I am experiencing problems with our Cisco 2960-S POE switch.
Our scenario is as follows: Cisco 2960 -> Avaya 4606 IP phone -> Windows XP PC
Whenever I plug the phone initially the phone comes online without problem as does the PC. However once the user logs on to the PC and their user profile starts being pulled down and the phone loses its connection to the Avaya phone system.
I am transitioning over to Cisco switches from 3com (4400,4500,5500) switches. Both the PC and the Avaya 4606 work fine with our 3com equipment but not with our Cisco equipment using the same scenario as above
Has anyone else come across this problem or come has come up with a fix??
Thanks your your help in advance!
Can you post the configuration of the interface?
Can you also post the output to the following commands:
1. sh interface
2. sh controller e
3. sh version; and
4. sh log
3Com and Cisco use slightly different port and voice vlan configuration settings. Can you please post the working 3Com port configuration (e.g. for 4500 family) and your current 2960S port (interface) configuration.
Thanks & Regards,
This is the config for the Cisco 2960S
switchport access vlan 60
switchport mode access
switchport voice vlan 61
This is the config for the 3COM 4500
stp edged-port enable
port link-type hybrid
port hybrid vlan 61 tagged
port hybrid vlan 60 untagged
undo port hybrid vlan 1
port hybrid pvid vlan 60
broadcast-suppression PPS 3000
packet-filter inbound link-group 4999 rule 0
Thanks for the information supplied. Will you please add the "auto qos voip trust" command under the Gi1/0/3 interface and see if there is any progress.
Thanks for the post of the output requestion. Here are my questions:
1. Uptime of the switch is 25 minutes. This sure doesn't give me a good picture.
2. What is connected to Gi 1/0/3? Why did the interface negotiate to HALF duplex?
3. Have you tried configuring Gi 1/0/3 as a trunk port and allow VLAN 60 and 61 only?
phone loses its connection to the Avaya phone system
If the Avaya phone system is connected to the Cisco switch, can you post the "sh interface
We have 4625IP phones connected to 3560 POE switches and work with no issues (even when users connect laptops via phone PC port). Below is working config...
description 1st Flr IP Phones
switchport access vlan
switchport trunk encapsulation dot1q
switchport mode access
switchport voice vlan
auto qos voip trust ---> This command will automatically generate multiple QOS related config on interface and in
global mode as well.
Also, the phones negotiate at 100/Full.
@amikat - I have tried that but it does not seem to make any difference. The PC still loses connection.
@leolaohoo - As for your questions -
1. - The switch has been bounced multiple times - that is why there is such a low uptime.
2. - An Avaya 4606 IP phone and Windows XP pc is connected to this interface. The Avaya 4606 are know to only negogiate to 100/half(in some of my buildings that only negogiate to 10half!). I have tried hard-coding speed/duplex to see if that would solve the problem, but it did not.
3. - Yes, I have tried that as well, and the PC will still lose its connection.
This is the config I used:
switchport mode trunk
switchport trunk native vlan 60
switchport trunk allowed vlan 60,61
spanning-tree portfast trunk
The Avaya phone system itself is connected to a 3com core switch in our data center.
@mvsheik123 - Thanks for the config. I've 4610's and 4620's in my enterprise and they are not having issues like the 4606 are. In talking with Avaya, the 4606 phones are actually hubs and anything greater than 4610's are switches.
I've even tried disabling QOS and spanning-tree and tried a 'vanila' config and it still isn't working.
Thanks to everyone who has chimed in help so far, it has much appreciated!! I'm glad to know you guys have been thinking along the same lines I have been trying. I've just never been this stumped with a Cisco problem like this!!
Thanks for the information supplied.
You can try few other configs if you like:
1) Activate the egress expedite queue via the "priority-queue out" interface command (under the Gi1/0/3 interface).
2) I understand that your Avaya 4606 is configured via DHCP. I would guess that you have already set Option 176 L2Q and L2QVLAN parameters. You can also try to explicitly set L2QAUD and L2QSIG parameters - Cisco recommends the values 5 and 3 respectively.
Will you please post the "sh mac add dyn | i Gi1/0/3" command output.
Thanks & Regards,
Thanks for the suggestions.
I activated the 'priority-queue out' and it does not seem to have an affect.
I explicitly set the L2QAUD and L2QSIG parameters and it not help the situation. Also, when I ran the command
"sh mac add dyn | i Gi1/0/3" I did not get any output. Should I have?
Thanks for the reply. Via the requested command output which should display the mac addresses of both the Avaya phone and PC I wanted to check the VLAN assignment. None of my 2960S switches run 15.0 release of IOS so perhaps the output changed at that particular version - can you please modify the command as "sh mac add dyn | i 1/0/3" and see if you get any output.
In addition to that I have checked again the output you have provided earlier and noticed that your Gi1/0/3 is negotiated at 100Mbps/half duplex with lots of collisions. As your speed and duplex settings are at default auto/auto this could be symptom of the fact that the other end settings are hardcoded as 100Mbps/full. So can you please try to configure "speed 100" and "duplex full" under the Gi1/0/3 interface and see if there is any progress.
Thanks & Regards,
Was this issue ever resolved? I am having the exact same problems with Avaya 4612 phones and a 2960S switch. I have tried all the things mentioned here but it still disconnects when the pc becomes more acitve.
Actually no - Cisco was not able to give a resolution to us. They seemed to be stumped and pointed the finger at Avaya saying I needed to talk with them.
Are you using a 2960S with full POE (740 W) or 370 W ?.If you are using a 370W PoE switch,it might be drawing more power..i had the same kind of issue with Avaya Phones...it was more to do with the PoE not with the configuration..