DI configuration modification

If so, how does that effect the data that has already been gathered as well as future data?

Drahrig:

If I am understanding you the idea would be to discover and audit data currently upon each clustered node versus the cluster itself. We will have to discuss the caveats and assumptions.

First when we scan a cluster we base it upon the resources (up to windows 2012 where the underlying logic is changed) and not the local file systems. The files are associated to a cluster and do not depend on which is the active node controlling the advertising of the share. When you switch to a winnas agent it will be based on the local filesystem and the resources currently on the node will be captured. Upon failover they will no longer be valid for that node and will instead be associated with the active node.

Path definition would be the \\Node_name\share\filesystem_hierarchy and the Node_name change would result in new objects being created and obsoleted for removal based on retention policy in the application. This would affect the past as well as the future association of objects to path definitions which would also influence reporting and the outcome of indexing as the newly discovered shares would reregister as they changed between the advertisers.

Auditing of the access events would also be affected as a user's context would change with the value of the path making consistent reporting difficult to gauge. Also as you mentioned the DFS servers it would be desirable to audit at the referral point of the location shared since the end user would actually be using this location.

I would suggest that you initiate a support case and discuss the issue interactively with Symantec support personnel to allow for a true understanding of the environment you are providing details for and the many consequences as there are a lot of assumptions that must be made in a question of this type on the forum format. I feel you should seek exact answers for your proposal.

Rod.