PXE Freezes
-
I have setup FOG on a Ubuntu 12.04 server and PXE boot freezes giving many random characters such as
" [2.670399] <co 00 01 00 c0"The server seems be running correctly, the image has been created and task started. The only thing I noticed was that homepage dashboard does not show Disk Information. It says "failed to connect to DefaultMember.’
I have double checked that all redirects are set correctly to the secondary drive, “/media/ImageDisk/” yet it does not detect it either.
-
What FOG version and kernel are you using?
-
FOG .32 default Kernel. The PC is a HP Copaq G1 600.
The display that it shows is many lines such as this one:
[2.625149] call trace:
[2.625149] [,c0428a5a.] ? vsnprintf+ 0x2da/ox430 -
eddieduce,
Have you tried creating a link to /images from /media/ImageDisk?
Also,
Did something happen that has created all this havoc with your system? Possible a bad disk, rebuild, etc…
It sounds to me like either the /tftpboot/pxelinux.cfg/default file may be corrupt, or the pxelinux.0 file.
-
@Tom- Ubuntu is Server 12.04 server, clean install. PC Desktop GUI was added. There were problems getting the server to take a static IP. That was finally resolved.
I followed the Change NFS location instruction from the FOG wiki. Reviewing that again, it does not seem to show a link creation section. I’ll try that.
-
Try my kernel, it looks like your kernel is in panic/freeze mode.
To get my kernel, open a terminal on the FOG Server. Login as root or, as you’re using ubuntu, make sure your user has sudo permissions.
To test sudo try:
[code]sudo touch /tmp/someTestFileIAmMakingUp[/code]It should ask for your password, enter it. If you get no errors, you should be set.
Then cd into the tftpboot/fog/kernel folder.
[code]cd /tftpboot/fog/kernel[/code]
Copy your old kernel just as a backup.
[code]sudo mv bzImage bzImage.orig-29JAN2014[/code]Then get my kernel and try again:
[code]sudo wget --no-check-certificate https://mastacontrola.com/fogboot/kernel/bzImage[/code]
Hopefully all will work again.
-
I got a ‘unrecognized option’ message for ‘–no-get-certificate’.
-
I just updated it, sorry thinking of wget where I wanted a check
-
message ‘failed. Name or service unknown.’
-
try prepending with sudo, I updated the code.
The next thing, make sure wget is actually install.
[code]sudo apt-get update
sudo apt-get install wget[/code] -
‘sudo apt-get update’ failed. with many file not found messages.
‘sudo apt-get install wget’ is updating but is barely on 9%.
Would you recommend just re-installing FOG. This is the production server. The downtime has extended too far now. Would you recommend circling back to .29?
-
How did you upgrade your system?
Did you just upgrade or create a new “Fresh” install?
-
Fresh install. What I mean to say is,
is it a FOG .32 issue or more a kernel corruption issue?
The original system is giving problems and I decided to rebuild a new one (the one we are discussing).
-
And you’re behind a very slow internet connection?
I don’t have the fastest of upload speeds, but they are 5MB, plenty enough that a normal download of my kernel should take about 5 minutes (max). It’s only just shy of 6 MB
-
no, I am installing wget. …28% at the moment.
Correction, I am updating packages, and it is barely at 42%.
-
The process finished. W-get for the kernel was unable to resolve host address for your site .
-
Try again, dns for somereason was lost, fixed now.
-
No go. ‘Name or service not known. Unable to resolve host address <… >’
-
Okay, try again.
I had to reboot both of my routers. I was testing multicast and it brought down my network.
-
Still unable to resolve.
