UFS related panic (daily <-> find)

[rank1seeker at gmail.com]

> On Wednesday, October 02, 2013 5:40:02 pm rank1seeker at gmail.com wrote:
> > > > > > Ok, here is another one, same case, just this time under 
> > 9.1-RELEASE-p7
> > > > > > 
> > > > > > ==============================================
> > > > > > Fatal trap 12: page fault while in kernel mode
> > > > > > fault virtual address   = 0x25
> > > > > > fault code              = supervisor read, page not present
> > > > > > instruction pointer     = 0x20:0xc082c552
> > > > > > stack pointer           = 0x28:0xe7eed7a8
> > > > > > frame pointer           = 0x28:0xe7eed7ac
> > > > > > 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         = 63645 (find)
> > > > > > trap number             = 12
> > > > > > panic: page fault
> > > > > > Uptime: 11h16m47s
> > > > > > Physical memory: 1014 MB
> > > > > > Dumping 143 MB: 128 112 96 80 64 48 32 16
> > > > > > 
> > > > > > #6  0xc0898d4c in calltrap () at 
> > /usr/src/sys/i386/i386/exception.s:169
> > > > > > #7  0xc082c552 in inodedep_find (inodedephd=Variable 
"inodedephd" 
> > is 
> > > > not 
> > > > > > available.
> > > > > > )
> > > > > >     at /usr/src/sys/ufs/ffs/ffs_softdep.c:2073
> > > > > 
> > > > > Please go to frame 7 and do 'x/i $rip'.
> > > > > 
> > > > 
> > > > (kgdb) up 7
> > > > #7  0xc082c552 in inodedep_find (inodedephd=Variable "inodedephd" 
is 
> > not 
> > > > available.
> > > > ) at /usr/src/sys/ufs/ffs/ffs_softdep.c:2073
> > > > 2073    /usr/src/sys/ufs/ffs/ffs_softdep.c: No such file or 
directory.
> > > >         in /usr/src/sys/ufs/ffs/ffs_softdep.c
> > > > (kgdb) x/i $rip
> > > > Value can't be converted to integer.
> > > 
> > > Oh, this is i386, use "$eip" instead of "$rip", so 'x/i $eip' at 
frame 7.
> > 
> > 
> > (kgdb) x/i $eip
> > 0xc082c552 <inodedep_find+13>:  cmp    %ecx,0x24(%eax)
> 
> Ok, so %eax must be 1.  I think you probably have failing RAM with a 
stuck bit 
> or some such.
> 

Today I've just finished HDD scan with recoverdisk and there were 3 bad 
sectors.
It was stuck on them for a 15 hrs, until it finally did read whole disk,
Then I've run it again and it read HDD 100%, without a glitch.
I don't know was it a firmware realocated those or those looooong read 
attempts fixed a thing.

Then reboted into single user and run fsck, which detected a LOT 
unreferenced inodes at /usr, which it successfully reconected.
Finally fsck again to get clean, non error output.

Could that caused a panics?

PS: I'll run a memtest86+ when I get some time. For how long do you advise?


Domagoj