CentOS and Redhat problem: NFS mount at boot up fails with error: “System Error: No route to host”

On our CentOS 4.4 box, I set up the nfs mount by putting the setting in fstab.

10.0.0.2:/export/files /files intr

After that I tested the setting by typing: mount /files Everything worked as expected and files was mounted successfully. At this point I wanted to make sure server will do nfs mounts automatically at boot up. To my surprise once I rebooted the server, nfs mount did not happen. I checked the logs and found this:

May 11 11:00:24 www3 mount: mount: mount to NFS server '10.0.0.2' failed:
May 11 11:00:24 www3 mount: System Error: No route to host.

But once server booted up, I typed mount /files and again, mount worked great. After poking around for a while I found out that when netfs script runs, netfs can not find network routes (due to network is still initializing). I tried few ways to get around this issue and settled on following solution:

vi /etc/init.d/netfs
insert: action $”Sleeping for 30 secs: ” sleep 30
right after: [ ! -f /var/lock/subsys/portmap ] && service portmap start
and right before: action $”Mounting NFS filesystems: ” mount -a -t nfs,nfs4

Basically we are going to give enough time for network to start and set up routes etc. You can play with that number after sleep command (we are basically pausing netfs script execution for 30 secs) and see what works for you. Once you are done, save/exit and reboot the server to see if number you used is sufficient sleep time. You can type df -h to see if it was mounted. If it wasn’t mounted, you can check /var/log/messages to see if nic was starting after nfs mount. If it is after, you will have to increase that number after sleep. Remember that you are only pausing the netfs script execution. Rest of the system continues to boot up. After I made changes, I checked my df output after reboot and my mount was there. I also checked /var/log/messages to see if everything looks ok there:

May 11 11:22:14 www3 kernel: bnx2: eth0 NIC Link is Up, 1000 Mbps full duplex
May 11 11:22:16 www3 netfs: Sleeping for 30 secs: succeeded
May 11 11:22:17 www3 netfs: Mounting NFS filesystems: succeeded

11 thoughts on “CentOS and Redhat problem: NFS mount at boot up fails with error: “System Error: No route to host”

  1. Lawrence Fung

    Thank you very much for the solution. I google a lot because of this same problem happened to my server and now it’s fixed.

  2. Glenn Johnson

    Thank you for this fix. I spent a few hours troubleshooting this problem before I found your blog! It works like a champ!

  3. Reza

    I am so happy that there are people like you who share their experience and knowledge.
    One more thanks to you.

  4. Pingback: CentOS5 (RHEL5) + 9TB Drive + AD + Win/OSX clients + BackupEXEC : Chatswood Computer Consultants Ltd.

  5. Nick

    Try using the “_netdev” option in fstab:

    ctfs1-p:/srv/eis /mnt/nfs nfs intr,_netdev 0 0

    That works for me to make sure networking is up before the mount.

    Nick

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>