We currently have approx. 500 1140e IP phones in a production environment. Prior to this week we were using DHCP from a W2K3 server. However, we just moved DHCP for the voice vlan to the Cisco switch. The challenge we're experiencing with DHCP, regardless the source, is that the phones send a tremendous amount of DHCP requests. We've attempted changing the lease times from anywhere between 8 & 90 days and have not seen any change in the number of requests.
This issue was discovered while attempting to isolate a problem with call recording. In the troubleshooting process we ran wireshark against the voice vlan. This is where we identified the enormous amount of DHCP requests. After looking at the DHCP logs on the windows server, we saw one phone send 350 DHCP requests in a 24 hour period. 350 requests x 400 phones lends to a lot of DHCP traffic.
We use SPAN ports on the Cisco switches to record the calls by mirroring the vlan to the recording servers. We get varying results in the quality of the captured calls.
To introduce yet another variable into the equation, the phones are about 50% new Avaya branded phones, and 50% Nortel refurbished sets.
We've identified several different versions of the phonesets with varying power requirements (class 2 vs class 3). Additionally, we've noticed that some of the phonesets have differentiating mac addresses. The phone gui even conflicts with the mac addresses.. whereas the set info menu reflects one mac address, but the services, services, Local Diagnostics --> IP Set Information menu --> hardware id string indicates the correct mac address, found printed on the back of the set. We validated which address was correct against the arp table in the switch.
Has anyone had experience that would lend some wisdom to this challenge?
