Hey everyone. I am having some really weird issues with my Metro Ethernet sites and Shoretel. Please note I am not the greatest at this as I have never had to deal with VoIP issues at my last job. But here is the breakdown and what I have done.
Our company strictly runs HP ProCurve Routers/Switches and a few older 4500/4510g 3com switches. We have 4 locations using Metro Ethernet 100mbps links and the phones (regardless of type) will randomly go into "Failover" mode and reboot. The Shoretel tech said this was because they were losing connectivity to their ShoreGear T1k switches at our main location and our network was the cause. I have ran Wireshark and captured VoIP traffic from our most troublesome site and could find no such occurrence of dropped packets. Everyone conversation I analyzed had minimal jitter and seemed fine. But then again I am not a wireshark professional. I enabled SNMP and have been running Solarwinds Engineer's toolset using NPM, Interface Monitor and monitoring all the routers and switches in between the locations up to the Shoretel switch. None of them show any dropped packets, errors or discards what so ever. Their logs are all clear of any out of the norm errors. The CPU stays low on all the units as well as the memory and the switches have more than enough buffer for the network traffic. I checked to make sure the ports connecting the Shoretel switches were 100mpbs full as I had seen that could be an issue. I also switched the HP switches that the Shoretel switches were plugged into to make sure that wasn't the issue. The weird thing is, the one location that connects via IPSec VPN and has IP Phones (hollister on the map) does not have this issue. The only common denominator is the Main Internal Router and our Metro Ethernet itself. I have attached a generic Netmap to kind of show you the layout. Any help or ideas would be much appreciated. If you need more information please let me know. Thanks.