@Infojoe And a real question : How can I mark this as “solved” ?

@rodluz I tried a manual installation of the updated init.xz as outlined in your instructions, and there was no change. I also upgraded my FOG server and reinstalled your experimental XCP bzImage from earlier in this thread but did not see any change in VM behavior after doing so.

Out of curiosity, @BPSTravis, what hardware are you running XCP-ng on? I know it’s virtualized, but maybe there’s some CPU feature I have disabled in BIOS on my XCP hosts or something like that?

Also, thanks for your patience. I’m only available to work on this once or twice in the middle of the week as I’m part-time with this organization.

@devle I’ve made a couple of edits, though I still cannot replicate the problem you’re seeing.

You shouldn’t see the errors any more (just saying), though I don’t know if these things will fix whatever problem you’re seeing.

@lucamathuse No, I would not move the files manually yet.

If the capture reports “Cloned successfully” but the task remains In-Progress, then the capture itself may have finished, but the post-capture process is not completing correctly. Under normal conditions, the host should finish the task and then reboot or shut down automatically based on your settings.

So, before moving anything by hand, I would want to know whether the host is actually stuck there indefinitely or whether it is still finishing something in the background. If it never updates the task status, that could also explain why your image set ends up incomplete or inconsistent.

@lucamathuse
I don’t remember if it’s related to Win11 we have this in our postdownload script :

if [ -n "$hd" ]; then if [ "$hd" = "/dev/nvme0n1" ] ; then partsys=$hd'p2' partdata=$hd'p3' else partsys=$hd'2' partdata=$hd'3' fi

We use Virtualbox, install classic WIndows 11, disable Bitlocker, and capture with FOG with image set to “Multiple Partition Image - All Disks”
My folder capture :

d1.mbr d1p1.img d1p2.img d1p3.img d1p4.img d1.partitions d1.size

Maybe try capture debug to see which partition FOG get during your capture

@Tom-Elliott
I did not update yet to 1824, but i did SELECT * FROM HISTORY;

MariaDB [fog]> SELECT * FROM history; +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ | hID | hText | hUser | hTime | hIP | +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ | 1 | [2026-04-01 08:33:55] MulticastSession ID: 1 NAME: test has been successfully updated. | fog | 2026-04-01 08:33:55 | 10.10.3.116 | +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ 1 row in set (0.000 sec)

the new session i create are not shown, and dont work until i run the maintenance queries

then it shows in history table:

MariaDB [fog]> SELECT * FROM history; +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ | hID | hText | hUser | hTime | hIP | +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ | 1 | [2026-04-01 08:33:55] MulticastSession ID: 1 NAME: test has been successfully updated. | fog | 2026-04-01 08:33:55 | 10.10.3.116 | | 2 | [2026-04-01 09:10:37] MulticastSession ID: 1 NAME: test has been successfully updated. | fog | 2026-04-01 09:10:37 | 10.10.3.116 | +-----+----------------------------------------------------------------------------------------+-------+---------------------+-------------+ 2 rows in set (0.000 sec)

@hariskar

I didn’t really catch the described use case…
But anyway, if application stores something in the database, then the database need to be backed up, of course. Fully and incrementally also (separate logic).

If you use application that stores data in the same machine where it is installed, better would be to configure it to store data on the remote database, e.g on specific database server or some other way.
You can create network drive or other remote storage and point your database files there - if needed to be separate (for each PC).

This way you have all the data in the same place, and you can make backups of this server or drive more easilly.

Backuping entire PC with database in it - this can be done, but why?
It feels strange.

Example:
Lets say, employees in your company are working with important documents.
Do you really want to backup of entire PC (with os and applications and data), instead of storing those documents on secure and reliable remote storage?
Better solution is to simply make network location (local or in cloud) and let the users to store documents there - so the documents would not be lost if some pc dies.
If PC dies, then you simply replace the pc and deploy ready to use system image.

@Tom-Elliott Hi, has there been any development with this?

I’m facing a situation that’s not quite identical but very similar in my 1.6.0-beta.2297.

I already posted to the Forum with the details I thought were relevant, but noticed my FOG status is showing the same messages that are posted here.

@BLT I don’t necessarily have a timeline for the final release but I did just release an experimental kernel with a newer nic driver. It may work for you if you want to give it a try.

@Exsival When you say “Causing the fog server to crash” what exactly do you mean?

If it’s just a “saying” I can understand that but from a technical standpoint what this means is completely different and would need logs. If all you mean is the machine in quesiton is hanging, that’s a different issue altogether. I highly doubt the whole FOG Server is crashing.

@kratkale said in Windows 11 | 65x HP Z2 Tower G1i:

Can I specify in FOG which hard drive to clone to? For example, using the serial number? That could be stored in the host entry

Hi, this is a known issue when using NVMe, as you’ve probably already discovered. I’m not sure whether FOG 1.5 supports using serial numbers to target disks, but someone from the team may be able to confirm that. FOG 1.6, however, does support targeting disks by serial number.

I ran into the same situation. As a workaround, I customized my init.xz to use the PCI path instead, since those appear to be more persistent.

@Infojoe I forgot an important information : version 1.5.10.1826

@toalalife Hi there! Sorry for the late reply. I’ve been on holiday and I forgot to check up on this. That is interesting. There are a couple of things I can think of to double check/try off the top of my head.

That particular error happens when iPXE can’t execute the binary, usually because either an architecture mismatch (e.g arm64 on x64) or because secboot fails to verify. Given that disabling secboot fixes it, I’m leaning towards that. (https://ipxe.org/err/2e0080)

So I would say you should double check that your kernel is signed. If you’ve updated them you’ll have to resign the kernel to ensure it keeps working.

The other would be to double check that the shim command is being invoked at some point prior to boot.php being chained. There’s a none zero chance that if you’ve updated FOG, it may have overwritten the modified default.ipxe

Other than that, if you could try and record the boot process I’d be happy to take a look and see if I can spot anything out of the ordinary, I’m also happy to take a look at your kernel or any ipxe scripts etc if you want me to double check if they’re signed or bootable.

As a final note, I don’t think I see iPXE loading the initrd.xz file there, which contains the ram filesystem that FOG uses on boot. I could be misremembering the boot process (I can’t recall if it’s normal for it to not do that if the bzImage fails to verify, or if it loads it prior to bzImage), but if that’s failing to load it might also be worth checking that out, though it shouldn’t have to be signed!

@lucasgfaj have you turned off secure boot?

@trent_potter Yes you figured it out. Sysprep is a Windows utility, it has nothing to do with FOG. It is always recommended to sysprep before capturing an image.

@Tom-Elliott All good the last update fixed the issue!

My workarounds for update on Debian Trixie are:

A)
git checkout master ; git pull
userdel fogproject
cd bin ; ./installfog.sh

or
B)
apt install lastlog2
git checkout master ; git pull
ln -s /usr/bin/lastlog2 /usr/bin/lastlog
cd bin ; ./installfog.sh

Peter