Crashing every day after Update 8

[dkawahig]

Hello,

I can't keep the pbx up after it updated to v20 update 8. The whole machine seems to go awol - I cannot connect via the management console or via SSH to the OS. The only think I can do is reboot via the AWS Lightsail console.

Very difficult to troubleshoot.

The one think I see in the syslog file is this:
2026-02-17T08:21:15.015261-06:00 ip-172-26-5-92 3CXMediaServer: assertion failed: /Sources/3rdParty/resiprocate/resip/stack/TransactionState.cxx:781: !sip || !(state->isFromTU(sip) && sip->isRequest() && method == ACK)

Didn't see these prior to the upgrade to update 8. Not sure if it related to the system going brain dead. It doesn't seem to stay up for more than an hour after reboot.

It is running on a small AWS lightsail instance with ~50 users.

Any help is appreciated. May need to roll it back to a previous snapshot.

Don K.

 

[Alphabetic]

Can you confirm system spec?

are you able to monitor performance at a hosting level?

do you have email alerts on (System > Alerts)

are you running out of disk space?

From the dashboard click on "overview" and check your disk space.

 

[dkawahig]

Here is the system info:


[General]
ProcessorCount = 2
TotalDiskSpace = 78 GB
FreeDiskSpace = 60 GB
TotalPhysicalMemory = 3.765 GB
FreePhysicalMemory = 2.562 GB
OSArchitecture = X64
OSDescription = Debian GNU/Linux 12 (bookworm)
ProcessArchitecture = X64
FrameworkDescription = .NET 8.0.21

[3CX PhoneSystem Info]
Number of Extensions = 44
Number of Queues = 0
Number of RingGroups = 10
Number of IVRs = 4
MULTICOMPANYMODE = 0
ALLOWUSEBUSYOPTFORGROUP = 0
FIREWALL_CHECKER_RESULT = 1
SYSTEMLOGGINGLEVEL = 2
VERSION = 20.0.8.1109
RESTORED_FROM_VERSION =
ADJUST_FROM_VERSION = 20.0.8.1109

 

[Charles_3CX]

Regarding your question and the issue you mentioned, if you work with a partner, please contact them to open a ticket with us at 3CX Technical Support so we can check this. If you self-manage, you can open a ticket directly in the systems admin console so we can check this for you.

 

[dkawahig]

I think I figured it out ... networking.service was failing, thus when the dhcp lease expired, it would not acquire a new address. On Lightsail the interface lease is in the 20-60 minute range.

Apparently the update added a number of ipv6 directives in the /etc/network/interfaces file that caused the issue, as ipv6 is shut off on the instance level. It looked like dhclient was not initializing. I have removed the ipv6 entries, and rebooted. It has stayed up longer than it has before.

So ... hopefully this is the issue!

Don K.