Fog 1.5.10 capture images stops at "saving partition tables (MBR)......"

[robert70569]

I installed on a fresh debian 11 the new fog version 1.5.10 and get an hangup while capturing with “saving partition tables (MBR)…” . I tried several clients, also clients (legacy BIOS) which work fine with fog 1.5.7. I also tried all Kernel Version (Version 4.x.x ends with kernel panic). The fog server is running as LXC on proxmox 8.0.4 with loaded module nfsd and the lxc.conf is set to “lxc.apparmor.profile: unconfined” What else can I try?

[robert70569]

@Sebastian-Roth I have to add, that I tried to use the FOG-Server in two VLANS (with 2 interfaces) and I didn´t believe that FOG is not running with two interfaces. This is probably the reason for “saving partition tables…” I changed the configuration to only 1 interface/network and capturing works perfectly and no errors occur. So my problem is “solved” in a way.

[Sebastian Roth]

@robert70569 How long have you been waiting for it to proceed at this stage?

[robert70569]

@Sebastian-Roth It never proceeded. I Waited from 30 min. till 60 min. before I shutdown the clients hard.

[Sebastian Roth]

@robert70569 Do you have clients that proceed beyond that point? As far as I can remember we did not have reports about this since 1.5.10 was released.

To further debug this issue I suggest you schedule a capture in debug mode, then boot up the machine, set a root password (command passwd) when you get to the shell, check the IP it got from the DHCP (command ip a s) and make sure you can connect via SSH. Then start the capture (command fog) and step through the process. When it hangs you wanna try the SSH connection again. See if you can login and see what it is working on (commands top and ps ax).

[robert70569]

@Sebastian-Roth I have to add, that I tried to use the FOG-Server in two VLANS (with 2 interfaces) and I didn´t believe that FOG is not running with two interfaces. This is probably the reason for “saving partition tables…” I changed the configuration to only 1 interface/network and capturing works perfectly and no errors occur. So my problem is “solved” in a way.

[Sebastian Roth]

@robert70569 This is interesting. I think it would fail earlier when trying to NFS mount the images share on the FOG server in that case. Strange.