Network and Storage Protocols

Disable Hotspare requirement, extend degraded mode, commands?


I am managing a NetApp FAS2020 filer for a small business with 50 desktops and 5-6 servers.

The company isn't ready to purchase additional storage right now and I would like to make use of one hot spares as part of an active aggregate.

We have two heads, HEAD1 is just being used for CIFS traffic. HEAD2 is being used for NFS and iSCSI traffic. There are a total of 12 15k SAS drives. They heads configured active/active and all four NICS are in-use and redundant. The company doesn't have any time/mission critical data and is okay with 3-4 hours of emergency downtime once a year (which we have done a good job of avoiding). We have 4-hour premium support in NetApp and are in a large metropolitan city. In a few months we'll have SnapMirror of all our data out-of-state for a full-on disaster.

HEAD1 has 2 data, 1 parity, 1 dparity and 1 hot spare.

HEAD2 has 3 data, 1 parity, 1 dparity and 2 hot spares.

I want to place the hot spare for HEAD1 into an aggregate.

I want to extend the functioning degraded time period for HEAD1 to be 48 hours (so in the case of a failed drive we have plenty of time to replace it).

I will also place 1 hot spare into an aggregate on HEAD2 and leave the one hot spare. In a severe case I would move the hot spare from HEAD2 to HEAD1 if we can't get a replacement drive and it is critical. Another highly unlikely scenario.

Eventually the company will purchase another disk shelf of SATA drives and move all the CIFS data to it to free up the SAS drives on HEAD1 for something else.

What commands do I need to use to accomplish this? Can I remove the ONTAP requirement for a hot-spare and extend the degraded mode to 48 hours?

Any help is appreciated. Thank you!

- Elan



options raid.timeout 48   # default is 24 ... be careful extending, but you definitely have a handle on this from the great explanation of what you are doing.


Note that this option timer goes into effect when a drive fails and no spare to fail to as you described, but also in the case of NVRAM battery failure...


Also if removing spares completely it makes sense to adjust raid.min_spare_count to avoid nagging warnings.


Can anyone point to the command-line commands I use (or where in the System Manager 2 GUI) I do the following?

1) Add the hot-spare as a data disk to an aggregate for HEAD1

2) Run the re-allocation command to spread the old data on the new drive