Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

CUCM cluster issue over the WAN

Hi Expert,

One of my client is having CUCM cluster over WAN and the two sites connecting via two different link the main link is using for all traffic but its not reliable I mean the delay can reach sometime more than 100 ms which is not recommend by Cisco “SRND”. I want to utilize the second link “less than 10 ms” to

be only for sync traffic between CUCM publisher and Subscriber, what is the type of port, size of traffic and what is the internal time for each update,


Saeed Idris

Everyone's tags (5)
Cisco Employee

CUCM cluster issue over the WAN

Hi Saeed,

The maximum one-way delay between any Cisco Unified CallManager servers for all priority ICCS traffic should not exceed 20 ms, or 40 ms round-trip time (RTT).

The ports used are:

TCP 8002 - intracluster communication service (SDL Link/Intra-Cluster Communication)

TCP 8003 - intracluster communication service (Used for signaling from CCM to CTIManager)

Please refer to:



If you find this post helpful, please rate!

Regards, Tere. If you find this post helpful, please rate! :)
Cisco Employee

Re: CUCM cluster issue over the WAN

Actually depending on the version that can go up to 80 ms round trip, review same chapter from the SRND for your cucm version to see which one applies

Sent from Cisco Technical Support iPad App



if this helps, please rate
New Member

CUCM cluster issue over the WAN


If my links are are 70-72 ms under normal conditions but if my provider has an outage or there is congestion on the network the latency could increase above 80ms. Will TAC support this.



CUCM cluster issue over the WAN

Hi Altaf

The SRND documents what TAC support (80 ms). If your round trip falls outside what is in the SRND, then, no, TAC will not support you.

My own experience is that clustering will generally work ok even at higher than 80ms, but you are on your own from a support perspective.


Barry Hesk

Intrinsic Network Solutions