Some Avaya 9608 handsets work, others don't - FIXED

matty-boy · ‎09-15-2014

Hi,

We're looking to roll out Avaya 9608 handsets to a site that currently do not use IPT.

Another team look after the Avaya side of things. I'm on the LAN/WAN side. Site have one large data VLAN but we need several smaller voice VLANs for each building so rather than using DHCP option 242 in the data VLAN to inform the handsets of the voice VLAN, we are using LLDP.

The switches are mostly 3560s and 3750s running 12.2(44)SE6 IPBASE. We are currently testing using a single port/cable.

The issue we have is that some phones (brand new out of the box) pick up the voice VLAN, get an IP, register to the call manager and work perfectly while others don't.

I'm completely stumped as to why they don't work??? There are plenty of spare IPs in the DHCP pools for the data and voice VLANs. There is no port security so we not hitting an MAC limit.

The bad phones correctly pick up the voice VLAN and DHCP discover out but don't appear to get a response. You'd think this was a fault with the DHCP server but why do some phones work and others not?? Helper addresses on the SVI are good. The DHCP servers are offsite but work fine for all other VLANs. Remember - some phones work in this very same port, whilst other do not.

We've not implemented IPT using LLDP before (usually just use DHCP option 242 in data VLAN) so folks are starting to point their fingers at this but I'm not convinced - especially as the phone appears to pick up the correct voice vlan.

Any help is very much appreciated!

Thanks in advance,

Matt.

Manish Prasad · ‎09-15-2014

Can you enable packet capture when you plug phones to port and attach it here.

It would be helpful if you capture on working and non working condition.

matty-boy · ‎09-16-2014

Sorted.

We use "Proteus" for IPAM here. Most sites are configured in a centralised configuration whereby they use a primary DHCP server at one core site and secondary DHCP server at another core site.

This particular site which is moving to IPT is large and has it's own Proteus appliance as primary and one of the core sites as secondary.

Turned out that the local DHCP server at site was randomly ignoring DHCP requests from some phones (don't know why yet) - seems like a software fault on the server.

In Proteus the secondary DHCP server was configured to be core site 'x' but the second helper address on the SVI was pointing at the core site 'y'. Adding a help for core site 'x' got it going.

Manish - thank you for the suggestion.

Matt.