UCS does not start properly after Proxmox backup

UCS - Univention Corporate Server
, ,
#1

Until recently, we had a UCS4.4 in use here as a VM in Proxmox.
Backups were carried out in Proxmox via stop mode every evening. This worked quite well until now.

Now we have upgraded the UCS system to 5.0-3.
The backup was kept the same as before. Now it looks like the VM does not start properly after the backup.
Today, for example, the service BIND9 was not running. After manually starting the service in the UMC, everything ran properly again.

If I shut down the VM, then manually start the backup in Proxmox and restart the VM after the backup, everything seems to run.

Has anyone ever seen this kind of behaviour?

Reiner

#2

I have done some more tests. It looks like the UCS VM is going to stand still for a very long time…

  Booting `Univention Corporate Server GNU/Linux'
Loading Linux 4.19.0-24-amd64 ...
Loading initial ramdisk ...

I started the backup at 12:40. Then the UCS is stopped for a short time, a snapshot is created and the UCS is started again.

According to syslog 20 minutes pass between server stop and restart. At 13:00 messages come slowly again into the syslog.

May 27 12:40:01 ucs-server blkmapd[261]: exit on signal(15)
May 27 12:40:01 ucs-server systemd[1]: Stopping Availability of block devices...
May 27 12:40:01 ucs-server systemd[1]: Stopped target RPC Port Mapper.
May 27 12:40:01 ucs-server systemd[1]: Stopped target Graphical Interface.
May 27 12:40:01 ucs-server systemd[1]: Stopping Disk Manager...
May 27 12:40:01 ucs-server systemd[1]: Stopping pNFS block layout mapping daemon...
May 27 12:40:01 ucs-server systemd[1]: Stopped target Multi-User System.
May 27 12:40:01 ucs-server systemd-memcached-wrapper[592]: Signal handled: Terminated.
May 27 13:00:48 ucs-server kernel: [    0.000000] Linux version 4.19.0-24-amd64 (debian-kernel@lists.debian.org) (gcc version 8.3.0 (Debian 8.3.0-6)) #1 SMP Debian 4.19.282-1 (2023-04-29)
May 27 13:00:48 ucs-server kernel: [    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.19.0-24-amd64 root=UUID=77ed3c80-18ff-425f-a390-736dd57249c1 ro quiet loglevel=0 rootdelay=5 sp$
May 27 13:00:48 ucs-server kernel: [    0.000000] x86/fpu: x87 FPU will use FXSAVE
May 27 13:00:48 ucs-server systemd[1]: Started Create list of required static device nodes for the current kernel.
May 27 13:00:48 ucs-server kernel: [    0.000000] BIOS-provided physical RAM map:
May 27 13:00:48 ucs-server kernel: [    0.000000] BIOS-e820: [mem 0x0000000000000000-0x000000000009fbff] usable
May 27 13:00:48 ucs-server kernel: [    0.000000] BIOS-e820: [mem 0x000000000009fc00-0x000000000009ffff] reserved

The backup slows down the system. I am aware of that. But can it be that something runs into the timeout. During this attempt BIND9 and the DHCP server was not started.

Reiner

Mastodon