We run a 100down/100up fiber connection with link failover to a copper cable network. Working with RC has produced few positive results based on what we have found. RC is not the issue, but apparently, my internal network is. We run two HP 2530-24G switches with full data passthrough that should not be causing any QOS issues.
What we have completed already;
- Used PingPlotter to record quality issues on client computer using 18.104.22.168 as the target.
- Ensured all UDP traffic going to 22.214.171.124/22 and 126.96.36.199/22 have a minimum 450 Kbps.
- Set all traffic with DSCP 101110
- Ensured sip-helper, sip-nat-trace, and RTP processing are disabled in the FortiGate
- Ensured the SIP entry in session-helper is disabled (not present at all)
- Ensured all traffic is flowing without issue during and not during phone calls.
So the question stands; should traffic on computers be somehow separated into different VLANs to account for the RC quality issues we have been experiencing?