Multiple SBC Status Change Notifications

[Tbulgrin]

First, I need to know why it is that when I search terms associated with my SBC status questions, i get directed to this web page:

Second, I keep getting multiple email notifications that my SBC status has changed. Specifically, "System wide: Trunk SBC 'SBC880704' (xx.xx.xx.xx/xx.xx.xx.xx) has been changed". I have searched these forums for answers; and there appears to be little or nothing written about it. There are acknowledgements that it "does happen". However, there is nothing leading to any resolution. Please assist!
If I have posted in the wrong thread/forum, please excuse the intrusion. I am new here. Thanks!

 

[NicholasP_3CX]

This would indicate that your SBC is losing access to the PBX. You will receive this notification about a minute after the change of status. You will need to see on your network where the SBC is, why it is losing the connection.

The SBC normally should not lose connection, but if it is happening regularly, you may bneed to investigate why.

 

[Tbulgrin]

This would indicate that your SBC is losing access to the PBX. You will receive this notification about a minute after the change of status. You will need to see on your network where the SBC is, why it is losing the connection.

The SBC normally should not lose connection, but if it is happening regularly, you may bneed to investigate why.

Thank you. I'm not finding anything of note in my logs. I was actually hoping that someone on here had experienced this issue. Perhaps they could offer some insight as to "why" these interruptions might occur. My ISP acknowledges some service interruptions; but they, of course, say the problem is not on their side.

 

[NicholasP_3CX]

This might not be your ISP per se, but their upstream provider. There is also the chance that there might be some sort of filtering on your network that might cut a connection, like IDS/IPS for example.

It would require a good look at the network.

This might be as simple as interference on a cable in the network, where someone fires up the microwave to heat up their lunch and this causes a disruption.

A whole can of worms waiting to be opened when troubleshooting a network.

 

[rayfield]

I've seen this happening, with some systems (not all), which are on different ISPs and even in different geographical areas (100 miles apart, for example. I never saw this happening with 3CX v18. It started when I updated systems to 3CX v20.

I started to troubleshoot this problem, but I got tied up with a lot of other work that's higher priority. So this problem had to go on the back burner for a while. I'll be getting back to it, maybe next week.

John

 

[SteveITS]

Specifically, "System wide: Trunk SBC 'SBC880704' (xx.xx.xx.xx/xx.xx.xx.xx) has been changed".

Up or down? Or both? Typically "up" happens at every single reconnection, and "down" is sent after a few minutes. Check for packet loss, even 1% could mean sporadic drops.

 

[NicholasP_3CX]

I've seen this happening, with some systems (not all), which are on different ISPs and even in different geographical areas (100 miles apart, for example. I never saw this happening with 3CX v18. It started when I updated systems to 3CX v20.

I started to troubleshoot this problem, but I got tied up with a lot of other work that's higher priority. So this problem had to go on the back burner for a while. I'll be getting back to it, maybe next week.

John

We now send the notifications 1 minute after the drop. It used to be 5 minutes. This could be the cause of the increase. Previously the drops would be reported less frequently.

 

[rayfield]

We now send the notifications 1 minute after the drop. It used to be 5 minutes. This could be the cause of the increase. Previously the drops would be reported less frequently.

How long does the drop in TCP connectivity have to be before it triggers a notification?

 

[NicholasP_3CX]

In update 5, anything below 1 minutes will not trigger any notification. Neither down, nor up.

Above 1 minute, the down notification will be sent and the UP notification will be sent when the status changes to up.

In prior updates, including 4, it is 5 mins.

This is assuming you have the notification for SBC/Trunk notifications active.

 

[Tbulgrin]

Thanks to all who have responded. It's been a minute; but I wanted to update my thread. I seem to have tracked the issue to a potential bandwidth problem. These "down/up" errors seem to occur only when i am running cloud backups on some of my in-house file shares. At the risk of inviting much derision, are there any thoughts in this area?

 

[NicholasP_3CX]

3CX does implement QoS on the SBCs, so you might need to configure your firewall to honour this and pass the voice traffic first

https://www.3cx.com/blog/news/sbc-eol-upgrade/

 

[SteveITS]

I'd even make the backups a lower priority which will also assist with normal-priority traffic.

 

[Mooney]

I am commenting as we too are experiencing this issue. Our SBC's are losing connectivity to OVH. Sometimes 20x a day only for a moment and coming straight back online. This is impacting around 10% of our base, and the one common they share is the ISP. They are all BT Wholesale circuits, with a mix of FTTP/FTTC/EAD. They all have different routers/firewalls, but when we check the event log on each instance they are all dropping SBC connectivity at the same time. One endpoint actually has two locations, one with a TTB connection and the other with a BTW connection. Only the BTW SBC drops at the same moment as the other instances on BTW. This issue is not replicated on any other ISP. Getting BTW or OVH to investigate at all is proving really difficult. Has anyone else experienced anything similar?

 

[SteveITS]

We've seen cases where Comcast has packet loss or routing problems to our data center but no other ISPs do. I've even seen cases where a backbone provider has a problem and it's neither end. That in particular can be difficult. You could try using pathping or mtr but would presumably need to catch it happening.