NFS mount problems on AIX

UNIX AIX
1

Hi,

I have two machines (AIX) each on a different VLAN.

Need to mount a filesystem using nfs on the other one.

When I export the nfs file system its a breeze. But when I try to mount it on the other machine the smitty command hangs on "running" and i get an OK from smitty but with this error.

mount: 1831-010 server APPSERVER not responding: RPC: 1832-018 Port mapper failure - RPC: 1832-008 Timed out
mount: backgrounding
APPSERVER:/datadb

I have restarted nfs deamons. I have checked the /etc/hosts values. Opened the 2049 and 111 ports on the network but still cant do.. :frowning:
Help! :wall:

2

@aixromeo
check out those links to trouble shoot , if nothing works post here again
link1
link2
link3

3

try to telnet from client to servers nfs ports (2049 111 Idont remember listening port numbers of nfs)
or remove access list and check if it success

4

Connect to the NFS client (not the server with the exported filesystem) and run the command

stopsrc -g nfs; startsrc -g nfs

Then try to map the filesystem on the client again.

Might work as it appears that part of the NFS system is not running.

Good luck!

5

Thank you for you replies.

Well I have done one thing. Opened Telnet sessions for both the machines and tried to telnet with the nfs ports one by one from each machines:

telnet <IP> <2049>
telnet <IP><111>

well from the node that has exported the mount point, I get the escape characters and it connects.

From the node where the mount point needs to be mounted
I run the same telnet commands but it doesnt work. Asked the network team to open the ports lets see. Will post if it gets fixed. :slight_smile:

6

Note that there are some auxiliary services (for instance NLM) NFS uses which use random ports.

Also: NFS uses either the portmapper- *or* the rpc.bind-daemon to advertise service endpoints to the clients. rpcbind uses 111 as you said already but you have to make sure this (and not the portmapper) is used.

You might want to consult this link for some more information and several other ports maybe involved.

I hope this helps.

bakunin

7

Thank you everybody who replied :slight_smile:

The issue has been resolved. The network team had opened the nfs ports one way. i.e Machine #1 to Machine #2 was ok but vice versa that was not the case..

It is mounted now :slight_smile:

Thanks again