Most organizations using Cisco's CallManager actually use two of them, and they're configured to fail over for each other. But sometimes, when there is a failure or you need to do some maintenance, all your devices will wind up registered to the backup. Then the question arises, "How do I get them back on the primary?" The phones will likely take care of themselves, and their configuration is simple enough, but you'll have to put some thought into the gateways.
For IOS gateways using MGCP, you have 4 options. The first is called "graceful fallback." After the gateway sees that the primary CallManager is available, it waits until there are no active calls before reconnecting itself to the primary. Cisco recommends this option, but it's not always appropriate, particularly if you have a busy environment; it may be a long time before you have no calls.
The second method is "immediate" which is fairly self-explanatory. As soon as the gateway detects the primary is again available, it connects immediately.
The third option is scheduling a specific time of day for the switch, like a convenient quiet period; lunch or evenings, perhaps.
Finally, you can configure the gateway to reconnect to the primary CallManager based on a configurable period of time after it detects the CallManager's is availablity. This would be useful if your CallManager is unstable for some reason and you're not confident it's up for good. You can configure the gateway
Requires Free Membership to View
All of these options are configured with the command "ccm-manager switchback". When making the decision about how to switchback, consider your particular environment, and how important it is for you to get off the backup.
One other thing to keep in mind: The rate at which your phones switch is very important. By default, they are rate-limited to 10 phones per second. This means that a thousand phones would move in roughly 2 minutes.
Thomas Alexander Lancaster IV is a consultant and author with over ten years experience in the networking industry, focused on Internet infrastructure.
This was first published in August 2003
Network Management Strategies for the CIO

Join the conversationComment
Share
Comments
Results
Contribute to the conversation