Re: nfs mount painfully slow in %post

["Taylor, ForrestX" <forrestx taylor intel com>]

Andrew M. Williams wrote:
> redhat 8.0
>
> I've done a ton of digging and a bunch of fiddling but I can't figure
> out why issuing an nfs mount in the %post section takes about 5 minutes
> to return.  (the mount does work)
>
> I was able to get it to return normally but I had to do the following:
>
> * get to the shell in <alt>f2, chroot to /mnt/sysimage
> * run /etc/init.d/network start
>    this kills the ip on eth0
> * manually set the ip on eth0
> * manually reset the routes
> * start portmapper
> * start nfs
>
>
> Shouldn't this work better than it does?  Is there anyway around this?
> I think I'm probabbly missing something dumb but I can't for the life of
> me figure out what it is.

It is trying to get a lock.  Try mount -o nolock,wsize=8192,rsize=8192 
server:/dir

The -o nolock option should keep it from trying to lock it, causing a 
10-20 minute timeout.

Forrest