I have just installed Redhat 8.0 on a home LAN. Lan seems to work fine - I can ping other machines using either their symbolic name (from etc/hosts) or their IP address.

However NFS mounts fail with a RPC Portmap message. I also tried telnet, to get connection refused. Firewall is disabled and hosts.accept and hosts.deny are empty so I don't believe this is a security or permissions issue. etc/exports is also set up correctly.

I removed the entries from etc/hosts and found that NFS mount worked fine using the IP address of the box I wanted to link to. But put the entries back in and NFS mount fails even if I use the IP address rather than the host name.

If this wasn't so basic, I'd suspect a bug. Any thoughts on further options to try?

(Sadly, smbmount for the one remaining windows machine works well throughout.)

Try:

# showmount -e <ip from the machine that is exporting>

Also, do you have portmap runing? are you usinf nfs or unfs?

showmount show the correct exported file list. I simplified it to the bare minimum bto elimante this as a cause of the problem.

/ *.localdomain

rpcinfo -p shows portmapper, nfs, mountd running.

I've disabled the firewall (iptables -F) and emptied the hosts.access and hosts. deny files.

In the machine that you are mounting the directory the portmap is runing? The domain in this machine is localdomain? Can you try to remove from the machine that is exporting from the /etc/exports the *.localdomain to make a test?

re telnet - i think newer linux distros do not include telnetd as they want you to use ssh
are rpc.statd and rpc.mountd running on the server?

It is running on my red 8.0

for the NFS , first check if the nfsd is running !!!
ps aux | grep nfs

normally you get something like:

oot 4514 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4515 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4516 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4517 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4518 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4519 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4520 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]
root 4521 0.0 0.0 0 0 ? SW 09:51 0:00 [nfsd]

otherwise run :
/etc/init.d/nfs start


after that check
if you have the export directory in /etc/exports
something like ( put the directory you want to export )
/opt/var/ftp/pub
without option you export for the everybody


otherwise run
exportfs -a ( for all the export directories)

now check to mount the nfs on your own machine (the nfs server) by using the IP of your own machine to avoid the connection problem.

for the telnet, you run it via the xinetd.

To run it simply, run the setup, choose the services, and check the telnet service (same thing for ftp ) save the configuration.
restart the xinetd with

/etc/init.d/xinetd restart


hope this help you

bela

/usr/sbin/in.telnetd has to be present on a server for you to telnet to it.

telnet-server-0.17-7.i586.rpm or similar contains it.

Many thanks for all the suggestions, esp bela whose step by step guide was rather better than the manual!

I had already followed the steps outlined but with no success. I fixed temporarily by taking all local name entries out of etc/hosts and using the IP address directly.

Incidently there is a bug (76543) on buzilla that seems to have symptoms very close to the problem