--=.)yeIcQBQNg.hh8
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

an interesting while-we're-waiting-for-a-debugging-kernel tidbit...

saw this on stderr while untarring srcsys.tar.gz:

crime: memory error address 27bc8c00 status 2040000
Aug 14 23:34:58 neptune /netbsd: crime: memory error address 27bc8c00 status204
0000 
crime: cpu error 4 at address 46048c4
Aug 14 23:35:01 neptune /netbsd: crime: cpu error 4 at address 46048c4
crime: cpu error 4 at address 46048c8
Aug 14 23:35:03 neptune /netbsd: crime: cpu error 4 at address 46048c8


the system seems to be unaffected, but who knows.
  --david



On Sat, 14 Aug 2004 23:06:31 -0400
david l goodrich <dlg@dorkzilla.org> wrote:

> On Sun, 15 Aug 2004 00:24:15 +0200 (CEST)
> Hubert Feyrer <hubert@feyrer.de> wrote:
> 
> > On Sat, 14 Aug 2004, david l goodrich wrote:
> > > root file system type: ffs
> > > /: bad dir ino 270370 at offset 8192: mangled entry
> > > panic: bad dir
> > > db>
> > >
> > > so... a filesystem error?
> > 
> > looks like it... what does fsck(8) say?
> > 
> 
> fsck on the next boot says:
> 
> --[start]--
> Starting file system checks:
> /dev/rsd0a: INCORRECT BLOCK COUNT I=287202 (2 should be 0) (CORRECTED)
> /dev/rsd0a: FREE BLK COUNT(S) WRONG IN SUPERBLK (SALVAGED)
> /dev/rsd0a: SUMMARY INFORMATION BAD (SALVAGED)
> /dev/rsd0a: BLK(S) MISSING IN BIT MAPS (SALVAGED)
> /dev/rsd0a: 99522 files, 569211 used, 919958 free (902 frags, 114882 blocks,
> 0.1% fragmentation)
> /dev/rsd0a: MARKING FILE SYSTEM CLEAN
> --[end]--
> 
> (yes, sd0a is the only partition.  small disk, didn't want to limit myself)
> # df
> Filesystem  1K-blocks     Used     Avail Capacity  Mounted on
> /dev/sd0a     1489169   569216    845494    40%    /
> kernfs              1        1         0   100%    /kern
> #
> 
> i have the whole boot sequence logged if anybody wants it.
> 
> i'll set to work on building a kernel with debugging enabled.  never done it
> before, should be an adventure :]
>   --david
> 
> > 
> > > > but here's the backtrace:
> > > > db> bt
> > > > 80343004+214 (87fff000,bf390000,0,d) ra 8029b6bc sz 0
> > > > 8029b52c+190 (87fff000,d,0,72) ra 802401c0 sz 40
> > > > 802401b4+c (87fff000,d,0,72) ra 0 sz 0
> > > > User-level: pid 633.1
> > 
> > Um, that's not useful.
> > Can you make a kernel with debug-symbols and check that, possibly by
> > forcing a crash dump ("reboot 0x104") first, then running gdb on the saved
> > core.
> > 
> > 
> >  - Hubert
> > 
> > -- 
> >                          ,,_
> > If wishes were wings,  o"   )~  would fly.            -- Go www.NetBSD.org!
> >                         ''''
> 
> 
> 



--=.)yeIcQBQNg.hh8
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (OpenBSD)

iD8DBQFBHuvPsw9Pt+GRQPURApv0AJ9hGzkUlgNQJgM9z0NYyqOmzSUrpQCgvrng
iXjCBEeuvz6v85YM16tUu2w=
=2YgP
-----END PGP SIGNATURE-----

--=.)yeIcQBQNg.hh8--