Just scheduled an image from a stock FOG 0.32 install. Unsurprisingly, it worked fine. So there must be something in the backup/restore process that I’ve munged.
Is there an authoritative backup/restore procedure somewhere? That is something I’d really like to have in my pocket.
In my case, I’m not even changing the IP address. So this should be as simple as it can get.
Blackout,
Thanks for the explanation and suggestion. I’ve verified those details and logged in to the fog account using ftp. It works. The problem must lie elsewhere.
The strange thing is that everything else works. Machines can PXE boot just fine, but scheduling a deploy using the web interface fails. What I mean by this is that, when the “Image all computers” button is pressed, there is a longer than normal pause (maybe two seconds) before the web interface complains:
[CODE]None of the machines were able to be queued!
[delld620-12] Unable to connect to tftp server.[/CODE]
This is happening with FOG 0.32, with the backup data coming from a 0.30 server.
I’ve wiped FOG from the new machine and am reinstalling it. I’ll confirm that it works from scratch (ie. adding a user, image, and workstation by hand before doing a test deploy). If that works, I can point the finger at something coming from the backup data.
Mark
Just to be clear, I used FOGBackup.sh to generate the backup of the existing FOG server.
The restore instructions I’m following ([url]http://fogproject.org/wiki/index.php?title=Restoring_FOG_from_Backup[/url]) talk about removing/restoring the FOG stuff under www and opt directories. However, the backup I made doesn’t contain anything for those directories.
Maybe the backup didn’t complete?
I just moved a FOG server onto different hardware. The IP address remains unchanged.
After restoring from the backup, attempting to deploy through the web interface gets me: Unable to connect to tftp server.
Looking at the restore steps, I think I may have missed something. I’ll try that and check back here.