Help: NFS error (No route to host)

Help: NFS error (No route to host)
Please help me I am new to this forum and i really need help.
I have set a NFS server and exported the /var/ftp/pub directory to all clients on the network 192.168.0.0/255.255.255.0 and have run the the following commands:
exportfs -a -v
service nfs start
chkconfig nfs on
service portmap restart
chkconfig portmap on

they all where successfuly done, but still when I try to mount this directory
/var/ftp/pub i get the an error message similar to this nfs "192.168.0.254" server down

also I made the same shared directory on a client PC and when I try to mount it from the server i get this error message:
mount: RPC: Remote System Error - No Route to Host

Please Help Me

can any one help me ????

Yipeeeeeeeeeeeeeeeeeeee
I have solved the problem

Hi, i noticed u have solved the problem. Would u mind tell me how to solve it ? I'm also having it. Been cracking my head but still no answer. Thanks

Regards
Daniel

Originally posted by systemhackerz Yipeeeeeeeeeeeeeeeeeeee I have solved the problem

Yipeeeeeeeeeeeeeeeeeeee I have solved the problem

It would be greatly appreciated if you posted how you fixed this problem, so other people who are sitting in your former shoes can get it working too.

Originally posted by linux-rulz It would be greatly appreciated if you posted how you fixed this problem, so other people who are sitting in your former shoes can get it working too.

Well Dont be sad my friend the solution was very easy just disable your iptables
$service iptables stop

If you want this service to keep runing with the NFS just add it as a rule to the iptables allow list

Or in rhel run the command

% system-config-securitylevel

and add the firewall exception

nfs:tcp

This still didn't solve the problem for me, but reading this, I recognized it as one of the things I needed to do to make it possible.

I'm still getting

mount: mount to NFS server 'myhostname.mydomain.edu' failed: System Error: No route to host.

System Error: No route to host

This means the machine that you are connecting from is not getting a TCP/IP ACK back from the server.

If turning off/adjusting iptables and/or selinux didn't fix it check your dns entries (use dig or nslookup) to make sure the ip address is correct or use the ip rather than the hostname.

Hi guys,

I'm having the same problem and I've done all the above steps. I'm running Cent0S 4.4

Even stopping the iptables doesn't help.

The client server is able to ping the nfs server. I am using the mount command with the IP address directly.

I have in my /etc/exports file:


/root/install 192.168.1.112(ro,sync)


I ran exportfs -a -v

I started (and re-started) the nfs and portmapper, nfs is in my firewall rules,


I'm not sure what else to try (?)



Thanks
Nat

oops my bad, I had a type in /etc/hosts on the IP address on the nfs server.

All is good now.

Nat

I was having similar problem. It was only happening when I rebooted the server tho. So i figured it out and put a solution on my blog @ http://crazytoon.com/2007/05/11/cent...route-to-host/

Hopefully it will help somebody. I ran into this thread and wasn't able to get my answer from it when i was searching for answers.

Originally Posted by systemhackerz Well Dont be sad my friend the solution was very easy just disable your iptables $service iptables stop If you want this service to keep runing with the NFS just add it as a rule to the iptables allow list

This is ABSOLUTELY NOT good advice. Don't just go turning off iptables unless you know what you're doing. And even if you do know what you're doing it's not a good idea.

The correct answer to this post is to say, it's probably your firewall setup, add nfs tcp/udp to your firewall configuration to allow your client access to the server. If you don't know what that means, dig around a bit, and replace the redhat iptables service with something like shorewall.

/me does the "2005 originated thread brought back to life" dance

The Answer to this as in what the previous person stated is correct. DO NOT just disable IPtables...

the issue is that the NFS needs a tcp/udp port through the firewall on the host from the client. Once the (for testing purposes only) firewall was disabled I was able to mount the NFS share. Next step is to learn which port is needed opening to permit MFS mounting from the client to the host

I'm having the same problem upgrading from RH-9 (Shrike). I have no problem mounting from my old "Shrike" box, but can't connect to the "Werewolf" box unless iptables is disabled.

Looking at the nfs man page, it says


(for option port=n)
>
The numeric value of the port to connect to the NFS server on. If the port number is 0 (the default) then query the remote host’s portmapper for the port number to use. If the remote host’s NFS daemon is not registered with its portmapper, the standard NFS port number 2049 is used instead.
>

It would seem, then, that the server would have to have a portmapper running in order to tell the client what port to come in on. Shrike has a portmap service but Fedora does not. Something else must provide this service, but what is it?

By the way, I added 2049/tcp and 2049/udp with system-config-securitylevel to no avail.