottom) > > > > This is the same system that I reported a bsd.mp panic on last night. > > Sometime over night the single processor generic bsd kernel panicked as > > well. As you'll see below, when I got my trace and ps I entered a boot > > reboot at the ddb prompt. The system did not reboot. Instead, it > > panicked again. The second trace and ps are for this second panic, and > > the ps shows a lot of processes that should not yet be on a rebooted > > system. > > > > I had someone at the shelter power cycle the server and it successfully > > rebooted. > > > > gnats <at> openbsd.org bounces sendbug e-mail: > > > > Connected to 192.43.244.163 but sender was rejected. > > Remote host said: 553 5.1.8 <jross <at> cas_server.openvistas.net>... > > Domain > > of sender address jross <at> cas_server.openvistas.net does not exist > > > > Thanks, > > > > > > Jeff > >
<snip a lot of trace and ps> > ahc2: target 0 using 16bit transfers > ahc2: target 0 synchronous at 80.0MHz DT, offset = 0x3f > ahc2: target 1 using 16bit transfers > ahc2: target 1 synchronous at 80.0MHz DT, offset = 0x3f > ahc2: target 2 using 16bit transfers > ahc2: target 2 synchronous at 80.0MHz DT, offset = 0x3f > ahc2: target 4 using 16bit transfers > ahc2: target 4 synchronous at 80.0MHz DT, offset = 0x3f > ahc2: target 5 using 16bit transfers > ahc2: target 5 synchronous at 80.0MHz DT, offset = 0x7f > root on wd0a swap on wd0b dump on wd0b > > Right now the system is off. It seems like it will reboot and find > everything if we wait a bit. > > Thanks, > > Jeff > > Final followup... I replaced the 3rd drive and the system has been up for 24 hours now with no more panics. Some googling for the real panic message (included in an e-mail that hasn't made it to the list) "panic: softdep_deallocate_dependencies: unrecovered I/O error" found this message from Henning Brauer in 2001: http://openbsd.monkey.org/tech/200107/msg00043.html and this reply: http://openbsd.monkey.org/tech/200107/msg00044.html I'm still keeping a close watch on the system but hopefully that took care of the panics. Am I correct in believing that *any* panic is a bug? If so, I'll try to get this info into sendbug, even though I can't use it directly from that server. Jeff
