Re: DFly-stable as of April 1st crashed, but no crash dump was possible due to ips driver being unable to do crash dumps

[Tomaž Borštnar <tomaz.borstnar@xxxxxxxx>]

Matthew Dillon wrote:
> :Matthew Dillon wrote:
> :>     Turn off sync_on_panic and see if you can get a dump.
> :> 
> :>     sysctl kern.sync_on_panic=0
> :
> :OK, same error as before (in less than 12 hours!):
> :
> :Warning: cache_inval_vp: race-A detected on pgstat.stat
> :
> :
> :Fatal trap 12: page fault while in kernel mode
> :mp_lock = 00000000; cpuid = 0; lapic.id = 00000000
> :fault virtual address = 0x28
> :fault code = supervisor read, page not present
> :instruction pointer = 0x8:0xc028feb7
> :stack pointer = 0x10:0xe04afc04
> :frame pointer = 0x10:0xe04afc08
> :code segment = base 0x0, limit 0xfffff, type 0x1b
> :	     = DPL 0, pres 1, def32 1, gran 1
> :processor eflags = interrupt enabled, resume, IOPL = 0
> :current process = Idle
> :current thread = pri 44 (CRIT)
> :interrupt mask = none <- SMP: XXX
> :trap number
> :panic: page fault
> :mp_lock = 00000000; cpuid = 0; lapic.id = 00000000
> :boot() called on cpu#0
> :
> :syncing disks...
> :
> :
> :--------------------------
> :
> :Tomaž
>
>     Do you get a debugger prompt? 

nope, "syncing disks..." is last thing printed on console. I have 
DDB_UNATTENDED in kernel config file.

> We need at least to get a stack
>     trace ('trace' command from the debugger prompt).  A kernel
>     core would be even more useful.
Never been able to get into ddb on this machine if I remember correctly. 
Maybe define DDB_TRACE to at least get something?

>     I also need your kernel binary for any of the numbers to make any
>     sense.
Ok, will mail you link to kernel file.

Tomaž