Hello again,
I finally got the nodes to build successfully! In the process I
learned two things that I think might be good additions to some sort
of FAQ.
[1] Ensure that your /boot is mounted prior to attempting to create a
image. Many distributions unmount /boot on successful boot as a
precautionary step. I created two images which had the boot partition
empty, thats when I realized that the boot was not mounted.
[2] For some very very strange reason, when I successfully created a
new image (with boot mounted), in
/var/lib/systemimager/images/<myimage> the proc folder of this image
was behaving strangely. It was actually behaving exactly like a
regural proc file system. During the node installs the strange error
messages I was getting was because the proc folder was being copied
over and that was causing problems. When I noticed this, I knew that
proc should not be copied/rsynced. I went to the image directory and
noticed that the proc folder of the image also had the same entries as
my head nodes proc. What was worse was that it would not let me
delete it, just like you cannot delete your real proc folder. Finally
I had to reboot the headnode, and when it came back up the images proc
folder was empty. I was finally able to build nodes without strange
proc error messages.
I hope this helps someone out!
Vijay Avarachen
On Mon, 7 Mar 2005 13:59:21 -0600, Vijay Avarachen <[EMAIL PROTECTED]> wrote:
> Hello,
> Here is a overview of my setup, 5 HP Visualize workstation (all
> identical), Headnode running RH9.0 with no updates, Using Oscar 4.0.
>
> I get to step 6 with no problems. I am attaching my rpm list file and
> partitioning scheme to this email.
>
> The problem I run into is building the nodes.
>
> [1] The nodes boot successfuly from PXE
> [2] node boots the pxe supplied kernel, loading all the necessary modules
> [3] Partitions the three SCSI drives according the the partition
> scheme I specified in image
> [4] Initially when its starts copying files from headnode, it spits
> out some error messages about /proc
> [5] Continues to copy a lot of files from the headnode, then it pauses
> at exactly /usr/lib/Omni
> [6] After 10-15 mins (during which the Headnode shows a lot of rsync
> activity) the node starts to copy files again.
> THEN IT FAILS :-(
>
> error messages on screen look like:
> stat proc/827/fd: No such file or directory
> rsyn: recv generator: mkdir "proc/835": No such file or directory (2)
>
> I have no idea why it would rsync /proc in the first place. I though
> /dev and /proc were dynamically generated, then why rsync it? I have
> been struggling with various issues (all related to unreliable NIC
> cards in the past) for a very long time. This is the farthest I have
> gotten. I tried this with FC2 and RH9 and I get the same issue. Also
> I just noticed that the node froze when I tried to <shift>+<pg
> up/down> too many times. Any help would be appreciated.
>
> Vijay Avarachen
>
>
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Oscar-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oscar-users
