When connecting a new shelf of storage, it's always a good idea to verify the storage connectivity - at both the shelf and disk level.
First - check the shelf. The shelves are numbered, but you also need to know the SAS loop. You can get a list of all the disks, and find the SAS loop number:
sysconfig -d
Look for XX.YY.ZZ, for example:
4d.25.0 4d 25 0 SA:A XXXXXXXXXX
In this case, 4d is the shelf loop connection. Therefore, do:
storage show shelf 4d.shelf25
You should see two pages of output, similar to the below. You will want to look for anomalies, such as drive 10 below.
Partial Path Link Invalid Running Loss Phy CRC Phy
Disk Port Timeout Rate DWord Disparity Dword Reset Error Change
Id State Value (ms) (Gb/s) Count Count Count Problem Count Count
--------------------------------------------------------------------------------------------
[SQR0] OK 7 3.0 0 0 0 0 0 1
[SQR1] OK 7 3.0 0 0 0 0 0 1
[SQR2] OK 7 3.0 0 0 0 0 0 1
[SQR3] OK 7 3.0 0 0 0 0 0 1
[CIR4] OK 7 3.0 0 0 0 0 0 1
[CIR5] OK 7 3.0 0 0 0 0 0 1
[CIR6] OK 7 3.0 0 0 0 0 0 1
[CIR7] OK 7 3.0 0 0 0 0 0 1
[ 0 ] OK 7 3.0 0 0 0 0 0 1
[ 1 ] OK 7 3.0 0 0 0 0 0 2
[ 2 ] OK 7 3.0 0 0 0 0 0 1
[ 3 ] OK 7 3.0 0 0 0 0 0 2
[ 4 ] OK 7 3.0 0 0 0 0 0 2
[ 5 ] OK 7 3.0 0 0 0 0 0 2
[ 6 ] OK 7 3.0 0 0 0 0 0 1
[ 7 ] OK 7 3.0 0 0 0 0 0 2
[ 8 ] OK 7 3.0 0 0 0 0 0 2
[ 9 ] OK 7 3.0 0 0 0 0 0 2
[ 10 ] UNKWN 7 NA 0 0 0 0 0 1
[ 11 ] OK 7 3.0 0 0 0 0 0 1
[ 12 ] OK 7 3.0 0 0 0 0 0 2
[ 13 ] OK 7 3.0 0 0 0 0 0 2
[ 14 ] OK 7 3.0 0 0 0 0 0 1
[ 15 ] OK 7 3.0 0 0 0 0 0 1
[ 16 ] OK 7 3.0 0 0 0 0 0 3
[ 17 ] OK 7 3.0 0 0 0 0 0 1
[ 18 ] OK 7 3.0 0 0 0 0 0 1
[ 19 ] OK 7 3.0 0 0 0 0 0 1
[ 20 ] OK 7 3.0 0 0 0 0 0 1
[ 21 ] OK 7 3.0 0 0 0 0 0 1
[ 22 ] OK 7 3.0 0 0 0 0 0 2
[ 23 ] OK 7 3.0 0 0 0 0 0 1
[SIL0] DIS/UNUSD 7 NA 0 0 0 0 0 0
[SIL1] DIS/UNUSD 7 NA 0 0 0 0 0 0
[SIL2] DIS/UNUSD 7 NA 0 0 0 0 0 0
[SIL3] DIS/UNUSD 7 NA 0 0 0 0 0 0
Since there is something funny going on with drive 10 - let's have a closer look. This will give us the multi-pathing status of all the drives.
storage show disk -p
This command will show you all the disks and their respective paths. In this example, you will want to scroll to shelf 25.
4d.25.0 A 5c.25.0 B 25 0
4d.25.1 A 5c.25.1 B 25 1
4d.25.2 A 5c.25.2 B 25 2
4d.25.3 A 5c.25.3 B 25 3
4d.25.4 A 5c.25.4 B 25 4
4d.25.5 A 5c.25.5 B 25 5
4d.25.6 A 5c.25.6 B 25 6
4d.25.7 A 5c.25.7 B 25 7
4d.25.8 A 5c.25.8 B 25 8
4d.25.9 A 5c.25.9 B 25 9
4d.25.10 A 25 10
4d.25.11 A 5c.25.11 B 25 11
4d.25.12 A 5c.25.12 B 25 12
4d.25.13 A 5c.25.13 B 25 13
4d.25.14 A 5c.25.14 B 25 14
4d.25.15 A 5c.25.15 B 25 15
4d.25.16 A 5c.25.16 B 25 16
4d.25.17 A 5c.25.17 B 25 17
4d.25.18 A 5c.25.18 B 25 18
4d.25.19 A 5c.25.19 B 25 19
4d.25.20 A 5c.25.20 B 25 20
4d.25.21 A 5c.25.21 B 25 21
4d.25.22 A 5c.25.22 B 25 22
4d.25.23 A 5c.25.23 B 25 23
As you can see, drive 10 in this shelf is missing a second path - indicating either a problem with the disk shelf backplane, or the drive itself. If you have a known good spare - try swapping drives and checking the paths again.
If the failed path moves to the new slot where you swapped the drives - the problem is with the drive. Just replace it.
If the failed path stays in place, you have a larger issue - most likely a failed backplane or connector. Try blowing compressed air into the slot where the drive goes, and try again. If it still fails, contact NetApp support for a new shelf.
No comments:
Post a Comment