Skip to main content

Hi,

I am currently testing with Airgap technology to make aux/offline copies to an isolated environment with WORM enabled. To explain the situation bit more this is how it works -- see picture below.

FYI; to Power On/Off the MA’s I use the Workflow: Airgap which commvault provided.
To make it more intelligent I created an PoSH which checks all the jobs that is being initiated by the AirGap Schedule Policy and if all the jobs contains: ‘completed’, ‘completed with errors’ or ‘failed’, the machines are going down using the Workflow.

Now I know there are some Administrative Jobs that has to run as well like;
Data Aging
DDB Backups
Data verification
Reclamation etc...

Ofcourse I can make this part of the PoSH script before Power Off the MA’s but is it smart to do? 
Is there anyone who has experience with the administrative processes and this technology below?
 

 

 

I’m tagging in @Chris Sunderland who would know best.  I’m not sure if the WF already takes this into account (and if not, ensure your idea will work) 🤓


Anyone???


Adding a few more people:

@Jos Meijer , @Onno van den Berg 

@christopherlecky might have some ideas


@Vjduuren What are your exact worries and or requirements? Because you are adding a lot of complexity e.g. a process chaining who will have to be monitored and maintained also introducing risks.  Depending on what you are trying to achieve there might be other options who can be picked out-of-the box.  

 

 

 

 

 

 


@Vjduuren  I will have to think about certain aspects of your question. Because as @Onno van den Berg already mentioned you are adding a lot of complexity to the workflow.

For example data aging. The data aging job maybe finished, but the actual pruning process on the MA is being run in batches on the background so you need a way to identify the work queue in order to assess if the aging has been completed. As this is not possible via the job manager a different information source needs to be used.

On top of that, when a lot of pruning needs to be performed the work queue will take more time to complete and therefore increases the risk of possible compromise which you are actually trying to minimize by shutting down the MA.

You could mitigate this by ensuring a maximum uptime for the MA after which it is forced to end the loop and shut down the MA.

But again, I need some time to think about your question and if possible test such a construct


Tape 😀


@Vjduuren following up to see if you were able to get anywhere based on the advice above.

Thanks!


Reply