09-30-2011 06:08 AM - edited 03-19-2019 03:42 AM
Hi Guys,
Appreciate if someone can help!!
Can we cluster two BE 6000 nodes running CUCM, Presence & Unity Connection across two sites over WAN for enhanced resilience. The SRND suggests, only CUCM may be clustered over WAN, remaining co-residing apps may not be clustered over WAN. A second UCS C200 server should be co-located with the primary UCS C200 server to provide redundancy for these remaining co-resident applications.
Thanks in advance.
JS
09-30-2011 08:59 AM
As you've already read, only CUCM can be clustered over the WAN, if you want to cluster CUC, CUPS and UCCX over the WAN then you need the "regular" versions of each product and not a CUCMBE 6000 solution.
Cisco Unified CMBE 6000 may be deployed using the clustering-over-the-WAN call processing local failover model. In this type of deployment, two Unified CMBE 6000 server nodes are deployed at each of two sites to provide geographic redundancy for the Unified CM call processing application. The two Unified CMBE 6000 server nodes may both be UCS C200 Rack-Mount Servers, or alternatively one of the servers may be a regular Cisco Media Convergence Server (MCS).
Only the Unified CM call processing application may be clustered over the WAN. The remaining Unified CMBE co-resident applications (Cisco Unity Connection, Cisco Unified Presence, and Cisco Unified Contact Center Express) may not be clustered over the WAN. However, a second UCS C200 server should be co-located with the primary UCS C200 server to provide redundancy for these remaining co-resident applications.
HTH
java
If this helps, please rate
www.cisco.com/go/pdihelpdesk
09-30-2011 12:50 PM
Thanks Java!!
If this is true, tbh, not very impressed with BE6000, looks to me a big functionality gap in the product at this juncture for SMB's looking true business continuity from lower capex solution. No doubt, product ticks 'cost-effective & feature rich' box of any customer requirements, but I believe still lags of most critical requirement around 'resilience'. It would have been nicer if Cisco has kept the consistency by bringing similar HA features as of 'regular' CUXn & UCCX!
I will wait for a while for others comment to see if someone else has anything to add!! Or otherwise, I may have to revise the design using C200 (or may be C210) with 'regular' products, with a hope C200 running full CUCM, CUXn & CUP can be clustered across WAN to offer true business continuity. Do you have any comments on the revised design??
Ta, JS
03-16-2016 08:03 AM
Hi Java,
Have you a document about this restriction of the cluster over the wan for the BE6K ?
Best regards,
Matthieu
03-16-2016 08:05 AM
Actually right now this has changed, you have no problem clustering over the WAN and following the regular CoW guidelines. However, the number of users/devices you can add to a 6K, does not change because of this.
03-16-2016 08:09 AM
Thanks for your fast answer,
Ok for the cluster over the Wan, and for the cluster on ONE BE6K ? I don't find the text about this configuration...
03-16-2016 08:51 AM
The SRND covers CoW requirements. There is no difference in the config.
03-16-2016 09:05 AM
Ok and two CUCM node on the same BE6K ?
03-16-2016 09:09 AM
Assuming you don't mind having a single point of failure for HW, and you are within the number of apps/resources the server supports, yes.
03-16-2016 09:38 AM
Thanks for your answer,
Assuming we have 2000 devices, we can put 1000 devices on each node of the same BE6K server?
I am in trouble with the limitation of the 1000 users
03-16-2016 09:47 AM
Then you should be looking at a 7K or regular UC on UCS, the 6K does not fit every scenario, and if there are chances of growth, those should be accounted for since the beginning.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide