IMP Servers HA 10.5

[Wilson Samuel]

Hi All,

I have had configured the HA on IM-P Servers 10.5 and although the failover works fine (i.e. whenever the Primary Server goes down, the Secondary one takes over the roles) and all works great.

However what I have not been able to locate is, once the failed Server comes online, both nodes do not come back to Normal Mode, and it remains in the "Running in Backup Mode" mode unless I go back to the CCM Presence Redundancy Group and do it manually.
 

Am I missing something here? As I want to configure the systems in a way, that, whenever the failed node comes back, it should automatically go to Normal Mode.

Regards

PS: It may not be the most correct forum (being IMP) but, IMP is now part of CCM, so here we are :-)

[Rob Huffman]

Hi Wilson,

Hope all is well! If I'm following along correctly here, there are dependencies that determine whether or not the Fallback to normal mode takes place automatically;

Automatic Fallback

IM and Presence Service supports automatic fallback to the primary node after a failover. Automatic fallback is the process of moving users back to the primary node after a failover without manual intervention. You can enable automatic fallback with the Enable Automatic Fallback service parameter on the Cisco Unified CM IM and Presence Administration interface.

Automatic fallback occurs in the following scenarios:

• A critical service on Node A fails—A critical service (for example, the Presence Engine) fails on Node A. Automatic failover occurs and all users are moved to Node B. Node A is in a state called "Failed Over with Critical Services Not Running." When the critical service recovers, the node state changes to "Failed Over." When this occurs Node B tracks the health of Node A for 30 minutes. If no heartbeat is missed in this timeframe and the state of each node remains unchanged, automatic fallback occurs.

• Node A is rebooted—Automatic failover occurs and all users are moved to Node B. When Node A returns to a healthy state and remains in that state for 30 minutes automatic fallback will occur.

• Node A loses communications with Node B—Automatic failover occurs and all users are moved to Node B. When communications are re-established and remain unchanged for 30 minutes automatic fallback will occur.

If failover occurs for a reason other than one of the three scenarios listed here, you must recover the node manually. If you do not want to wait 30 minutes before the automatic fallback, you can perform a manual fallback to the primary node.

From;

http://www.cisco.com/c/en/us/td/docs/voice_ip_comm/cucm/im_presence/configAdminGuide/9_0/CUP0_BK_CFF5B189_00_config-admin-guide-imp-90/CUP0_BK_CFF5B189_00_config-admin-guide-imp-90_chapter_011101.html 

Cheers!

Rob

[Wilson Samuel]

Thanks Rob for the reply, however I initially (before posting on the forum) dismissed the above info as IMP 9.x, I guess this document stands true for 10.x as well :-)

More over there is almost NOTHING on the IMP 10.x on the CCM Maintain and Operate Guides, more over they have taken away IMP Webpage (well most of it) as it is now "CUCM-IMP".

Hope Cisco learns it and makes is more readable for a product like IMP.

http://www.cisco.com/c/en/us/support/unified-communications/unified-communications-manager-im-presence-service-version-10-0/model.html#~tab-documents

Regards