AFF

Why is there no command line for a failed disk drive ?

pippen23
5,129 Views

In the past couple months, we have had a couple disk drives fail, replacement parts ordered and shipped to us. But, there is no Netapp command to tell which disk has failed from the command line...so, I end up walking down to the basement, checking for an amber light, why ? Why is there no disk show command status =failed ???

 

Grrr

1 ACCEPTED SOLUTION

pippen23
4,616 Views

Failing over the cluster and a giveback fixed the issue, we needed to do s SCSI reset, which cleared the error.

View solution in original post

8 REPLIES 8

SpindleNinja
5,105 Views

Try... 

cluster::>storage disk show -container-type broken  

(there's several other variations too) . 

 

 

pippen23
5,049 Views

-container-type -container-name
MDC-1N01::> storage disk show -container-type broken
There are no entries matching your query.

MDC-1N01::>

 

This is a false positive after changing out a failed disk, and, the amber light is on the disk and shelf. I am looking for a command to determine what is the status of the amber light on a disk(or any disk), without having to march down to the datacenter to check the status.

andris
5,036 Views

Here's a few commands to help understand about storage/disk faults.

::> storage disk show -broken

 

::> storage disk error show

 

::> node run -node <name> -command "storage show fault -v"

 

SpindleNinja
5,032 Views

If there is an amber light on after the disk change it's probably not assigned.    

 

But as far as disk status in the shelf.   check out this command:   

 

CLUSTER::> storage shelf show -bay

Shelf Name: 2.0
Stack ID: 2
Shelf ID: 0
Shelf UID: 00:00:00:00:00:00:00:00
Serial Number: xxxxxxxxxxxxx
Module Type: IOM6E
Model: DS4246
Shelf Vendor: NETAPP
Disk Count: 24
Connection Type: SAS
Shelf State: Online
Status: Normal

Bays:

Has Operational
ID Disk Bay Type Status
--- ----- ----------- -----------
0 true single-disk normal
1 true single-disk normal
2 true single-disk normal
3 true single-disk normal
4 true single-disk normal
5 true single-disk normal
6 true single-disk normal
7 true single-disk normal
8 true single-disk normal
9 true single-disk normal
10 true single-disk normal
11 true single-disk normal
12 true single-disk normal
13 true single-disk normal
14 true single-disk normal
15 true single-disk normal
16 true single-disk normal
17 true single-disk normal
18 true single-disk normal
19 true single-disk normal
20 true single-disk normal
21 true single-disk normal
22 true single-disk normal
23 true single-disk normal

Errors:
------
-

 

storage shelf show -instance will give you a lot of detail about the shelves / disks 

tekievb
4,976 Views

You can run the commands below:

 

cluster::> storage show disk -broken or system node run -node NODENAME "vol status -f"

 

If you have quite a few shelves/disks, I always ensure that I turn the led_on just to ensure the lights are on and disk is easily identified.

 

system node run -node NODENAME  (will take you to the old prompt like back in 7-mode)

priv set -advanced

led_on diskID

 

Once the disk has been replaced you may need to assign disk ownership if auto assign is not enabled.

 

RajeshPanda
4,826 Views

@pippen23 Following up to see if you are still looking for the solution.

pippen23
4,788 Views

working with support, turns out I may have a bad bay in a shelf...ouch!

pippen23
4,617 Views

Failing over the cluster and a giveback fixed the issue, we needed to do s SCSI reset, which cleared the error.

Public