Not enough spare disk

[DBARAJAS123]

A hard drive failed and we replaced it with, but we continue to get:

"There are not enough spare disks. /aggr3/plex0/rg0: Please add spare disks supporting block checksums with minimum size: Used 4"

When it was placed in the the netapp, the disk had a bad label, which I unfailed it and then zero'd it.

I'm fairly new to this, any suggestions would be helpful.

[jeras]

The specific RAID group listed in the error message (/aggr3/plex0/rg0) is the key to the message.  In looking at the sysconfig -r output that you supplied, that RAID group has 15K 450GB disks in it and there are no 450GB 15K disks in your spares pool that are listed at the bottom of the sysconfig -r output that you posted here.  If you do have spare 450GB 15K disks in the system, you may want to check the output of the disk show -n command to see if those disks are not assigned to be owned by this particular controller.  If the spare 450GB disks are not assigned, you'll need to assign ownership to those disks with the disk assign command.

Ken

[DBARAJAS123]

Thanks for the reply. I'm not able to run disk show -n could be related to the version we are using 7.3.3. I checked managed disk - Unowned disks and empty. What would you suggest?

[jeras]

What's the FAS model for your controllers?  Software disk ownership was introduced with the FAS3020/FAS3050 models and dates back to at least Data ONTAP 7.1, so the disk show command is available in Data ONTAP 7.3.3.   What is the message you receive when you enter the disk show -n command?

Ken

[DBARAJAS123]

This is the message I get:

netapp1*> disk show -n

disk: Did not recognize option "show".

usage: disk <options>

Options are:

        fail [-i] [-f] <disk_name>  - fail a file system disk

        remove [-w] <disk_name>     - remove a spare disk

        swap                        - prepare (quiet) bus for swap

        unswap                      - undo disk swap and resume service

        scrub  { start | stop }     - start or stop disk scrubbing

        unfail [-s] <disk_name>     - unfail a disk  (-s  not valid in maintenance mode)

        power_cycle [ -f ] { [-d <disk_list>] | [ -c <channel_name> [ -s <shelf_number> ] ] } - power-cycle one or more disks

        replace {start [-f] [-m] <disk_name> <spare_disk_name>} | {stop <disk_name>} - replace a file system disk with a spare disk or stop replacing

        upgrade_ownership           - upgrade disks to new ownership scheme

        shm_stats [<disk_name> | asup | clear_errors]     - Storage Health Monitor statistics for a disk

        zero spares                 - Zero all spare disks

        checksum {<disk_name> | all} [-c block | zoned]

        sanitize { start | abort | status | release } - sanitize one or more disks

        maint { start | abort | status | list} - run maintenance tests on one or more disks

netapp1*>

[jeras]

Would you run a sysconfig -a and post that output?  I'd like to confirm the FAS model and that output will provide that.

Thanks!

[DBARAJAS123]

I have included the a screenshot of the output. Let me know if you need anymore information.

[jeras]

Thanks for posting the sysconfig -a output.  This helped clarify why the disk show command is not available.  Your controller model is the FAS3020 and that controller model supported hardware or software disk ownership. By default, hardware disk ownership was used on the FAS3020 and that is your current configuration.  The disk show command is only available when software disk ownership is used.

This then brings us back to you not having a 450GB 15K disk as a spare in this particular system.  You need to add an additional 15K FC disk to eliminate the warning message.

[DBARAJAS123]

Could I replace one of the spares with a 450 GB 15K drive? I'm not too familiar with the system itself.

[jeras]

That is an option.  Doing so will drop the number of spares of your other disk types to 1.  I would also check the value for the minimum number of spares option (options raid.min_spare_count) and set that value to 1 if it is currently set to 2 otherwise, you'll get warnings from Data ONTAP that you do not have enough spares for the disk type that only has 1 spare in the system.

Ken

[DBARAJAS123]

Ken

I just checked how what the min spare count is for our netapp.

netapp1> options raid.min_spare_count

raid.min_spare_count         1

netapp1*>

So, my only option is just to replace one of the current spares with a 450 correct?

Thanks again for your help!

[jeras · Accepted solution]

Correct.  There needs to be at least 1 450GB 15K FC disk in the system as a spare, so one of the other spares would need to be replaced with a 450GB 15K FC drive.

Glad to help!

Ken

[DBARAJAS123]

Thanks again.