LinuxQuestions.org - NFS: System Error: No route to host (retrying)

- Linux - Networking (https://www.linuxquestions.org/questions/linux-networking-3/)

- - NFS: System Error: No route to host (retrying) (https://www.linuxquestions.org/questions/linux-networking-3/nfs-system-error-no-route-to-host-retrying-481472/)

NFS: System Error: No route to host (retrying)

Hi,
I am configuring NFS for my LAN as following:
The server(192.168.2.101) has nfs, nfslock, portmap running. The /etc/exports file is:

Code:

#this has only one line

/home      192.168.2.103(rw,sync)

On the client(192.168.2.103), nfs,netfs,nfslock and portmap are running
The /etc/fstab on the client is:

Code:

#the local mount directives is cut

192.168.2.101:/home          /home    nfs    bg,soft,nfsvers=2  0  0

Trying to mount NFS volume on client give the following error in /var/log/messages:

Code:

localhost mount[3323]: mount to NFS server '192.168.2.101' failed: System Error: No route to host (retrying).

On the client, issuing

Code:

ping 192.168.2.101

give correct result, dropping rate =0%
On the server,

Code:

ping 192.168.2.103

also give droping rate =0%
Does anyone have any idea what may goes wrong?

PS:The document i used to config NFS is
http://www.linuxhomenetworking.com/w...FS#NFS_And_DNS
which also mentions that the client must be able to use host to look up server name and the reverse but does not say how to implement it. Does it actually cause this error?
Hope anyone can give a hint
Thai

Have you run

Code:

exportfs -ra

on the server to actually to insert the share into the table of exported file systems?
Well, actually this is just a wild guess.
No route to host errors is what I get when I try the wrong IP adress.
Since you can ping, maybe it's a firewall issue?
Are your rpc.nfsd and rpc.mountd started?
Have you tried mounting the share somewhere in /mnt (where it should be)?
What happens when you type

Code:

mount -t nfs 192.168.2.101:/home /mnt

Hi, thank you for your reply.
I did use

exportfs -ra

serviceconfig shows:

nfs
rpc.mountd (pid 1995) is running...
nfsd (pid 1992 1991 1988 1987 1986 1985 1984 1983) is running...
rpc.rquotad (pid 1970) is running...

Try mouting NFS volume manually as you advise yield the same error.
Any other suggestion? Since both user and client use portmap, i guess there's nothing to do with the firewall since we never know what port will be assigned to NFS after the handshaking.

2002

There is a way to force the port to always be the same (that what I had to do). You will also need to check your /etc/hosts.allow file. I think everything you need should be in this link.

http://forums.fedoraforum.org/forum/...ight=nfs+setup

Lazlow