Hi Reza,

Let me try to help out here.

(1) You can in both scenarios uncluster a DMA from cluster B and join it to cluster A, but the big difference is that in the second scenario, the DMA you join, will need to be reconfigured to use the database cluster, and this means it will lose the information from the local database (element data, trend data, alarm data). So, the first scenario would be better, since you would not loose that database information.

(2) In both scenarios, you will be able to carry over the elements as they are stored on the DMA you carry over. The related data in the database (element data, trend data, alarm data) will only be carried over in the first scenario since the second scenario requires the DMA to be reconfigured to start using the cluster database. Then you also have some data in files. During the clustering process, if the data is stored in separate files, it will merge all the data. E.g. all protocols and templates will still exist afterwards. If there is a naming conflict, the data in the cluster first wins. If the data is stored in a single file, then the data in the cluster first wins. E.g. the view structure of your DMS will remain the same as it was originally, the view structure of DMA you add will be lost and all elements will reside under the root.

I hope I was able to bring a little bit of clarity on this subject.

Bert