Management Console service randomly crashes (Debian 9.13, 3CX 16.0.9)

[Allinco]

Hi 3CX Community,

We've got a minor issue with a newly deployed 3CX installation on Debian 9.13, the management service (3CXPhoneSystemMC01.service) randomly crashes after being up for a while.
The customer has no impact, but whenever we need to change something we have to start the service again.
We are running 3CX version 16.0.9 (Professional Perpetual).

When checking the status we get the following:
systemctl status 3CXPhoneSystemMC01.service

I have censored the hostname of the unit due to it containing the name of an end-customer.
Our Linux experience is not exactly that great, is there anyone that has seen something similiar? Google was not exactly that helpful.

 

[Saqqara]

How did you install 3CX, did you use the 3CX Debian ISO image ?

Info here on installing the debian version - https://www.3cx.com/docs/manual/installing-debian-linux-pbx/

 

[NickD_3CX]

Very important, first what @Saqqara asked.

Beyond that, do you have to manually restart the service after this happens, or e.g. you just get an email that the service stopped?

For example, in the screenshot I see that a few minute prior, the system checked for Debian updates, so if it found something outdated, the Linux machine might have restarted because the installation of a package may have required it.

You said it happens often, when you do the same command, do you always see it stopping after it checks the pinning file?

 

[Allinco]

Thank you @Saqqara and @NickD_3CX for your replies.
I have just checked the install and it is an official install, the official debian-amd64-netinst-3cx.iso was used for installation.

We do not get an email message saying the service stopped, we detect it by getting an error when logging into the management console. Then we have to manually start the service, not restart as it is stopped.

It happens completely randomly. Sometimes it takes an hour, at other times it is a day. With fresh install I meant it is running for about a month now, and the issue has happened since the installation.

It's always the same service failed message(s), just different timestamps.

 

[NickD_3CX]

How much RAM does the system have? It is quite common, especially on Deb9, if the OS starts running our of RAM, the OS OOM Killer may stop the Management Console service.

 

[Allinco]

It's a VM with 2 CPU cores and 4GB RAM on ESXi 7.0.0, it's above the minimum requirements.

 

[NickD_3CX]

If you access the Debian via terminal, what files do you see in /var/lib/3cxpbx/ ?
Could you send me a screenshot?

 

[Allinco]

Hi Nick,
I just checked the directory, it contains the following:


My further replies might be delayed due to our business hours ending soon. But I will pick this back up tomorrow.
Thank you kindly for your assistance and time so far.

 

[NickD_3CX]

Thanks, OK, please check the creation date/time of the "core" file and see if it matches approximately the last time the Management Console service stopped/crashed.

If it does let me know and I'll send you a PM with some things I'll need to check.

 

[Allinco]

After checking the core file I indeed noticed it last changed when the service crashed for the last time. The timestamps might be a while back but we left the service in failed state for a while due to not needing to log into the management console.

 

[NickD_3CX]

Il'l send you a PM to ask you for some information I'll need to investigate, hopefully the logs haven't rolled over.