How to handle space issues on Netapp

There might be situation where the aggregate got filled up and there is insufficient space on netapp to carry out the regular tasks like SIS and snapmirror and this might result in performance degrade. Indeed i faced this kind of situation once.

The following would be the actions you can try immediately on the your filer to make sure they are up and running without any issues.

1) Do a datamotion of the volumes across the aggregates( I have explained this in my older post- http://kumaraysun.wordpress.com/2013/04/04/moving-volume-across-aggregates/ )

2) Identify the oldest volume snapshot of the filled up aggregate and delete the same

you can achieve this with the following command

snap list -V <volumename>

hourly.0 would be latest snapshot

say for example your oldest snapshot is hourly.8 then you can delete that. this would regain some space on the aggregate.

to delete it, the command would be as below

snap delete -V <volumename> hourly.8

the same can be performed across multiple volumes in the aggregate and regain some considerable space.

Note: Please bear in mind deleting the older snapshot means you are loosing your backup( exercise with caution and only in a situation when there are no other option)

3) We can also delete the aggregate snapshot in the similar way.

snap list -A will list the snap shot for all the aggregates in the filer

snap delete -A <aggrname> hourly.2

this will delete the oldest snapshot of that aggregate

Deleting the aggregate snapshot could be disastrous ( exercise with caution and only in a situation when there are no other options)

4) Reduce the aggregate snap reserve

The default snap reserve on an aggreate is 5% which is 5% of the total aggregate space. we can reduce this to 3% and regain a considerable space on the aggregate.

snapreserve -A will list the aggregate snapreserve  for each aggregate

to change to 3%

snapreserve -A <aggr name> 3

5) There might be cases where the SIS is running which has drastically increased the aggregate space after a huge change on the filer.( dedupes would be running for a long time when there is’nt sufficient space on the aggregate)

we can temporarily stop all the Dedupes and try the steps 1 -4 and then start the dedupes once we get space on the aggregate.

6) Deleting the volume snapshot would increase the aggregate snapshot( so you will have to plan accordingly)

Do share with me if you have come across a aggregate full situation and how you handled it better.