I'm just wanting to make sure my understanding of Celerra\VNX placeholder 'pass-through' behavior (which I believe is the default) is accurate. The followiing is from the EV guides:
The syntax of the-read_policy_overrideparameter is as follows:
-read_policy_override [none | full | passthrough | partial]
The effect of the values is as follows:
none (the default value). The setting has no effect.
full. Recall the whole file to the Celerra on read request before the data is returned.
passthrough. Retrieve the data without recalling the data to the Celerra.
partial. Retrieve only the blocks that are required to satisfy the client read request.
For pass-through, the Celerra uses the same cache on the Enterprise Vault server that you set up for Enterprise Vault to use when retrieving files for the Celerra.
So the 'passthrough' option will retireve the data without recalling the data to the Celerra but will use a cache are on the EV server?
Am I correct in thinking that (with passthough configured) if a large number of placeholders are copied from one directory to another that this will not result won't result in the any files being recalled back to the Celerra? Instead EV will open these files in the EV cache location to allow the copy process to read the file and then, once the move is complete, these files will then be deleted from the EV cache location?