Moebius wrote:Furthermore, I think I like better a script that lists protected VMs on the basis of which are currently included in my jobs, rather than on the basis of the past backup history. I believe it's a matter of preferences here.
Regarding this portion, do you only add VMs to your jobs individually? If so, then that's a pretty easy use case, however, many, many customers choose to select container objects such as datastores or folders rather than individual VMs. This makes the script very complicated to determine which VMs are actually in a job, since I would then have to get the container object that are in the job, and then enumerate that object to find it's child VMs. Then I would have to take into account any possible exclusions, which can also be based on individual VMs or container objects.
On the other hand the session data from the last run would generally include all of the VMs that were backed up based on those enumerated objects so that why I selected that method as it was simply far easier to code and covered all user cases. Also, I wanted to catch cases where VMs were in a job, but failed since the idea was that it would be a scheduled report. It sounds like your use case is different enough but the overall concept is about the same.