We fixed this by adding the following to our dhcpd.conf
[CODE]class “pxeclients”{
match if ((substring(option vendor-class-identifier, 0, 9) = “PXEClient”) or (substring(opti
filename “pxelinux.0”;
max-lease-time 120;
}
[/CODE]

Alright man,

Sorry I wasn’t able to help out too much but good luck. Maybe I can help with your other issues then.

The IOCTL error isn’t particularly important then. I see it with my home-made kernel because it has RAID drivers built in, but the devices don’t have raid drives. If you’re having issues with imaging a specific drive, it may point you in the right direction, but by itself it’s perfectly fine.

edit (nano or vi) " /etc/default/tftpd-hpa " and add “sleep 30” on the top line so it looks like :

[I][B]sleep 30[/B][/I]
[I][B]# /etc/default/tftpd-hpa[/B][/I]
[I][B]# FOG Modified version[/B][/I]
[I][B]TFTP_USERNAME=“root”[/B][/I]
[I][B]TFTP_DIRECTORY=“/tftpboot”[/B][/I]
[I][B]TFTP_ADDRESS=“0.0.0.0:69”[/B][/I]
[I][B]TFTP_OPTIONS=“-s”[/B][/I]

will have to restart the service or reboot

easy way of finding default gateway and subnet mask is to set the workstation to dhcp - wait until it gets and ip address - fo to the information - note down the dg and subnet - pop out to a terminal try pinging and address up the top or down the bottom of the ip range (to find out a address that hasn’t been allocated out) find a non responding address and use that along with the dg and subnet mask details… most if not all subnet masks will be 255.255.255.0 …

One of the things I’ve learned with building a custom kernel is to Read the friendly Wiki page or (RTFW) as other so like to display.

If you’re building the kernel on a 64 bit system, make sure your make commands are:

make ARCH=i386 <command>

So when you’re doing the menuconfig do:

make ARCH=i386 menuconfig

Then make your changes and save the new config.

Then when you create the bzImage file do:

make ARCH=i386 bzImage.

I would almost recommend making this an actual part of the WIKI rather than just a note, as it wouldn’t hurt anything to do this from any platform whether i386 or 64 bit.

Thank you,

Try this: [url]http://www.fogproject.org/wiki/index.php/Setting_up_ProxyDHCP[/url]

When you get to step 7, modify line [B]dhcp-range=192.168.1.10,proxy[/B] (almost at the end of the file) to match your DHCP server IP address (dhcp-range=x.x.x.x,proxy).

This is a solution when you cannot -or you don’t want to- modify DHCP options 66 and 67. I have it working this way because options 66 and 67 are already busy.

Don’t think there’s a way to do it with “chain.32”. I found other references to “pxechain.com” from the linux utilities.
I added the following and it seemed to start booting but unfortunately didn’t boot WDS:

LABEL Windows Deployment Services
kernel pxechain.com
append IPADDRESSGOESHERE::boot\x64\pxeboot.n12
MENU LABEL Windows Deployment Services
TEXT HELP
Loads the Windows Deployment Services Imaging Environment.
ENDTEXT

So I’m going to give up on this one and do what everyone else seems to have done.
Which is to put the linux boot files onto the WDS location and modify from there instead.
Reference: [url]http://www.syslinux.org/wiki/index.php/WDSLINUX[/url]

Actually, someone found this as well (multi-PXE boot) which I might try out first:
[url]http://www.vcritical.com/2011/06/peaceful-coexistence-wds-and-linux-pxe-servers/[/url]

OK it seems to be a grub2 issue, this thread explains it [url]http://fogproject.org/forum/threads/grub-issues-reboot-loop.3904/[/url]

view hidden files:

[CODE]ls -a

ls -al

ll[/CODE]

Turns out the power outage caused the original (physical) fog server to turn on when the power was restore. I arrive at this conclusion after I removed the network cables form the machine that fog should be on and stopped the another old virtual fog server on another, yet I was still able to PXE boot with fog. After calling the previous tech guy we came to the conclusion that the original fog server had to be on, and it was. It was a combination of miss-information and mother nature. I did find that the original fog servers mysql database was not running and I was unable to start it again. Now the virtual fog server is up and running fine. Thanks for the help and fast responses, sorry wasting your time 😄

I am having an issue where I can’t upload a duel boot system because the ubuntu system is a extended partition. Anyone have any ideas how to do that?

