rpc.mountd dies when server accessed through automounter

[Wendy Stabenow]

HPUX 11.11 rpc.mountd dies when server accessed through automounter.
cd /net/hostname takes a long time to come back, but it will connect to the server. Once connected, however, the rpc.mountd dies making it impossible for any other connection. Seems like an NFS issue. Most of the NFS patches I've seen have warnings issued with them. Anyone know of a fix for this?

[RAC_1]

I do not know much about automount. Recently there was a reply from "Dave Olker-The NFS Guru" about debugging rpc.mountd. Send a kill -17 to rpc.mountd. This puts in debugg mode
and logs messages to a file (/var/adm/syslog and some files under that)

This should give you some information to start with. Also you may want to search the thread which talks about debuuging rpc.mountd.

The other thing to look at is "patch - nfs, autountd"

Also you may want to look at "completely eliminating atomountd and go for autofs" autofs is latest and off course free.

Hope this helps you.

Anil

[Anil C. Sedha]

Wendy,

Do you see any messages in /var/adm/syslog.log or

message "ABORT in
add_reclock" is logged in the /var/adm/rpc.lockd.log file

This might require a patch.



Willy

[Wendy Stabenow]

I am seeing no errors in either syslog.log or mountd.log. I saw another thread on this problem that said nsswitch.conf was wrong. I copied that file from a working server and this had no effect either. Tried the kill -17 on rpc.mountd, but didn't see anything different. I'm not familiar with how it is supposed to work though. Still searching for the rpc.mountd debugging thread you spoke of Anil, but have not found it.

[RAC_1]

I did some goof up. Following is the thread. (but is about rpc.locked)

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=702900

May be you can put an email to Dave for this.

Also did you look at automountd patches and completely elimanating automound and use autofs.

Hope this helps.

Anil