How to configure hot spare disks.

[DEEPSEA678]

Hello All,

We have a storage Data ontap 7.3. When I try to create an aggregate using GUI it is throwing an error “ There must be two hot spares to create an aggregate” .  I Could not find any option to configure spare disks in GUI and I am not familiar with netapp commands.  I searched on internet for commands to configure hot spares but no luck.

  Can anybody help me to configure hot spares and aggregate.

Thanks in advance.

Deep

[RAVI_PALADUGU]

Hi Deep,

First assign a owner for the disks, the disks that are assign a owner and don't have any data in them are taken as hot spares. Check if the disk auto assign feature is enabled from "options disk" if not do this "disk assign <disk id>".

If you want to create a aggregate you can just specify the number of disks you want it to contain and use this command.

Examples:

=======

The following command creates an aggregate called newaggr, with a RAID group size of 8, consisting of the disks with disk IDs 8a.16, 8a.17, 8a.18, and 8a.19:

aggr create newaggr -r 8 -d 8a.16 8a.17 8a.18 8a.19

The following command creates an aggregate called newfastaggr, with 20 disks, the default RAID group size, and all disks with 15K RPM:

aggr create newfastaggr -R 15000 20

The following command creates an aggregate called newFCALaggr. Note that if SAS disks are present, they might be used, because FC and SAS disks are considered to be the same type.

aggr create newFCALaggr -T FCAL 15

After you finish

You can use the aggr status -r command to verify the RAID groups and disks used in the aggregate you just created.

[RAVI_PALADUGU]

I am sharing a example that I typically do on my filers.

The below command creates aggregate "aggr1" with raid as "raiddp" of raid group size 24 with 48 SAS disks of size 300 GB each. Also I have made snap reserve to 0 on that aggregate and also made no schedule for snaps on that aggregate.

aggr create aggr1 -t raid_dp -r 24 -T SAS 48@300

snap reserve -A aggr1 0

snap sched -A aggr1 0 0 0

For making disk assign auto:

=======================

disk assign auto

verify:

=====

disk show -v

[DEEPSEA678]

Thanks Ravi for your inputs. I tried the following command but no luck. I could see one aggregate when i run 'sysconfig -r' command. That aggregate is having one data and one parity disks. Is it mandatory to have an aggregate before creating another. If not, when I try to delete the existing aggregate it is throwing an error 'it contains root vol'.   

dave2> aggr create aggr1 -t raid_dp -r 12 -T FCAL 12@134

aggr create: Couldn't create aggregate: 12 disks needed, but no disks of the requested size are available.

[bondbhola]

Hi Deep,

First chect that how many disk is there which is free. use the below command.

aggr status -s

then create the aggregate.

aggr create <aggr name> <number of disk> for example if i want to create aggregate with 10disk then command would be.

aggr create <aggr name 10 < here please we sure that you have enough disk  and also do not forget that you have enough spare disk. If you have 10 disk only then don't craete the aggregate with 10 disk. keep two disk as free/spare.

Here is the output from your side.

dave2> aggr create aggr1 -t raid_dp -r 12 -T FCAL 12@134

aggr create: Couldn't create aggregate: 12 disks needed, but no disks of the requested size are available.

Its say that there is not enough disk < same size of disk>  is availabe in filer to create the aggreate. First please check that how much disk is avaialbe for you.

and i am sure that you want to create the aggregate with same size/same kind of disk not the misc one.

Please get back me if you have still question.

Thanks,

Bhola Gond

[DEEPSEA678]

Thanks Bhola for your inputs. Please do find the output of above said commands. I believe it does not have any spare disks to create an aggregate or; a spare pool. I could see a few disks are present on storage. Please do find the below commands output. As i said you earlier, Its already having an aggregate called aggr0. What is the role of aggr0 aggregate. If it does not hold any configuration info then we will wipe out that as well to create a new aggregate. How to assess raid group size. Are there any recommendations for raid group size. 

dave2> aggr status -s

Spare disks (empty)

dave2> sysconfig -r

Aggregate aggr0 (online, raid4) (block checksums)

  Plex /aggr0/plex0 (online, normal, active)

    RAID group /aggr0/plex0/rg0 (normal)

      RAID Disk Device  HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks)

      --------- ------  ------------- ---- ---- ---- ----- --------------    --------------

      parity    xx.xx   0b    1   11  FC:A   -  FCAL 15000 136000/278528000  137104/280790184

      data      xx.xx   0b    1   12  FC:A   -  FCAL 15000 136000/278528000  137104/280790184

Spare disks (empty)

Partner disks

RAID Disk       Device  HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks)

---------       ------  ------------- ---- ---- ---- ----- --------------    --------------

partner         xx.xx   0b    1   5   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   0   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   3   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   1   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   8   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   2   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   13  FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   10  FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   7   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   6   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   4   FC:A   -  FCAL 15000 0/0               137104/280790184

partner         xx.xx   0b    1   9   FC:A   -  FCAL 15000 0/0               137104/280790184

[RAVI_PALADUGU]

You may not want to delete aggr0, it may contain "root volume" so better don't do that.

Also I think I clearly explained how spares will be on DataONTAP above.

[bondbhola]

This out put show that you do not have any disk owned by controller dave2. All disk is in list is for partner contoller.

Please give us the below coman doutput.

dav1> storage show disk

dav2> storage show disk

dav2> aggr status -s

dav2> cf status

Thanks,

Bhola Gond

[SATISH_PAMARTHI1983]

try to unassign from the spare disk which was in spare pool of the partner which will be moved to unowned disks pool.

From there you can assign to the filer which u're looking for spare disk pool.

Hope it explains..