Subject: panic: ffs_nodealloccg: map corrupt
To: None <current-users@NetBSD.ORG>
From: Erik Bertelsen <erik@sockdev.uni-c.dk>
List: current-users
Date: 11/17/1997 09:45:25
On an i386 (newly installed within the last two weeks), where
the kernel and userland has been sup'ed and rebuilt on 13 Nov,
we saw a panic this morning:

cg=15 irotor=60 fs=/usr
panic: ffs_nodeallocg: map corrupt


_panic(...)
_ffs_nodealloccg(f084e500,10,0,81a4,f084e500)
_ffs_hashalloc(f084e500,f,6d53,81a4,f019b828) + 0x57
_ffs_valloc(f2e33db4,f2e33e0c,f2e33e20,f2e33ee0,25) + 89
_ufs_makeinode(81a4,f0853200,f2e33ef0,f3e33f04,f2e33e84) + 78
_ufs_create(f2e33e0c) + 24
_vn_open
_sys_open
_syscall(5)

Does this ring any bells? Any suggestions or explanations ?

The machine died during the weekend, probably about 22:08
Sunday evening, and its main task has been to run the perl
based mrtg application, i.e. retrieving some SNMP  statistics
from a bunch of routers.

The above mentioned backtrace was written down on a piece of
paper and then keyed into this message -- no guarantees for
100% correctness.

Also I regret to say that I do not have a dump (DDB's reboot
did reboot, but produced no dump, I'll have to see how to
get the dump, if it happens again).