Hi everyone,

This is my first post on this forum so hopefully I'll clear and concise. I am running RH9 on a hard disk which is acting as a server. A separate PC (the client) PXE boots to that server and NFS mounts his file system. This was done by installing RH9 onto his (the client) hard disk. Then an image of the client was taken and exploded onto the server. The server was then setup for DHCP, TFTP, and NFS. The client boots up fine however when I go to shut it down or reboot it, I'd get some RPC errors. I moved the following kill scripts in the rc0.d directory: K75netfs -> S00netfs, K86nfslock -> S01nfslock, K90network -> S02network, S01halt -> S03halt. In addition, I removed S00killall. Now when I do a shutdown, the RPC errors are gone and here is an excerpt on my screen:

Sending all processes the TERM signal... [ OK ]
Sending all processes the KILL signal... md: recovery thread got woken
up... [ OK ]
Syncing hardware clock to system time [ OK ]
Turning off quotas: [ OK ]
Unmounting file systems: lockd_down: lockd failed to exit, clearing pid
[ OK ]
Halting system...
flushing ide devices: hda hdc
Power down.
lockd: cannnot unmonitor 192.168.1.1


So I guess my questions are, does the messages look good? Is this really just a hack to see less errors? Is there a better way to setup a diskless client?

Any input is greatly appreciated. Thank you.
-Gil

Everything to me seems fine, only the lockd error is a little problem.. To fix that problem look at this http://lists.debian.org/debian-user/.../msg00837.html and then go to the bottom of the message and hit followup..

What were the RPC Errors? If we know what those errors are then we can get rid of those too.. Which the list of new errors everything else will be fixed, and you will be able to shutdown without any problems..