> 1. Set up your dhcp server to dole out unique IP addresses to your > clients based on MAC address. Ok, so the actual IP's I would have otherwise installed on each server. > 2. Mount your GFS partition as /tftpboot My setup includes multiple storage devices. I can either have multiple devices or I can create one large filesystem for all of the nodes in this cluster. Which do you think is best since I can keep adding or taking from a volume group. > 3. In your /tftpboot, create root directories for your clients, and put > what you need there. As in server images and such? This I have to read about, the whole idea is new to me as of this morning :). >You still need a minimal set of these libs and such on the client root fs in >order to boot far enough to where the fstab can actually mount these things >though. My machines share CDROM/Floppy so can't boot from those automatically. I do have network boot on each server however and of course, I'd like not to have to use any drives on any of the blades. Then I could put more into the central storage. > 4. Of course you need nfs serving out the data, and xinetd accepting > tftp connections. I'm using fibre channel so all of the nodes see the storage as their own. Thanks for all this info, I'll continue looking for more and build one of these. I am sure it's exactly what I need over having to build multiple machines all the time. Mike -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
