topic Re: nfs problem in Operating System - HP-UX

nfs problem

Binu_5 — Tue, 12 Jul 2005 00:16:47 GMT

Hi aLL

Iam facing a stange nfs problem

In the system /local directory is exported in the server
we will be able to access from other systems also but intermintantly ig give nfs server not responding error for this local directory

All the nfs services are running in this systems ..
If I give showmount -a it gives the following error

showmount: mfptest1: RPC: Program not reGISTERD

Is this error has anything to do with
the NFS problem

Thanks
Binu

Re: nfs problem

Deepak Kulkarni — Tue, 12 Jul 2005 01:24:53 GMT

Hi Binu

have you disabled the sunrpc port?

Regards
Deepak

Re: nfs problem

Binu_5 — Tue, 12 Jul 2005 01:30:54 GMT

Hi Deepak

How to check wheather the port is disabled

Binu

Re: nfs problem

Sudeesh — Tue, 12 Jul 2005 02:28:35 GMT

Hi,

Usually you will receive 'Program not registered' error when there is an issue with daemons running.You can verify nfs related process using rpcinfo command.

rpcinfo -p
rpcinfo -u mountd : to verify rpc.mountd is running fine.

You can receive â server not respondingâ messages when the server or
network is heavily loaded and the RPC requests are timing out. Try
doubling the timeo mount option for the directory.

/etc/fstab
nfs_server:/local /nfs_mount nfs nosuid,timeo=14 0 0

If lot of clients are mounting nfs shares from this server try increasing the number of nfsd daemons running by increasing NUM_NFSD variable in /etc/rc.config.d/nfsconf.

Sudeesh

Re: nfs problem

Sudeesh — Tue, 12 Jul 2005 02:30:17 GMT

Something more that will be helpful.

This is an abstract from 'troubleshooting nfs service' document.

To Diagnose NFS Performance Problems
1. Issue the following command on several of your NFS clients:
nfsstat -rc
2. If the timeout and retrans values displayed by nfsstat -rc are
high, but the badxid value is close to zero, packets are being dropped
before they get to the NFS server.
Try decreasing the values of the wsize and rsize mount options to
4096 or 2048 on the NFS clients. See â To Change the Default Mount
Optionsâ on page 44.
See Installing and Administering LAN/9000 Software for
information on troubleshooting LAN problems.
3. If the timeout and badxid values displayed by nfsstat -rc are of
the same magnitude, your server is probably slow. Client RPC
requests are timing out and being retransmitted before the NFS
server has a chance to respond to them.
See â To Improve NFS Server Performanceâ on page 326.
Try doubling the value of the timeo mount option on the NFS clients.
See â To Change the Default Mount Optionsâ on page 44.
4. If the null value displayed by nfsstat -rc is greater than 1%, the
automounter is trying too frequently to mount a directory from
multiple servers. It sends out null procedure calls to all the
configured servers and mounts the directory from the server that
answers first.
Increase the time between mount attempts by adding the -tm 60
option to the AUTO_OPTIONS variable in
/etc/rc.config.d/nfsconf, as follows:
AUTO_OPTIONS=â -f $AUTO_MASTER -tm 60â
Then, restart the automounter. See â To Restart the Automounterâ on
page 90.
Continue to increase the value of the -tm parameter until the null
value displayed by nfsstat is less than 1%.
5. Issue the following command on any machine on the network:
netstat -i
Troubleshooting NFS Services
Performance Tuning
Chapter 8 325
The number of collisions (Coll) divided by the number of output
packets (Opkts) is the collision rate. If your collision rate is greater
than 10%, consider dividing your network into smaller segments and
putting an NFS server on each segment. See Installing and
Administering LAN/9000 Software for information on dividing your
network.

Sudeesh

Re: nfs problem

vinod_25 — Tue, 12 Jul 2005 02:37:26 GMT

Hi Binu

The problem is that rpc.mountd on the HPUX machine is not being started.
This is started from the inetd program. To start this, edit the
file /etc/inetd.conf on the HPUX machine.

The line that follows should be uncommented:

rpc dgram udp wait root /usr/etc/rpc.mountd 100005 1 rpc.mountd

Then, issue the command:

/etc/inetd -c &

This will reread the /etc/inetd.conf file and rpc.mountd will now start when a mount request comes in from a foreign node.

Regards

Vinod K

Re: nfs problem

Deepak Kulkarni — Tue, 12 Jul 2005 02:41:09 GMT

Hi Binu,

"nmap" is the utility for this. this will be installed by default linux.

Reds
Deepak