Michael,
Your AIX update probably over wrote the AFS hooks you had installed when making
the machine into an AFS server. I don't know about AIX 3.1.5, but AIX 3.2 AFS
servers had to replace the "fsck" "program helper", /sbin/helpers/v3fshelper
with the one from Transarc.
------------------------------------------------------------------------
Bill Pitre Battelle PNL, Box 999 M/S K1-87, Richland WA 99352
[EMAIL PROTECTED] Phone: 509-375-2091 FAX: 509-375-6631
------------------------------------------------------------------------
"We all need to be salvaged every once in a while."
-Troy Thompson (Battelle PNL)
On Jan 24, 3:26pm, Michael D Sofka wrote:
> Subject: AIX partitions taken off-line.
>
> Has anybody else seen this? We have had two fileservers restart with
> the message:
>
> Assertion failed! file afsfileprocs.c, line 5588.
>
> in FileLog. This is with AFS 3.3a, and I've seen this before when
> there are disk media errors. But, when AFS was restarting multiple
> partitions could not be salvaged. The message in SalvageLog was:
>
> 01/23/96 14:53:27 SALVAGING FILE SYSTEM PARTITION /vicepq (device=lv17)
> 01/23/96 14:53:37 File system /dev/rlv17 is in a bad state.
> 01/23/96 14:53:37 Call your IBM representative.
>
> This happened to 5 partitions on 2 disks the first time, and 6 partitions
> on three disks the second time. When AFS came back up, large numbers of
> the volumes on these partitions could not be attached. (Yes, I'm also
> going to post this to comp.unix.aix.)
>
> Stopping AFS and running fsck gave me:
>
> # umount /vicepq
> # fsck /dev/lv17
>
> ** Checking /dev/rlv17 (/vicep)
> ** Phase 1 - Check Blocks and Sizes
> ** Phase 2 - Check Pathnames
> ** Phase 3 - Check Connectivity
> ** Phase 4 - Check Reference Counts
> ** Phase 5 - Check Inode Map
> ** Phase 6 - Check Block Map
> Superblock is marked dirty; FIX? y
> 52595 AFS files, 405 non-AFS files,
> 53000 files 2532432 blocks 285616 free
> ***** File system was modified *****
> # mount /vicepq
> mount: 0506-324 Cannot mount /dev/lv17 on /vicepq: There is an input or
> output error.
>
> For all the partitions. In order to re-mount the disks, I had to
> reboot the machines (after touching SALVAGE.fs to force a salvage).
>
> The servers are running AIX 3.1.5.1, with SSA and other disks. The
> first time the dirty superblock was on two SSA disks. The second time
> on three non-SSA disks. The media error associated with the first
> restart was *not* on the SSA disks which reported the dirty
> superblock. In the second case, the dirty superblock was on one of
> the disks reporting a media error, but the other two disks did *not*
> report media errors. I suspect AIX, since it is the only thing
> changed recently (we were running 3.1.5 prior to adding SSA and FDDI).
>
> Mike
>
>
> --
> Michael D. Sofka [EMAIL PROTECTED]
> ITS Systems Programmer TeX, AFS, NIS, NTP, Astronomy,
Epistemology.
> Rensselaer Polytechnic Institute, Troy, NY. http://www.rpi.edu/~sofkam/
>
>-- End of excerpt from Michael D Sofka
