Mitigate a loss of NFS resouce

[dbbyleo]

I'm trying to determine if ther are anything I can do to minimize the effects of lossing a NetApp that hosts one of the mounted NFS filesystem on my Sun V490 running Solaris 9.

ServerA using NFS filesystems from 2 NetApp filers - NetApp01 and NetApp02. If I loose NetApp02, the service on ServerA is configured to continue running. However, after doing testing... my problem is at the operating system level when I take-down (turn off) NetApp02 to simulate a complete loss of that filer. The O/S seems to "hang". someone told me it's because the O/S doesn't know that filer is no longer there and any request it makes to the filer, it will hang until the request timeouts, and in most cases, if not all, it will continue to retry.

Are there any configuration I can make to a specific mounted NFS filesystem so that the server doesn't behave this way.

I know that doing a umount -f of the affected mount point fixes it, but it would mean that the services/server would suffer until someone could intervene.

Thanks in advance...

 

[maximusGeek]

It's crude and there is probably a better way to do it but I have a cron job that runs every 15min that checks for the existence of a file on my NFS mount. If not there, then it umount -f and tries to remount the device.