iPXE boot not working after latest upgrade
-
Did you change the boot file in Option 67?
-
I tried undionly.kpxe. Should I be using a different boot file option?
-
Also,
Thanks Wayne! I didn’t, don’t worry. The errors I’m seeing go something like this when I boot:
CLIENT MAC ADDR: <mac> GUID: <guid>
CLIENT IP <IP> MASK: <SUBNET MASK> DHCP IP: <DHCP SERVER>
GATEWAY IP: <GATEWAY IP>
TFTP.
PXE-MOF: Exiting Intel Boot Agent.
Selected boot device failed. Press any key to reboot the system.I’ve tried this on 2 of our most common machines and I see the same on both. I’m testing on Dell Optiplex 9020 desktop and Dell Latitude 6430u laptop. Thanks all!
-
Please look through this: https://wiki.fogproject.org/wiki/index.php/Troubleshoot_TFTP
Please try restarting TFTP, and getting a file from TFTP.
Also, verify your Firewall is off.
For CentOS, the Fedora commands ought to work.
Why are you editing /var/www/html/fog/lib/fog/Config.class.php ??? Please don’t do that…
Ok, BEFORE you do any of the troubleshooting above, please re-install FOG. -
Before I go re-installing FOG, I have tested TFTP and that hasn’t worked. I do tftp <IP> get undionly.kpxe and I get a timeout. To clarify: to re-install, can I just run the trunk upgrade in fog/bin/install.sh on my current install?
-
@sarge_212 said:
Before I go re-installing FOG, I have tested TFTP and that hasn’t worked. I do tftp <IP> get undionly.kpxe and I get a timeout. To clarify: to re-install, can I just run the trunk upgrade in fog/bin/install.sh on my current install?
Yes.
It’ll re-use the settings you chose when you last installed. -
So iPXE is now working. I did restorecon -Rv /tftpboot and that did it. The problem I’m seeing now is it doesn’t seem to be pushing the image down to the machine. Here are some cell phone shots of that:
Thanks again Wayne and Tom for your awesome feedback and help. I sure appreciate it!
-
Please examine this:
https://wiki.fogproject.org/wiki/index.php/Known_FOG_Bugs#The_Image_Resize_BugI think you might be affected.
-
Luckily I had a snapshot of the VM before I upgraded. I reverted back to stable 1.2.0 and now I can push images to machines. However, I’m seeing an incredible slow down on spinning disk vs SSD. Spinning disk is going very quickly, however SSDs are taking twice as long to image. I tried the upgrade trunk method and that kind of borked things. I’m thinking I might try that again but if you all have any suggestions for speeding up deployment to SSDs, that’d be appreciated.
-
The only way I know of to speed up deployments (at all) is to upgrade to the latest Trunk version of FOG.
Knowing that you utilize snapshots, I don’t see why you shouldn’t (because you can always go back).
The only time threads don’t get any suggestions/comments is when everybody who hangs here are either stumped or busy; both are very uncommon. Not trying to be harsh but, in this instance, I did give you something to look at and 4 days later you come back saying you simply reverted and didn’t even comment on the suggestion whatsoever.
-
I get that, Wayne, and I don’t think that’s harsh. I should have commented on your suggestion. Better late than never: The weird thing I saw was when I did a download with debug, and then ran fdisk was I only saw one partition. I didn’t really understand that and still getting my feet wet with FOG decided to revert back to a snapshot before going any further. I really do appreciate the help and knowledge you guys have, I would be sunk without your help!! Thanks again. I’m going to try upgrading to the latest Trunk version and see what happens. Thanks!
