@Tom-Elliott

Great news!

I updated my server to Debian Bookworm, restarted the Fog installation, and the accent problem disappeared. Awesome!

Thanks for your help and happy new year 2026, all the best for the FOG project!

@Tom-Elliott Thank you, Tom. The init version is now 2025xxx. I’ll test soon to confirm the SSD info is showing in logs.

@brent could you specify exactly at which point of the installation you see this error? If you notice the installer hangs or fails specifically during the “pinning” stage, it is likely a name resolution issue.

I’ve found a solution for this: you need to manually edit the Windows hosts file at C:\Windows\System32\drivers\etc\hosts and add an entry with your FOG server’s IP address followed by the name “fogserver”. This ensures the SmartInstaller.exe recognizes the server and can download the certificate properly.

Just adding my case here in case it helps someone else with the same error.

I’m running FOG 1.5.10 on RHEL 9.5. In my case the problem was not iPXE, kernels or init.xz, but a broken NFS setup on the FOG server.

When I tried to rerun installfog.sh, it always failed at “Starting NFS server” with a systemd dependency error. Checking systemctl list-dependencies --failed nfs-server.service showed that one dependency was a failed mount coming from /etc/fstab. It was an old ISO mount pointing to a path that no longer existed.

Because that mount failed, nfs-server could not start, and FOS Linux failed to mount its root filesystem over NFS, causing the kernel panic on UEFI clients.

After removing the invalid mount from /etc/fstab, running systemctl daemon-reload, fixing NFS exports and starting nfs-server, I was able to complete installfog.sh and UEFI PXE boot started working again.

@pbriec Sorry that your post got lost in the shuffle, I saw it pop up with @raul’s response.
I imagine/hope you got past this in the last year but just in case, the fix would most likely be resetting the host encryption from the web ui (removes some token related database entries on the host record) then restarting the fog service on the client.

If that didn’t do the trick, then restarting the fog service along with resetting the hots encryption should fix it.

@nicolas-moraes I managed to solve it and in my case it was not a FOG or network issue. I verified that DHCP was working correctly (confirmed with tcpdump), options 66/67 were unchanged, TFTP was running, and the FOG web/ipxe boot.php were reachable.

The key test was that when the client stayed at “Start PXE over IPv4”, the FOG server was not receiving any DHCP or TFTP requests at all, which pointed to a client-side problem. Even though Secure Boot and Fast Boot were already disabled, the issue was caused by the BIOS state.

After restoring BIOS defaults on the HP laptop and re-enabling PXE IPv4 boot, everything started working again. It seems some UEFI/PXE flags were stuck, and resetting the BIOS fixed it.

@Kureebow I’ve had this happen to me before. These are the two cases that I’ve seen:

1 - I captured an image in UEFI and the computer that I was deploying it to was in Legacy mode. I also had the inverse of this, where I captured in Legacy and was trying to deploy to UEFI.
2 (the most common for me) - Some sort of RAID or volume management device is enabled in the UEFI. I only use HP so I don’t know what it would be called on Dell, but on HP there is an option that I have to disable “Configure Storage Controller for VMD”.

@romprager The simple answer is depends on the iso image and what OS will boot from the ISO image.

I do have several how tos in the forum that shows how to net boot several different installers.

https://forums.fogproject.org/topic/10944/using-fog-to-pxe-boot-into-your-favorite-installer-images

@Tom-Elliott Yep, I caught that workaround in the earlier post. I moved the image folder from /images/dev to /images, but still resulted in the same error.

One thing I noticed this morning, is that the image capture process is hanging at 99% on the task page, as in for at least 45 minutes. Not sure if that is a symptom or the cause.

The previous post was suspect of a very recent init version – but i’m on the most recent dev-branch.

Thank you!

Thank you george1421 that 's was the solution.
I was troubleshooting remotely, and the technician on site insisted that images had already been downloaded onto it.

I understand your point of view, but I also have to remain anonymous to my colleagues.

Solved !

@george1421 noted. When I was initially testing, I did not know this and it took some digging in the forums to find this tidbit of knowledge. Given that having separate storage groups stops replication between the nodes in my mind it makes most sense to have a more distributed setup of normal servers at each site to handle all of my requirements. Thanks for the input!

I am confident I have ruled out NFS. I can now connect to the NFS from localhost and Ubuntu to /images & /images/dev without issue and all the permissions that come with it. The only thing failing now is FOG, any guidance would be great!

@rodluz
Problem solved - the disk with the image was damaged

@lmoysidis I’m not using fortigate but check out the following article: https://docs.fogproject.org/en/latest/installation/network-setup/proxy-dhcp/?h=
I needed to dynamically assign a boot file based on vendor class and dnsmasq did the trick!

@Tom-Elliott Thank you so much Tom, as always I appreciate your support and guidance.

I reverted back to a snapshot I had on the same version and it’s working as expected. Unsure how things got corrupted but I would say it’s safe to assume this is working unless others report the same issue.