SBC Update from Debian 12 to 13 Not Detecting Current OS Version

Gavino

Customer
Joined
Jul 24, 2021
Messages
41
Reaction score
16
EDIT:

I shut down the VM, took another VM snapshot, and booted up again.
After boot up I ran it from my own bash env with just "sudo ./sbcdebian12to13.sh" rather than "sudo bash sbcdebian12to13.sh" and hey presto - it worked! I can't say if it was the reboot or calling the script directly but glad it completed OK. Something for the devs to check out.

I'll be doing a test install on a HA pair of my lab Raspberry Pi 4B 8GB devices sometime this week and then probably cutover the production Raspberry Pi SBCs with fresh installs starting on the weekend.


-----------------------------------

I am very excited to see support for Debian 13 for the SBCs - thank you devs! There is a long-standing issue when running HA setups and switches do firmware updates - the SBCs can get themselves locked up. I have been working around that by putting SBCs in maintenance mode "crm configure property maintenance-mode=true" then stopping pacemaker, corosync and 3cxsbc before network switch firmware updates. I am really hoping that Debian 13 fixes these lockups (as I have read about in other Linux forums) so am super keen to try this upgrade.

Anyway not off to the best start. I have a 3CX SBC set up on Proxmox amd64 install where I did everything standard and there's no SBC HA (only Proxmox HA) so should be fairly simple. I followed the guide here: https://www.3cx.com/blog/releases/sbc-beta-debian-13/

As per guide I downloaded the upgrade script, set it as executable and "sudo bash sbcdebian12to13.sh" ran it. Copying just the yellow log items.

Bash:
====== [21:31:33] Starting: Backup script /home/techsupport/sbcdebian12to13.sh
====== [21:31:33] Starting: Backup script version: e34b931681dc1326882b7f0c8963d84d793b8b88
====== [21:31:33] Starting: Current directory: /tmp
====== [21:31:33] Starting: Current user: root
====== [21:31:33] Starting: Current date: Mon 10 Aug 2026 21:31:33 AEST
====== [21:31:33] Starting: Current version:
====== [21:31:33] Starting: Architecture: amd64
====== [21:31:33] Starting: Hostname: GT3CXSBC01
====== [21:31:33] Prepare: Installing jq...
====== [21:31:35] Prepare: Installing curl...

====== [21:31:37] Prepare: Installing apt-transport-https...

====== [21:31:39] Error: Unsupported OS detected. This upgrade script can only be run on Debian 12 (Bookworm). Current OS version: Debian . Exiting..

But why?

Bash:
techsupport@GT3CXSBC01:~$ uname -a
Linux GT3CXSBC01 6.1.0-52-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.180-1 (2026-08-03) x86_64 GNU/Linux
techsupport@GT3CXSBC01:~$ cat /etc/debian_version
12.15

Strange.
 
Last edited:
Hi @Gavino the line ====== [21:31:33] Starting: Current version: appears completely empty. Because of this, the operating system variable is left blank (Debian .) and the script aborts the installation.

is lsb-release installed?

Keep in mind that the sbcdebian12to13.sh script only works on installations performed from the official 3CX ISO. If by chance you installed a generic Debian on the instance and then added the SBC package manually, the script's system validation check will fail.

If this was a manual installation on top of standard Debian and not from the official 3CX ISO, you will need to deploy a fresh machine using the 3CX ISO.
 
  • Like
Reactions: NikosT_3CX
SBC install was from the official 3CX ISO. Not sure if you caught my edit after the original post, but I called the script with "sudo ./sbcdebian12to13.sh" and that worked, when "sudo bash sbcdebian12to13.sh" failed. I did reboot after the failed attempt - not sure if that mattered. I had shut down the SBC just prior to the initial upgrade attempt, so that I could grab a powered-off snapshot.

lsb-release - I am confident that it would have been installed prior to the initial upgrade attempt, as I did a standard 3CX ISO installation without modifications.

Anyway, hopefully it helps others that might have the same issue. My suggestion would be to reboot after the first failed attempt and try that. If it fails again then call the script directly, so long as you run from a native bash shell, rather than calling bash from bash.

I've been running on the upgraded Debian 13 SBC for the last whole business day and no issues to report, so I'll upgrade the remaining ones - probably on the weekend.
 
  • Like
Reactions: Alejandro_3CX