on CentOS-5. I am trying to do diskless booting. I have configured one
machine as the Head node (with OS and other programs) with DHCP and its
running TFTP-server.
it's valuable to turn on the verbose flag to tftpd so you can
watch syslog to see clients requesting files. for instance, I found
that a bunch of crummy old whitebox 1U servers I have will actually
request "pxelinux.0\xff" rather than the \xff-less name I was expecting.
originally I hacked the tftp server to remove the byte, but this time
around I just made a symlink...
But when I boot the client I am getting the following error message,
"kernel panic - not syncing: VFS: Unable to mount root fs on unknown block
(9,2).
major number 9 appears to be md on my systems. what are you telling the
kernel its root device is? also, since you have all the relevant boot
files on the cd, you don't actually need tftpd, right?
root file system which is not in the CD-ROM. But I want the root file system
along with other directories to be mounted form the head node. Could anyone
guide me to solve this?
what are your kernel arguments? on a full pxe/nfsroot cluster of mine,
/proc/cmdline looks like:
ip=dhcp root=/dev/nfs nfsroot=10.50.5.5:/newclu/,rsize=4096,v3,nolock \
rootfstype=ext2 BOOT_IMAGE=bzImage-2.6.22.2 \
ip=10.50.4.1:10.50.5.5:10.50.63.254:255.255.192.0
(to tell you the truth, I don't know why rootfstype=ext2 is in there -
maybe it's obsolete...)
_______________________________________________
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf