06-12-2019 06:07 AM
Hi All,
I need to extend UCCX (in HA mode) capability from 300 to 400 agents.
Since the actual servers are based on OVA template for 300 agents the only way is via DRS backup by creating new VMs based on 400 agents if I'm correct.
Can be done without service interruption not switching both of the servers OFF?
my concerns are:
- Since the new VMs (Master and Slave) should be built by using the same network configuration (also administrator credentials, security password) and that old server definitely should be turned off, which one is under restoration onto the new VM otherwise IP address conflict can happen?
- Can cause any problem that if one server restored with template 400 but the other one is still with template 300?
- Will the license file also be restored?
so if service could not be stopped I think following steps can be done:
- make backup from UCCX Cluster
- create VMs for Master and Slave based on VM template for 400 agents
- shut Master server down (Server_1)
(slave will switch to Master role)
- install UCCX on one of the created new VM with the network config for Master server (Server_1)
(during installation set this server as first node)
- after installation of Server_1 do data restore on it
>> what to do after this point?
to follow the steps on link below(?):
https://www.cisco.com/c/en/us/td/docs/voice_ip_comm/cust_contact/contact_center/crs/express_10_5/maintain_and_operate/guide/UCCX_BK_CEE84C5F_00_cisco-unified-contact-center-express-operations-guide-10-5/UCCX_BK_CEE84C5F_00_cisco-unified-contact-cente...
Restore Only First Node in HA Setup (with Rebuild)?
after that
on same link
Restore Second Node in HA Setup (with Rebuild)?
Many thanks
Kind regards,
Tibor
06-12-2019 07:36 AM
Hi there
Unfortunately redeploying OVA and rebuilding UCCX is the only option for you.
Hope this Helps
Cheers
Rath!
***Please rate helpful posts***
06-12-2019 07:58 AM - edited 06-12-2019 08:06 AM
You do not need to re-build new UCCX cluster. Adding of vCPU and vRAM is supported for UCCX virtual machines (adding vDisk and vNIC is NOT supported), but you must match resources as specified in OVA Template. In your case, you only need to increase vCPU from 2 to 4 and vRAM from 8GB to 16GB for both UCCX virtual machines. The existing vDisk size of 2x146GB meets the specification for 400 Agents OVA template.
Check these sections for more details:
VMware Feature Support for Contact Center
06-12-2019 12:10 PM
Hi Vaijanath,
This is what I thought initially, that need to add vCPU and vRAM (as you also pointed HDD is same like for 400 agents), but saw lot of posts here that DRS is needed, since OVA locks the max number of ports.
So now I'm little bit confused.
Thanks,
Tibor
06-13-2019 05:49 AM
06-13-2019 07:51 AM
Hello Mike,
Many thanks for your valuable information again.
just one more addition:
I also want to upgrade the UCCX Cluster (from 10.5) to 11.6, so I guess after I have moved application into the new VMs (HA mode) I can do migration via PCD.
My question is around license before upgrade:
after moving UCCX Servers from the old 300 template VM to 400 template VM should I ask for license for the 10.5 from licensing team or that will not needed and will also be restored?
(Of course after upgrade to version 11.6 I need new license to install)
Thanks in advance!
Regards,
Tibor
06-13-2019 08:20 AM - edited 06-13-2019 08:21 AM
-> Migration task for UCCX using PCD is not supported. Only fresh install and upgrade task is supported for UCCX. Check Supported Tasks for Cisco Unified Contact Center Express.
-> When you deploy new UCCX virtual machines using new OVA template, it will generate new license MAC address. At this moment, you need new license file for version 10.5. When you do the upgrade from 10.5 to 11.6, then again you will need new license file for version 11.6.
06-13-2019 07:59 AM
If you want 400 simultaneous active agents then yes, you do need to re-build new UCCX cluster to increase the capacity.
06-14-2019 02:39 PM
One other thing to call out here: 400 agents, or even 300, is pushing the envelope of what CCX is capable of supporting. Be certain your solution will fit within the published limits. For example, the BHCC limit of 6000 will require call connection times of four minutes depending on your abandonment rate to stay under the cap at 400 people. Also be confident that CPS spikes are highly unlikely for your workload since CCX is only capable of 1.6 calls per second. Other limitations such as team size or quantity of supervisors also frequently become problematic at this scale. Review the SRND:
Also consider using the UC Sizing Tool to validate everything. I believe this tool still requires partner-level access:
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