@alireza2n May I include your configuration in this tutorial? https://forums.fogproject.org/topic/10944/using-fog-to-pxe-boot-into-your-favorite-installer-images

@bramen I can’t speak so much to the grub stuff.

What I can tell you is that FOG has the ability to run a bash script after the image has been pushed to the target computer and before the computer reboots after completing the deployment. You can use this feature to tweak settings on the target computer before the computer is first booted into the target OS. This is used commonly with the windows environment to place hardware specific drivers onto the target computer that is used during the OOBE/WinSetup phase of windows. In linux we’ve used these scripts to rename linux computer and do some post install activities.

The FOG imaging agent is Linux based so it can’t directly interact with ms windows, but it can leave bread crumbs behind for windows to act upon. For linux target computers it can do more things. I can’t say if it can run grub commands though.

The concept of the fog post install script is that the script will need to mount the target computer media/file system, do whatever is needed then disconnect from the target media.

For windows I have a tutorial here: https://forums.fogproject.org/topic/11126/using-fog-postinstall-scripts-for-windows-driver-injection-2017-ed While this is windows specific the setup program would be very similar for linux.

@sebastian-roth Thank you for the response! I’m assuming I should keep EXT4. I haven’t updated this thing in a while. I was planning to take a snapshot first but wanted to have a little better clarity. I’m running 1.58 so I will also update to 1.5.9 as well.

@wt_101 If I can suggest you look at this post: https://forums.fogproject.org/topic/11126/using-fog-postinstall-scripts-for-windows-driver-injection-2017-ed

it gives you an idea how to do this with windows. You can use the same concept, except for maybe look for the partition that has an /etc directory.

Here is a bit older script that shows a few different techniques. https://forums.fogproject.org/topic/7740/the-magical-mystical-fog-post-download-script

@vertimyst said in Running both CloneDeploy and FOG?:

but since I have CD on our server already I can’t get anything to boot to FOG over our network - it all defaults to CD.

Its not clear exactly what you are doing here. FOG doesn’t use a CD/DVD it boots over the network. FOG is similar to CloneDeploy or Clonezilla but different.

For FOG you will set dhcp options 66 to the IP address of your fog server and dhcp option 67 to the boot file name. The boot file name depends on what format your computer’s firmware is in. For a BIOS computer, dhcp option 67 is undionly.kpxe and for a UEFI computer the boot file name is ipxe.efi.

Once those values are set in your dhcp server then you pxe boot your target computer into the FOG iPXE menu.

I have a feeling since you referenced CD several times you might be doing something non-standard with FOG. Can you explain a bit more of what you are trying to do?

@george1421 Hello! I fired up a new VM and installed Debian 10.11 (buster). I did a modified installation from the YouTube video you shared, I did partitioning based on this video - https://wiki.fogproject.org/wiki/index.php?title=Debian_8

I installed 1.5.9 and setup the DB and changed the fog default password.

My next question is - can I export my settings off of my old FOG server into the new one? Currently I have both VMs fired up so I can simply toggle between the two IPs in a tabbed browser.

Once I get the new server setup, I am going to change my DHCP and PXE settings on my Windows network to test PXE booting my new computers that use the USB dongle. I am going to be beyond thrilled if this works as I have several hundred of these machines that need imaged.

Thank you so much for your help and look forward to your response!

Gratefully,

Michael

Hello, I came to this topic because I was having this same issue with some new Dell computers that arrived here where I work, in my case I didn’t needed to repair the grub, but I could solve this by changing some bios configuration where there’s a option to add uefi boot option and there appears the .efi file of the grub and mark it to boot first. Hope it helps!

@george1421 Ok, thanks.

If so, i will not touch it:)

@sebastian-roth said in The initial storage node is much smaller than the disk space I gave my VM?:

sudo resize2fs -p /dev/mapper/ubuntu–vg-ubuntu–lv

That all did the trick, VM is now showing the full allocated disk space!

Thank you so, so much for your help, and also for giving me a little homework! I’ll read up on LVM as you said.

Thank you again for your help, I can’t thank you enough!

@chris_unit said in Controlling exit:

That way we can book to windows using exit to refind, and exit to GRUB UEFI for ubuntu.

You are on the right track on the other thread. The two refind options are what you need. This way you can target where refind looks for the boot files. You can include and exclude file paths to get what you need.

The grub4win implementation inside fog is bios only. BUT that’s not to say you couldn’t use a similar concept to the dual refind hack to leverage grub too. There are a few ways to go about it. I don’t have to deal with dual boot so I’ve never looked into streamlining that flow.

Hi @Sebastian-Roth,

Now works fine the FOG client 🙂

Thanks for your fast reply

@sebastian-roth The error with chkconfig is still here and I dnd’t found any topic about it in the forum, is any alternativ possible to this package ?

@jc35
mount /dev/sda1 /mnt
mount --bind /sys /mnt/sys
mount --bind /proc /mnt/proc
mount --bind /dev /mnt/dev
mount --bind /dev/pts /mnt/dev/pts

chroot /mnt ./usr/sbin/grub-install /dev/sda

@chris_unit said in Intel NUC UEFI PXE menu:

and set the vendor class to PXEClient:Arch:00007
is that the right vendor class for an intel nuc?

I can’t answer that. I can tell you it will either be type 7 or 9 for 64 bit uefi computers.

If you want to find out what type the nucs are setup wireshark on a witness computer (second computer on the same subnet). Run a capture filter of port 67 or port 68 PXE boot the target computer. In the Discover packet sent out by the target computer, in the dhcp option 93 or 94 (sorry can’t remember at the moment) the target computer tells the dhcp server what type it is.

@sebastian-roth
I completed the upgrade without any major problems.
I had to reinstall some php packages that had been removed during the upgrade
(the comparison between the packages installed on the old version and the new one allowed me to reinstall them easily)

I also had to re-run installfog.sh because I had the same problem as here:
https://forums.fogproject.org/topic/15254/503-service-unavailable-after-debian-upgrade.

Also, unrelated to fog, I had to review the configuration of mariadb and more particularly the authentication mode of the “root@localhost” user.
Indeed the user “root@localhost” was still using authentication with a password, but the new method is to use auth_socket :

+-----------+------------------+-----------------------+-------------------------------------------+ | Host | User | plugin | authentication_string | +-----------+------------------+-----------------------+-------------------------------------------+ | localhost | root | mysql_native_password | XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX | +-----------+------------------+-----------------------+-------------------------------------------+ | Host | User | plugin | authentication_string | +-----------+------------------+-----------------------+-------------------------------------------+ | localhost | root | unix_socket |

this is all I can say

@george1421 Well I’m kind of answering my own question here.

php -m | grep -e gettext -e mysqli

If they are installed on the OPs system then they should be visible with the php -m command.

@JasonNaughton As we had another person run into a similar looking issue I had a closer look at the code and might have found why this happens. Please find all the details here: https://forums.fogproject.org/post/144691

Hi,

It was my firewall, I have rebooting a rule and now all is fine.

Thank u

@chris_unit With some solid Linux knowledge you surely can convert an LVM install to a plain one but probably re-installing is a lot faster if the system is not highly customized.

@chris_unit said in mysql not starting when trying to install 2nd time:

did you install on 7.2?

I haven’t. Somehow thought this was a typo when you first mentioned version 7.2. Quoting from: https://mirrors.xtom.de/centos/7.2.1511/readme

This directory (and version of CentOS) is deprecated. For normal users,
you should use /7/ and not /7.2.1511/ in your path. Please see this FAQ
concerning the CentOS release scheme:

https://wiki.centos.org/FAQ/General

If you know what you are doing, and absolutely want to remain at the 7.2.1511
level, go to http://vault.centos.org/ for packages.

Please keep in mind that 7.2.1511 no longer gets any updates, nor
any security fix’s.

Sorry but we don’t have enough people to work on FOG for it to support officially deprecated OS versions.