How to gracefully restore NFS drives and applications after network going down?

I’m having a bit of a problem with drives connected by NFS and KDE4.4.4. The setup is that a server running oS11.2 and KDE4.4.4 gives access to it’s shared drives via NFS to two laptops (one with 11.2 the other with 11.3, both running kde4.4.4) and a desktop PC (running oS11.3 KDE4.4.4). The clients have no problems accessing the shared drive on the server via NFS. The problem comes when the network goes down or the connection to the network is lost.

Any applications open at the time that are ‘looking’ at the shared drive just hang forcing me to kill the application. Dolphin is a classic example. Even after the network comes back up Dolphin will not start properly and just opens a blank window. Starting dolphin from a console reveals nothing and the console just drops back to the command line with no output. Logging out and back in doesn’t help so it’s something at system level.

My question is how can I make this more robust and allow the system recover gracefully and without taking down the applications when the network is lost?

I think


su -c 'rcnfs restart' 

is what you’re looking for.

But I’d rather take care that the network does not go down. All these network protocols need the network to be there.

Thanks Knurpht. I guess I could create a little script with that command with an icon to it so the users can restart the service themselves or can this be automated? I.E if the network is lost it’s automatically detected and the nfs service is restarted.

I agree about taking care about the network going down but sometimes there is no choice or control. Sometimes the laptops use wifi to connect and sometimes the signal drops. I’m surprised that the KDE apps don’t recover and just hang if a network drive is lost unexpectedly. In this regard WinXP did a better job of dealing with cases like this - one of the few things though.

Would autofs be appropriate here to try automate the reconnection of lost network drives?