IMP HA pair ServerAction setup

[cdevkota-sc]

Hello, 

I have two IMP Servers with HA enabled in CUCM Presence Redundancy Group Configuration. The serverAction for both servers is Failover. Is that the correct setup for the failover and when one server goes down so that users moves to another server automatically?

[Carlo Poggiarelli]

Hi,

You can balance users by assigning them to a different  IMP node through User Management --> Assign Presence Users menu.

Checking the presence redundancy group status, you'll see how many users have been assigned to one node and how many to the other one.

Once one of two nodes should fail, all users previously assigned to the failed node will filover to the remaining node.

HTH

Regards

Carlo

[Shalid Kurunnan Chalil]

Hi there,

I would like to add , In an ideal scenario, the High Availability (HA) setup resembles the configuration depicted in the attached screenshot.

Your interpretation is correct—during normal conditions, the Failover button will be highlighted, indicating the system running in a normal mode. As @Carlo Poggiarelli if one node fails, the users assigned to this node will fail over to the remaining node.  "Fallback" option will be available if the node is Idle or Failed over state. 

Regards,

Shalid 

Disclaimer:

Responses are based on personal knowledge and experience. Consider them as guidance. Other members may offer different perspectives or better approaches. No responsibility is assumed for outcomes; discretion is advised.

[Nithin Eluvathingal]

Below is a screenshot of how the users are distributed between the IMP nodes in a HA pair. If the nodes are in normal state, you will see a similar image on your Presence redundancy group. The important thing to check is the server state, which should be normal. If there is an issue, you will see it on the server state and the reason will be shown on the next column. Sometimes, you can use the failover button to bring the IMP nodes in the HA pair to a normal state. In other cases, you may need to restart the server itself.

When one of the nodes in the HA pair fails, the users on that node will be automatically transferred to the other node in the same HA pair, not to another server in a different presence redundancy group.

https://www.cisco.com/c/en/us/td/docs/voice_ip_comm/cucm/im_presence/configAdminGuide/14_su2/cup0_b_config-and-admin-guide-14su2/cup0_b_config-and-admin-guide-1401_chapter_0100.html