[cisco-voip] What debug should I be looking at..

[Wes Sisk]

Thanks Ed.

I'll look through these. In the mean time a packet capture even from CUCM perspective would be beneficial.  Our heuristic guess at this point is that some device in the network is interfering at the TCP level.

/wes

On Nov 3, 2011, at 8:57 AM, Ed Leatherman wrote:

Attached zip file of debugs w/ the tcp transaction over about an hour.

I've filtered out a TON of "TCP0: ACK timeout timer expired" messages,
and some setting "TCP_TOS Properties" messages.

This was over a 1 hour period, the only sccp profile that gets
mentioned in the sccp debugs is 888 which is the conference bridge -
also looks like that is application type 3.
Looks like a lot of keep alive time-outs for application types 2 and 3.

Not knowing anything more about interpreting the syslog messages, I
would guess that the gateway is not getting keepalives in a timely
fashion from CM and is initiating the switch itself. The same device
also has 6 MGCP controlled PRI's, seems odd that connectivity for
those is not a problem (as far as I can tell).

Still trying to arrange a packet cap at the gateways.

> 
> On Mon, Oct 31, 2011 at 5:08 PM, Wes Sisk <wsisk at cisco.com> wrote:
>> Thanks Ed,
>> 
>> Can you grab these debugs as well:
>> debug sccp message
>> debug sccp event
>> debug ip tcp transaction
>> 
>> deb ip tcp trans can be busy if this box terminates or originates many TCP sessions.  Use with caution.
>> 
>> Regards,
>> Wes
>> 




-- 
Ed Leatherman
<Syslogs.zi_>