Lol, I found the solution on the first link after googling…

Move your FOG* files to another directory than /etc/init.d/, install you package, and then replace the FOG* files 😉

Awesome info Dustin! We use static IPs on our printers, but we have quite a few HP 4000 series printers, so this is good to know. Good find.

I’m installing fog on a workstation to use as a server. I’m having problems setting up a static IP. I have a workstation that i will be using as a server to re-image about 30 laptops. Computer is connected to the local network such as veh.local domain. when i install Linux my internet connection lets me get to the internet, however when i try to configure my connection setting to a static IP, I’m no longer able to get into the internet. The connection reminds connected connected but when i open my Firefox browser the internet does not work. The Question is how do you properly configure a STATIC IP and if possible ,how would i be able to find my proper IP sub-mask and gateway to enter correctly. PLEASE HELP CLARIFY THIS PROCEDURE. Currently running Ubuntu 12.04 LTS with fog 0.32. internet connection is needed to download actual fog .032 after Linux install. what I’m i doing wrong ??? PLEASE HELP.

[quote=“xNRSVx, post: 9911, member: 3028”]I’ve been using FOG .32 for just about 2 weeks now, and absolutely love it. I’d like to integrate images for Apple computers (both Intel and PowerPC hardware), but don’t wish to setup DeployStudio or NetRestore; I’d prefer to utilize the bootable factory image that you’d normally get on DVD or through the app store (the .DMG file). My FOG server is running on an isolated network, and I believe is setup to be the DHCP. Would it be possible to setup FOG in such a way that I could NetBoot the Apple units directly to the server, customize the PXE menu, and simply add an option pointing to a DMG file (similar to adding ISOs with the “raw” flag)?[/quote]

You can incorporate clonezilla with FOG to image your MAC systems, in order to do this follow the instruction on the link provided below:
and as far as PXE booting MAC systems are concerned you can also do that as well. You need to create a CD/DVD image of gPXE-1.0.1+ and boot your MAC systems from the CD/DVD and then it will boot from your current PXE.

[url]http://www.clonezilla.org/livepxe.php[/url]

[quote=“william.nunn, post: 11251, member: 939”]We are currently looking for a solution to image our Imac 20’s. I can the IMac’s to gpxe but I can’t get them to do anything past the fog screen. I was wondering what kernel you were using and what method you was using to image your MACS off of the fog server?[/quot

Once you get the gPXE-1.0.1+ CD/DVD image then you can incorporate that with Clonezilla Live for PXE to image your MAC systems.
The following post will do the trick :

[url]http://www.clonezilla.org/livepxe.php[/url][/quote]

I did some firewall configs a while back and posted them [URL=‘http://fogproject.org/forum/threads/firewall-config.27/#post-75’]here[/URL]. They’re far from perfect, but they might help point you in the right direction.

[quote=“Mark Playford, post: 10120, member: 498”]
Change “TFTP_FTP_PASSWORD” and “TFTP_FTP_PASSWORD” to be the same as the fog user
[/quote]

Mark, that is the same variable… I believe what you meant was:

[QUOTE]
Change “TFTP_FTP_PASSWORD” and “STORAGE_FTP_PASSWORD” to be the same as the fog user
[/QUOTE]

[quote=“chad-bisd, post: 10634, member: 18”]When you install FOG on Centos, which OS do you choose? I don’t know the package manager that it’s trying to use or where it’s getting its packages.[/quote]

I choose 1 for Red Hat.

[quote=“chad-bisd, post: 10634, member: 18”]You might try installing UDPCast manually using your package manager, or just downloading it from the udpcast website and compiling it on your own, or getting a binary and see if it works on your version of Linux. If you can get udpcast setup, you can comment out the lines in the installer scripts that are trying to get it from the package manager.[/quote]

I was able to get a binary of a newer version (20080914) but couldn’t find one for 20071228 that would run on centos 4. I installed it, then commented out the line in the install script for udpcast and now the install fails on the FOGMulticastManager. Documentation said this might happen and to just rerun the install again and it should work but it fails everytime.

Thanks for trying to help by the way. Really appreciate it.