Only one reply to my question, which is included below. Thanks Todd B.
Acheson.
on Wed, Apr 14, 1999 at 09:21:15AM -0400, Todd B. Acheson <acheson_at_ohiou.edu> wrote:
> You didn't state what version of the OS you are using.
>
> There are ADVFS lock problems with patch kit 003 of 4.0D - a global
> lock was added to fix a race condition and they didn't quite get it
> correct. I am not sure of all the activity that might yield a lock
> problem but I know that "rmfset" will lock all activity on the domain
> and render a system "hung". I would consider this possiblity in your
> analysis.
The OS version (4.0e) was in the subject line but I missed the patch
level - the machine has the 1st patch kit installed. Ill have to check
with Digital about the problem and post a second summary - No time for
that now.
The original question was
>Hi
>I have been trying to run defragment on an advfs that is on one 20GB
>partition of a raid. The output of showfdmn and showfsets is appended
>to this mail. I am running the command:
>
>/usr/sbin/defragment -v ee
>
>via a cronjob that starts at 2am. The system hangs about 7 hours
>later! I know that defragmentation does occur 'cause I run defragment
>-nv later and see the stats improve (see output). The only messages I
>can find are:
>
>Apr 14 08:39:56 fs1 snmpd[494]: Closing subagent svrMgt_mib, reason: 0
>Apr 14 08:46:09 fs1 snmpd[494]: Sendto failed: 2
>Apr 14 08:46:09 fs1 snmpd[494]: Closing subagent advfsd, reason: 0
>
>These messages start occuring just after the defragment starts -
>nothing special just before the hang. I suspect that the defragment is
>trying to do something just as it finishes and this is causing the
>system hang.
>
>HELP
BB
--
Bevan Broun ph (08) 9380 1587
Computer Systems Officer fax (08) 9380 1065
Dept. Electrical and Electronic Engineering
University of Western Australia rm. G70
Received on Mon Apr 19 1999 - 01:46:57 NZST