V16 Phone system issues - ring group and silent calls

[boxrick]

Hello! I am having a bunch of issues which seem to have occurred at around 12pm over the last two days, the system itself has been running fine for years.

Self hosted: Google Cloud
Phones: SNOM D385
Trunks: Inbound / outbound via two different providers
Issue: When people ring in, the incoming call is silent on the caller side for a long period of time then starts randomly ringing around the call group. 'Ringall' is set to on, and nothing changed when this issue started.
After a number of things, such as rebooting phones, reregistering with call system, reboots, phone reboots, recreating ring groups etc it doesn't seem to help. Then randomly after a period of time it all goes back to normal again.

I am not entirely sure whats going on, but I would like to re-provision another instance running version 18 and re-configure it all to both try and resolve the issue and obviously get us working on version 18. What is the most sensible way of doing this with a completely fresh system? Would it be sensible to export / re-import or would a fresh setup be more sensible?

 

[ChrisC_3CX]

What SIP Provider are you using, is it a 3CX Supported SIP Provider? If yes, which one and, if not, is it a Register based SIP Provider or IP Based?

I think the first thing you should check is if the Call comes into the PBX on time, otherwise, it might be something that is out of the 3CX PBX's control. You could enable verbose logging mode from "Dashboard >> Activity Log >> Settings" and then, make the same call to replicate the issue, and check the activity log to see at what time the SIP INVITE reached the 3CX PBX. Remember that, in the activity log, you can use the "filter" button to filter which call you are looking at and hence make it easier to trace it.

 

[boxrick]

Thanks for the additional logging info, the issue is 'fixed' for the moment so the extra logging will not help at this time. But may be useful when it re-occurs again.

They are Register SIP based ( Numbergroup / voipfone ), I did try my outbound provider as an inbound briefly when the issue was happening and the same problem occurred. So the issue definitely lies with my 3CX system somewhere.

 

[ChrisC_3CX]

Thanks for the additional logging info, the issue is 'fixed' for the moment so the extra logging will not help at this time. But may be useful when it re-occurs again.

In that case, wait when it happens again, set logging mode to verbose, reproduce and check the logs see if you can spot anything and let us know.

 

[boxrick]

I have backed up, provisioned a new system under v18 and restored with the backup. Lets see how it goes...

 

[boxrick]

Seems it is still doing it. It has occurred for the past 3 days at around the same time... Frustrating that it is still doing it with a new provision and V18. I have enabled verbose logging.. but not entirely sure what I am looking for.

 

[ChrisC_3CX]

It sounds like this is something you might need to talk to the SIP Provider about too. In the meantime, you might also want to also consider how the IP Phones are provisioned. If you're using STUN for instance, the issue might be related to if they were configured correctly. If you want to eliminate this possibility, try routing inbound calls to an IVR instead. The IVR should answer the call immediately so if this is not happening either it further suggests that the problem is out of 3CX's control.

 

[boxrick]

So to put things in perspective this this system has been working fine for about 3 years, multiple phones provisioned etc.

This problem started just a few days ago some things I have tried:
- Used different SIP Trunks, this issue seems to affect multiple providers and even happens going directly to an extension rather than a ring group
- I have tried routing the phone calls through a second internet line.
- Diverting from the SIP Trunk directly to a mobile phone resolves the issues right away
- The first time I ring it seem to be a long delay on the inbound call, then it will finally ring but with no audio on either side. Subsequent calls from the same number appear to be ok.
- Outbound calls are taking a long time to initialise.

 

[OlivierD]

We are having the same issue at different customer sites, we noticed that it's only with devices that are registered via direct STUN.
Incoming calls are routed correctly, phones are ringing but they are unable to answer the call.

 

[NickD_3CX]

We are having the same issue at different customer sites, we noticed that it's only with devices that are registered via direct STUN.
Incoming calls are routed correctly, phones are ringing but they are unable to answer the call.

This is a classic STUN symtpom, when the remove firewall has not been configured correctly (at the sites where the phones are located).

