- Joined
- Sep 10, 2018
- Messages
- 15
- Reaction score
- 1
I have an odd situation that occurs with the following network configuration:
It has happened on multiple installations. My solution every time has been to replace the Netgear switch with a Cisco or HP Enterprise and the issue immediately resolves. Even on a super simple network where the switch has no configuration loaded on it - just a blank default switch. If I swap it, I fix it. Heck, I did a quick field repair once with a cheap desktop switch with no PoE and plugged the phones into power outlets - that has fixed it too.
The only thing I've ever been able to see in the 3CX logs, and only on verbose mode, is what looks to be simple packet loss. (As I don't have any clients having the problem currently, I don't have an example of a log to show.)
However: non-VOIP traffic exhibits no discernible issue. I can pull a throughput test between two computers on the network for 30 minutes at full port saturation and not see a single lost packet.
This leads me to conclude there is some odd setting in the Netgear switch that is messing about with my traffic that I haven't been able to find.
My questions are these:
1) has anyone else out there seen anything similar to this with Netgear switches?
2) if you have seen it, what on the netgear switch is causing this?
I would love to know in case I run into a situation where replacing the switch isn't an option.
- POTS based phone service (from literally any POTS provider - i've tried more than one) or SIP over Fiber (i.e. provided over a dedicated fiber, not through the general Internet)
- Wellgate, Patton or Grandstream FXO Gateway when using POTS. (I've used all three)
- Any model of Yealink Phone. (I've used everything from T19 up to T46 - firmware up-to-date)
- 3CX 15.5+ running on either Windows or Linux, VM or on bare metal, tiny Intel NUC up to a $10,000 server.
- Netgear PoE Switch (GSM7226LP, GSM7248P in particular - firmware up-to-date)
- Configure the switch with some VLAN settings -or- disable every smart setting I can see making it a dumb switch -or- leave it default out of the box - doesn't matter.
- Phones on the AutoVOIP thing, or their own VLAN or on the same VLAN as the rest of the network - doesn't matter.
- One switch on the network. Multiple switches on the network. Doesn't matter.
- Gateways and Phones on "PBX Delivers Audio" or not in any combination - doesn't matter.
It has happened on multiple installations. My solution every time has been to replace the Netgear switch with a Cisco or HP Enterprise and the issue immediately resolves. Even on a super simple network where the switch has no configuration loaded on it - just a blank default switch. If I swap it, I fix it. Heck, I did a quick field repair once with a cheap desktop switch with no PoE and plugged the phones into power outlets - that has fixed it too.
The only thing I've ever been able to see in the 3CX logs, and only on verbose mode, is what looks to be simple packet loss. (As I don't have any clients having the problem currently, I don't have an example of a log to show.)
However: non-VOIP traffic exhibits no discernible issue. I can pull a throughput test between two computers on the network for 30 minutes at full port saturation and not see a single lost packet.
This leads me to conclude there is some odd setting in the Netgear switch that is messing about with my traffic that I haven't been able to find.
My questions are these:
1) has anyone else out there seen anything similar to this with Netgear switches?
2) if you have seen it, what on the netgear switch is causing this?
I would love to know in case I run into a situation where replacing the switch isn't an option.
