Intermittent hard hangs on Debian 12 / KVM VPS since early November (PBX 20.0.10.1785)

pws

Free User
Joined
Sep 9, 2026
Messages
3
Reaction score
0
Our hosted 3CX PBX hangs hard, with no shutdown in the logs, and needs a manual restart.

Environment

3CX PBX 20.0.10.1785 (updated from 20.0.9.995 on 11-12) and 3cxai 1.6.91 (updated 11-17)
Debian 12.15, kernel 6.1.0-55-amd64 (6.1.192-1). 6.1.0-54 is also installed.
KVM guest (QEMU on a RHEL 9.6-based host, q35, virtio), 1 vCPU, 8 GB RAM
Resources are fine: CPU under 1%, RAM about 29%, disk 14%
Timeline

The first offline alert I have is 11-03. There were more on 11-06, 11-14, 11-17 (twice) and 11-18 (twice).
Before 11-16 the kernel was 6.1.0-54 or older. 3CX's unattended-upgrades installed 6.1.0-55 on 11-16 at 03:44 CST.
Hangs since the kernel change: 11-17 about 00:04 and 08:30 CST, and 11-18 about 00:00 and 12:37 CST.
The journal ends abruptly each time, with no OOM or lockup messages. One boot ended with a kernel BUG (NULL pointer dereference) on 6.1.0-55.
3CX update runs happen at about 03:00–04:10 CST and cause clean reboots. None of the hangs lines up with an update run.
Questions

Has anyone seen hangs like this on 6.1.0-55 under KVM, or on 3CX 20.0.10?
Is there a kernel version 3CX recommends or tests against?
Can the OS-update setting be limited so the kernel isn't changed automatically?
 
Please see existing threads. Debian released a kernel with a bug (.55), if that got installed you need to either downgrade to the 3CX provided .54 or move to kernel .56 .
 

Latest Posts

Forum statistics

Threads
112,542
Messages
593,112
Members
165,751
Latest member
Teresa Nguyen