 

[NickD_3CX]

So to put things in perspective this this system has been working fine for about 3 years, multiple phones provisioned etc.

This problem started just a few days ago some things I have tried:
- Used different SIP Trunks, this issue seems to affect multiple providers and even happens going directly to an extension rather than a ring group
- I have tried routing the phone calls through a second internet line.
- Diverting from the SIP Trunk directly to a mobile phone resolves the issues right away
- The first time I ring it seem to be a long delay on the inbound call, then it will finally ring but with no audio on either side. Subsequent calls from the same number appear to be ok.
- Outbound calls are taking a long time to initialise.

So just as a test, can you disconnected the snom Phones and have the users use the 3CX WebClient or 3CX Mobile apps for their extensions?
This may be something relating to STUN and the packets not reaching the phones properly, which is very likely if you are using STUN.

Let's do this to rule the phones out.

 

[OlivierD]

This is a classic STUN symtpom, when the remove firewall has not been configured correctly (at the sites where the phones are located).

This was all working properly for quite some time, all customers who are calling in have the exact same issue since today.

 

[OlivierD]

I just tested with a 3CX mobile app and this works, it's only the Yealink desk phones with STUN configuration that are giving problems. Is there a problem with the 3CX stun server?

 

[ChrisC_3CX]

Is there a problem with the 3CX stun server?

IP Phones configured via STUN have nothing to do with the 3CX STUN servers. As per @NickD_3CX , you need to make absolutely sure that the remote firewall has been configured correctly

Alternatively, if you have multiple IP Phones on the same remote site, what I would recommend is deploy a 3CX SBC instead. Using the 3CX SBC will make provisioning a breeze with the use of PnP, up security due to the use of the 3CX Tunnel and also ensure smooth operations of the IP Phones as opposed to when using the STUN provisioning method.

 

[Kristoffel Loos]

@ChrisC_3CX I had several calls of installs/customers with phones provisionned true STUN on V18 and I checked the parameters:



So are you sure the STUN servers aren't the issue, and if not can you tell me exact why.

Br
Kristof

 

[boxrick]

Ok given I have never set up STUN ports, port forwarding this is feeling like the issue. So any reasons why it has worked for so long?

- If I use the SBC method, is there a simple method to re-provision a STUN configured phone to that?

 

[boxrick]

I set up the SBC... wiped a test phone ( using holding # at boot ) and it appears to be working. But with the default template as opposed to custom template.

Given all my template does as add a second line as the shared voicemail line, is this functionality built in to 3CX yet?

 

[boxrick]

Right.... I have managed to re-provision all the handsets and have an SBC set up. Found some method to get my custom templates working and hoping this resolves my problems.

Thanks for the help all. Will update this thread with the outcome.

 

[ChrisC_3CX]

I know you've already sorted this out for the most part but here are the answer your questions:

So are you sure the STUN servers aren't the issue, and if not can you tell me exact why.

The stun servers you see there are actually only used by the PBX. The stun server for an IP Phone configured with the Stun provisioning server is the PBX itself.

Ok given I have never set up STUN ports, port forwarding this is feeling like the issue. So any reasons why it has worked for so long?

This we can't say for sure as firewall behavior varies from vendor to vendor and the state of active connections at any give moment will also most likely affect this.

If I use the SBC method, is there a simple method to re-provision a STUN configured phone to that?

You've already gone through this but, do bear in mind that we always recommend factory resetting an IP phone before changing it's provisioning method. One of the reasons for this is to get rid of any remnant configuration.

Given all my template does as add a second line as the shared voicemail line, is this functionality built in to 3CX yet?

Do you mean a BLF key to check another extension's voicemail? If yes, then, not exactly. If you want to get rid of the custom template and use the default one, what you can do is access another extension's voicemail by calling 999 and then dial #. This will have the Voicemail IVR ask for the extension number for which you want to access voicemail.

Thanks for the help all. Will update this thread with the outcome.

Let us know how it goes!