I am wondering if it is possible to setup a dataset with VSM -> SV cascade but only send 1 qtree to the SV? I have a situation where a qtree in a volume has long term backup requirement, but the non-qtree data does not. The entire volume also has a requirement for a shot-term DR copy. I have tried by only adding the said qtree as primary data and then importing the VSM and SV external relationships. The VSM imports fine, but when adding the SV into the dataset PM wants to build a SV relationship for the non-qtree data (which I do not want archived).
There is no way to avoid it. As in case of the SV source which is the VSM destination its the entire volume. So PM splits them into individual qtree and snapvault the entire volume including the non-qtree data. You cannot import qtrees as primary member if you wish to have a VSM.
But what is the problem you see in non-qtree data being snapvaulted ? It finish of in seconds a there should be no data.
Can you give me more details on what is your concern or reasoning on not having the non-qtree data snapvaulted ?
I have a NFS/CIFS volume setup for a specific department, most of the data does not have a long term backup requirement, but some of it does. I have created a qtree for the data that has the SV requirement. The entire volume also has a requirement for availability at a secondary site in case we must activate DR using a DR vfiler.