Hi,
I received a two replies and one "me too!" to my posting (see below for
original posting). It will take time to follow up on these suggestions and
verify success since I can't make the problem happen. I will post a follow
up posting if I get anywhere.
Tom Barkanic recommends:
>"Some things you mite try are increasing the number of nfsd's. Mine is set
>to 16. Another is make sure you have all the NFS/AdvFS patches for 3.2.
>The other is uping the NFS timeout value for your clients. Take a look at
>the mount man page for timeo and retry."
I'm not running AdvFS and I couldn't find any patches that address this
particular problem, will keep looking. My nfsd is set to 16
I will take a look at timeo and retry.
Robert L. McMillin asks:
>"You don't say whether you've checked the server daemons to see whether
>they still exist on the server machine. Have they crashed?"
Sorry I didn't check, I will the next time the problem happens.
Thank you,
Diane Ibaraki
University of Hawaii
High Energy Physics
<diane_at_uhheph.phys.hawaii.edu>
-------- original posting ----------------------------------------
>Hi,
> I'm having problems with one of my nfs servers:
>DEC 3000/600 with a PMAZC-AA (dual scsi)
>OSF1 uhhepi.phys.hawaii.edu V3.2 214 alpha
>servers data information only, *doesn't* serve home or mail partitions.
>
>At least once a month, this system appears to be having portmapping problems.
>The symptoms are, all nfs clients systems complaint with:
>
>Jan 7 06:16:16 uhhepj vmunix: NFS3 server uhhepi not responding still trying
>
>Anyone trying to log in after this point will get hung on login. The nfs
>server itself works happily along.
>
>Eventually, I attempt a shutdown of the server and shutdown hangs with this
>message: "Portmapper not responding still trying"
>
>I usually end up recycling the power, then all is well after the reboot. I
>check the error logs and systems messages on the nfs server, but I can't
>find anything that will give me a clue as to what is going on. Has anyone
>seen this behavior, or can anyone give me any hints how to troubleshoot
>this problem?
>
>I must admit that my users do have a tendency to hose the network (running
>jobs from the client systems that access disks on the nfs server). But
>would this cause something to break on the server? Any parameters I should
>be tweaking?
>
>TIA,
>Diane Ibaraki
>University of Hawaii
>High Energy Physics Group
><diane_at_uhheph.phys.hawaii.edu>
Received on Thu Jan 30 1997 - 20:38:27 NZDT