2011-12-05 06:45 AM
I was wondering what the best place is to have Data Ontap 8. At the moment the Data Ontap vol is placed in an aggr with other volumes. Would it be better to place Data Ontap vol in its own small aggr? The immediate reason is that Im going to destroy the same aggr to make new aggr so the Data Ontap vol is going to be moved/destroyed and for future work I was thinking if it wasnt better if Data Ontap vol were placed in its own aggr to avoid having to either move it or destroy it? I would appreciate either a comment or a reference to where I can read more?
At the same time I need to move Data Ontap from the aggr I need to destroy. Can I move it to another aggr and make Data Ontap boot from there? Then create the new aggr and move Data Ontap back? Is that the easiest way? Again I would appreciate a comment or a reference to where I can read more?
2011-12-05 07:01 AM
Do you mean moving root volume? Yes, that's doable and is fairly straightforward - you can find commands in this thread (in a Simulator context, but works on a 'real' filer as well):
Re dedicated aggregate: you can have one, but it has its downsides as well - less usable space & less spindles to serve you data. Have a look at this thread:
2011-12-06 07:41 AM
We've had some heated discussions on here about separating root...search and there are more discussions on it We tend to be more conservative and use a separate root aggr...the exception is smaller systems... 2000 series... 2200 series... usually not a lot of disk and it just isn't worth the edge case reasons to have a separate root aggregate. We should put a polling question on what % of people use a separate root aggr... I'll try to post that now.