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

Need Help with Unity Failover

I just ran the failover wizard on both primary and secondary server last night and all went well. This morning I found that the secondary server went active while Unity still running on the primary server and all voicemail ports on the primary server went unregister in CallManager. Can someone tell me why this happens ? and what I need to do to failback to primary ? and how prevent this from happen.

I am running Unity 4.0.3 sr1

thanks in advance !


Cisco Employee

Re: Need Help with Unity Failover

You should check the event viewer logs, especially on the secondary, to see why it failed over. Often it is because the maximum number of ports available on the primary server was exceeded and Callmanager was configured to call forward busy to the secondary. It could also be something else, where a unity port wasn't picking up for some reason and a call callforwad no-answer to the secondary.

You can go to the Failover Monitor tool on the secondary server and manually fail back or schedule it.

Depending on why it failed over, you may want to turn on the Skinny traces (there's a macro trace under the Unity Diagnostics Tool) should this happen again.

New Member

Re: Need Help with Unity Failover

I am seeing a red x on Replication Monitor in SQL Server Enterprise Manager, there is a red x on the publishers folder, the agents folder and the Queue Reader Agents folder. I think this could be why failover occurs due to replication errors.

Can someone point me to the redirection ?

Thanks !!!


Cisco Employee

Re: Need Help with Unity Failover

Replication errors themselves won't trigger a failover. However that can mean that not all properties are making it from the primary to the secondary (or vice-versa). So you might not have all the same subscrivers, call handlers, etc. But that won't cause failover. You'd still have to look at the application event viewer logs for that.