if you end up in ddb try boot crash. If your swap and /var/crash is big enough you might get a code that you can use to extract the crash.
On Thu, Apr 30, 2009 at 02:12:58PM +0200, Michael wrote: > Hi, > > Owain Ainsworth wrote: > > On Thu, Apr 30, 2009 at 09:53:54AM +0000, Stuart Henderson wrote: > >> On 2009-04-30, Michael <bele...@bsdmail.de> wrote: > >>> Hi, > >>> > >>> when using the current i386 snapshot kernel my system crashes during > >>> boot, shortly after network initialization. > >> sorry, this report is no use at all without details of the crash, > >> use pencil and paper if you have to. > >> > >> - last line/s printed before it happened > >> - what the panic/uvm_fault message is > >> - trace, as an absolute minimum the function names and offsets > >> - ps, as an absolute minimum show what entry is marked with * > >> > >>> kqemu: kqemu version 0x00010300 loaded, max locked mem=519228kB > >>> DDB symbols added: 339616 bytes > >>> scsibus1 at softraid0: 1 targets > >>> sd0 at scsibus1 targ 0 lun 0: <OPENBSD, SR CRYPTO, 003> SCSI2 0/direct > >>> fixed > >>> sd0: 50156MB, 512 bytes/sec, 102719531 sec total > >>> auich0: measured ac97 link rate at 48000 Hz > > > > reproducing it without kqemu would be good too if you want anyone to pay > > attention. > > The dmesg is with the working 4.5 kernel. kqemu gets loaded from > rc.securelevel there, but it doesn't even get as far as loading kqemu > with the snapshot kernel, already tried deactivating kqemu, but it > crashes before that part get reached, somewhere after network > initialisation and rc.securelevel. > > Big problem here is that the kernel crash output is scrolling really > fast, I could probably use pen and paper, but only for the last 25 or so > lines because scrolling up won't work. :-( > > > Michael
