From: Shaun.Racine@intier.com
Date: Wed Oct 23 2002 - 07:29:33 EDT
I knew it had to be simple.
Answer from Nagaraj S;
> If I have understood your querry following will script
> solve your problem
>
> -----start
> #!/bin/sh
> while true;
> do
> nfsstat -s >> nfs.out
> sleep 60
> done
>
> -------End
> run this with &
>
> tail -f nfs.out
>
> for your VIEW
Thanks to Yogesh Bhanu, not what I originally asked but provided me with
further analysis ideas;
> Try using "collect" which is better and If you have some
problems with nfs ..
> Well easiest way out is to check if all your NFSD daemons are being used
.
> Try lsof |grep nfsd . Also try upping the value under NFSIOD under
/etc/rc.config from default 8 to
> higher (better in muliples of 8 ). Check also using nettstat if your NIC
is not throwing up errors .
> Also try and look under /var/adm/syslog/ there are lot of log files
which can throw some light.
> Maybe the problem is somehwere else .
>
> If you have compaq analyze you can analyze binary.errorlog.
Thanks also auto replies from people letting me know they are out-of-office
;-)
Original question follows;
----- Forwarded by Shaun Racine/MISGroup on 23/10/2002 12:12 -----
Shaun Racine
To: tru64-unix-managers@ornl.gov
23/10/2002 cc:
10:06 Subject: how to capture output from nfsstat ?
Hi Managers,
This seems so simple but its driving me nuts. I want to record the output
from nfsstat in a file.
We have a regular system panic on our ES40 (v5.1A), Compaq support have
suggested it may be overload of NFS service, which I doubt.
I can use nfsstat to monitor this online, but I want to record output in a
file to view after system crash. The commands I have tried are;
nfsstat -s -i 60 > nfs.log 2>&1
echo 'nfsstat -s -i 60 > nfs.log 2>&1'|at now
nohup nfsstat -s -i 60 > nfs.log 2>&1 &
But in each case the logfile always remains 0 length with no data in it!
One that sort of works is;
echo 'nfsstat -s -i 60'|at now
and the output is captured in mail after the process is terminated, but I
can't guarantee the mail would appear in the event of a crash.
I welcome all suggestions.
Regards,
Shaun Racine
Systems Development Manager
email: shaun.racine@intier.com
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:57 EDT