2012-06-27 01:12 PM
I have a few questions regarding autosize.
Autosize is being considered for snapvault destinations at my customer. I am trying to find the gotchas to implementing autosize, hopefully i m asking the right questions. if there is anything else i should know about autosize please educate me i find the netapp documentation on autosize to be fairly light.
1. A-Sis and autosize = should the max size autosize can grow to be set to the max a-sis size if using A-sis?
2. is there a limit on how many times autosize can execute in a given time period?
3. is there a max number of times autosize can grow a volume?
4. As autosize grows volumes and consumes space from the hosting aggregate, when space is freed in the volume (snapshots rolling off, etc...) if the volume has a guarantee set to none will the free blocks be returned to the aggregate or continue to be trapped in the volume until an administrator shrinks the volume?
any help is appreciated, and supporting documentation would be ideal
2012-06-27 06:18 PM
That would be a good idea to not exceed asis size for vol autogrow (although up to 16TB with 8.0 for all platforms then even better with 8.1 there is no limit except for the platform flexvol limit)
The last I remember vol autogrow is based on the threshold being hit (wafl_reclaim_threshold?) so is based on that threshold when it is met
The max size and increment are set in the autogrow setting. For example to have 5000g maximum and grow in 5g increments below
You can also use snapshot autodelete with vol autogrow but might not be a fit with snapvault if no other snaps on the target.
If volume guarantee is set to none then free blocks do go back to the hosting aggregate.
The man pages and ontap documentation is very good and has command syntax.