RSM - Automated failover

[Kritao]

Hi,

Can somebody help to configure Replication Solution Manager. So that if the source fails automatically the clients should attach to DR.

Currently we are doing this manually. We have one eva at source and one at DR. the server connected to source and destination have the same application. each client is having two shortcut icon of the application one pointing to source and another to dr. once the source fail, users will use the second shortcut to use their application. Can we automate this using RSM, like which can be done through Storage mirroring.

[Ivan Ferreira]

I think that you need "HP StorageWorks Cluster Extension EVA" http://h18006.www1.hp.com/products/storage/software/ceeva/index.html

[Kritao]

Hi,

Thanks for the reply. The one you have suggested is for having a Cluster kind of solution. Can we achieve this without cluster using RSM.

[Sheldon Smith]

While I'm guessing it is possible using the command line interface, the key here is being able to determine if the triggering event actually happened or not.

I believe you would need (to develop) some sort of out-of-band mechanism to accurately determine why the dr site no longer hears the source site: Looking just from within the network, there is no difference at the dr site between "The source site was hit by a tornado (initiate failover)" and "Someone kicked the router power cable (wait a little for it to come back)".

Both are communications failures. Both result in transmission timeouts. How does a program at the dr site tell them apart?