Problem with node update

V

vikozo

Renowned Member
May 4, 2014
821
34
93
suisse
www.wombat.ch
Good afternoon,
I am aware that I am behind with upgrades.
Now I wanted to update one node in order to initiate an upgrade across all three nodes.

One node returns the error message ‘update not possible; use dpkg --configure -a’.
When I do this, the console freezes after a while and I can no longer see how to proceed.

Even when accessing and starting remotely, the connection is cut after a while and I am disconnected from the nodes.


Code: 
root@pve-04:~# dpkg --configure -a
Setting up pve-kernel-5.15.107-2-pve (5.15.107-2) ...
Examining /etc/kernel/postinst.d.
run-parts: executing /etc/kernel/postinst.d/apt-auto-removal 5.15.107-2-pve /boot/vmlinuz-5.15.107-2-pve
run-parts: executing /etc/kernel/postinst.d/initramfs-tools 5.15.107-2-pve /boot/vmlinuz-5.15.107-2-pve
update-initramfs: Generating /boot/initrd.img-5.15.107-2-pve
Running hook script 'zz-proxmox-boot'..
Re-executing '/etc/kernel/postinst.d/zz-proxmox-boot' in new private mount namespace..
No /etc/kernel/cmdline found - falling back to /proc/cmdline
client_loop: send disconnect: Broken pipe

Which log should I look at to see where it stopped working?
To provide you with further information so that I can complete the update?

have a nice day
vinc
 
Last edited:
that sounds like a previous upgrade got interrupted.

please try connection with SSH and using a tool like screen or tmux to have a persistent session to run the upgrade in. then post the full output here.
 
when you say disconnect, what does that mean? does the node crash? is network interrupted? is apt/.. killed?
 
I lose the SSH connection to the node, then I have to log in again, and when I go to /var/log/, for example, it freezes and I lose the connection again. some times it is shorter sometimes longer to be on the. The same if go in the GUI and there the node und shell same problem.
 
please post the full system log covering that time period, and any errors printed by SSH or apt
 
the SSH part is above

aus dkkg.log

2025-09-11 19:53:57 startup packages configure
2025-09-11 19:53:57 configure pve-kernel-5.15.107-2-pve:amd64 5.15.107-2 <none>
2025-09-11 19:53:57 status half-configured pve-kernel-5.15.107-2-pve:amd64 5.15.107-2
2025-09-11 19:54:20 configure pve-kernel-5.15.136-1-pve:amd64 5.15.136-1 <none>
2025-09-11 19:54:20 status half-configured pve-kernel-5.15.136-1-pve:amd64 5.15.136-1
2025-09-11 19:54:39 configure pve-kernel-5.15.116-1-pve:amd64 5.15.116-1 <none>
2025-09-11 19:54:39 status half-configured pve-kernel-5.15.116-1-pve:amd64 5.15.116-1
2025-09-11 19:55:00 configure zfs-zed:amd64 2.1.15-pve1 <none>
2025-09-11 19:55:00 status unpacked zfs-zed:amd64 2.1.15-pve1
2025-09-11 20:09:58 configure pve-kernel-5.15.158-2-pve:amd64 5.15.158-2 <none>
2025-09-11 20:09:58 status half-configured pve-kernel-5.15.158-2-pve:amd64 5.15.158-2

and a bit later the monitor freeze (with SSH Connected, but the same in the GUI shell)
which other log would be use full too?
 
Last edited:
As @fabian suggested, you should use tmux or screen in your ssh session.
In case you get disconnected, the dpkg process will continue und you can re-attach to the running tmux/screen session after re-connecting through ssh.

1. ssh into your node
2. start tmux
3. run you dpkg command

if you get disconnected:
1. ssh into your node
2. re-attach to tmux "tmux attach"

That way you should be able to examine the error messages and/or continue the update.
 
hallo
heute nochmals ein # dpkg --configure -a gestartet und wollte die log files durchsuchen.
heute kein freeze der Monitore und es lief durch.

dabei ist mir der fehler aufgefallen

Code: 
Copying and configuring kernels on /dev/disk/by-uuid/FB5D-B209
        Copying kernel 5.15.107-1-pve
        Copying kernel 5.15.136-1-pve
cp: error writing '/var/tmp/espmounts/FB5D-B209/vmlinuz-5.15.136-1-pve': No space left on device
run-parts: /etc/initramfs/post-update.d//proxmox-boot-sync exited with return code 1

woher kommt dieses no space left?
Code: 
# df -h
Filesystem        Size  Used Avail Use% Mounted on
udev               32G     0   32G   0% /dev
tmpfs             6.3G  1.9M  6.3G   1% /run
rpool/ROOT/pve-1  8.5T  2.4T  6.1T  29% /
tmpfs              32G   63M   32G   1% /dev/shm
tmpfs             5.0M     0  5.0M   0% /run/lock
pve-04-ssdPool    796G  128K  796G   1% /pve-04-ssdPool
rpool             6.1T  256K  6.1T   1% /rpool
rpool/ROOT        6.1T  256K  6.1T   1% /rpool/ROOT
rpool/data        6.1T  256K  6.1T   1% /rpool/data
/dev/fuse         128M   76K  128M   1% /etc/pve
tmpfs             6.3G     0  6.3G   0% /run/user/0


Es ist mir immer noch nicht klar weshalb die Updates und dpkg --configure -a bisher nie durchgelaufen sind???

ABER
heute
dpkg --configure -a lief durch und dann über das GUI auch das ganze upgrade ohne weitere Probleme!

Somit wäre das Problem gelöst weshalb und wie auch immer!

gruss
Vinc
 
Hallo
kann mir jemand erklären was diese Fehlermeldung bedeutet?!

cp: error writing '/var/tmp/espmounts/FB5D-B209/vmlinuz-5.15.136-1-pve': No space left on device

und wird mir das bei einem Upgrade auch Fehler machen?
 
Last edited:
das bedeutet dass auf deiner EFI System Partition nicht genug platz fuer alle installierten kernel ist..
 
die ESP wird nur waehrend kernel oder bootloader updates gemounted, wenn proxmox-boot-tool verwendet wird. daher scheint sie in so einem setup auch nicht in `df -h` auf.
 
poste mal "proxmox-backup-tool status". normalerweise reicht es aus, alte kernel pakete zu entfernen. wenn das nicht funktioniert, musst du eventuell haendisch aufraeumen (/dev/disk/by-uuid/FB5D-B209 wohin mounten, dann aufraeumen, dann unmounten)
 
You must log in or register to reply here.
Top Bottom