03-06-2012 08:15 AM
I need to move our clustered MAG6611 to a new facility and am considering the following to minimize downtime for new connections:
1. Shut down node 1, node 2 continues to run
2. Move node 1 and bring on line
3. Shut down node2
4. Move node2 and reconnect to cluster
I think this will work, however one caveat is that I need to renumber the internal interface of node 1 at step 2. Will I be able to do this in a degraded cluster state and if so will node 2 rejoin properly at step 4?
Solved! Go to Solution.
03-06-2012 09:39 AM
These steps seem okay, but I wanted to clarify if this is a A/P or A/A cluster. If it's a A/P, you just need to add a step to fail over to the corresponding device. If it's A/A, you just need to configure the load balancer to route traffic to one node.
03-12-2012 01:00 AM
If you are planning to change IP on the SA nodes you need to have a maintenance window planned.
You have to break the cluster , change the IP on the nodes and form cluster between the nodes.
03-23-2012 08:14 PM
Always good to have a maintenance window when doing anything like this; but looks like you should be ok with the plan you have to reduce the downtime as much as possible