I have a M4100 currently as configured as a core switch;
The overall issue which is occurring is packets are being discarded (as seen under Monitoring --> port detailed stats -> Receive Packets Discarded) when phones ring "After Hours" I can easily replicate the issue;
* During the working day only 10 phones ring in --> the receptionists then divert the call to the required person. - Works ok
Issue Exists and can be replicated here:
Currently the request is to have 52 phones all ring at the same time after 5:00PM --> This is when packets are dropped and is causing all sorts of things to occur with the phones.
Things Completed;
* I have confirmed the issue is not with the firewall or the WAN links - Packet captures show the traffic and packets entering through the firewall and forwarding onto the engress port of the switch.
* I have disabled STP for the port
* I have attempted to set a "Strict" Scheduler Type inter CoS Queue Configuration --> The packet loss / discard rate is a litte bit less, but it is still giving issues.
* I have given auto VoIP Traffic a classification of 6
* I have placed the "voip ingress" port to be of a higher priority of any port
* I have attempted to place all VOIP traffic through a LAG with no difference
* During the time the VoIP VLAN was part of the LAG I raised the LAG to be of a high priority.
* I have setup a syslog service to temporarily monitor this to try and get more information, when I can replicate the issue I get the following alerts;
03-15-2019 08:25:12 User.Debug 192.168.15.20 Jan 3 11:40:03 192.168.15.20-1 VOIP[50795004]: voip_util.c(2106) 306 %% Ageing timerTimer started for interface lag 1. 03-15-2019 08:30:51 User.Debug
192.168.15.20 Jan 3 11:45:42 192.168.15.20-1 VOIP[51243548]: voip_sip.c(290) 307 %% Removed SIP call id = 7737f91c-fc925689@192.168.11.142 Source IP = 192.168.11.142, Dest IP = 0.0.0.0 Source Port = 16498, Dest Port = 0
03-15-2019 08:35:39 User.Debug 192.168.15.20 Jan 3 11:50:30 192.168.15.20-1 VOIP[51243548]: voip_sip.c(290) 326 %% Removed SIP call id = 7875357b2908f6060dbdc9791a27d578@x.x.x.x:5060 Source IP = xxx.xxx.xxx.xxx, Dest IP = 192.168.11.142 Source Port = 11062, Dest Port = 16502
The issues end once the general working time starts and phones flick back and DID only rings 10 phones at once. My question here is
* Can the M4100 handle the traffic load of 52 phones calling at once
* How can I tell if the packets are being discarded due to queueing or buffering issues - If the current queue buffer is shared between all ports, would It benefit me to setup a LAG with more then 2 ethernet nodes and would the burst of packets be split between each port in the LAG's queue? - Previously only setup LAG with 2x ports.
* If after troubleshooting the findings are that the M4100 is not capable and the ports are being flooded, how can I determine what model of switch would be able to handle the packet load?
I found out that you have already opened a support ticket (case # 41076306) with NETGEAR Support about your concern described here on your initial post. We will appreciate if you would post updates here about the progress of your support ticket.
