Mailing List Archive

Art,

Issue the following at the command prompt:

fcadmin device_map

If you see an entry with xxx instead of a number, that is the missing drive. You should also review the message files - you will probably see a line about a missing disk, finding a spare and reconstructing. The "missing" disk is spun down, but not marked as failed. If you pull out the drive you find that the netapp thinks is missing, and put in a new spare, you will first see a "failed" message. Fun stuff!

--sam

-----Original Message-----
From: Art Hebert [mailto:art@arzoon.com]
Sent: Friday, August 09, 2002 1:49 PM
To: 'toasters@mathworks.com'
Subject: I have a volume without a spare...




I was just reading about how a disk can drop out and the spare gets used
but its not marked as failed.
So I thought I would check my filer and sure enough I have a volume that
has the following:

1 parity drive
6 data drives

RAID group 0

RAID Disk HA.ID HA SHELF BAY CHAN Used (MB/blks) Phys
(MB/blks)
--------- ----- ------------ ---- --------------
--------------
parity 4.2 4 0 2 FC:A 34500/70656000
35003/71687368
data 4.4 4 0 4 FC:A 34500/70656000
35003/71687368
data 4.1 4 0 1 FC:A 34500/70656000
35003/71687368
data 4.6 4 0 6 FC:A 34500/70656000
35003/71687368
data 4.0 4 0 0 FC:A 34500/70656000
35003/71687368
data 4.3 4 0 3 FC:A 34500/70656000
35003/71687368
data 4.5 4 0 5 FC:A 34500/70656000
35003/71687368

I always have a spare on a filer so I'm not sure if it has been used for a
failed disk (I go to our hosting once
a week to check all of the machines physically and didn't see any red
lights).

How can I tell if I have a failed disk? I do a df of the volume and it
shows 6x36gb disks.

Filesystem kbytes used avail capacity Mounted on
/vol/vol0/ 190657844 52445200 138212644 28% /vol/vol0/
/vol/vol0/.snapshot 0 3108820 0 ---%
/vol/vol0/.snapshot

My other conclusion is I screwed up when I created the volume, if so how
can I turn one of the data disks
into a spare?

Thanks

art

Sam -- thanks, everything has numbers assigned. I must of screwed up when I
created the volume.

Any way to make one of the 6 data disks a spare?

Thanks

art

-----Original Message-----
From: Sam Schorr [mailto:sschorr@homestead-inc.com]
Sent: Friday, August 09, 2002 2:08 PM
To: Art Hebert; toasters@mathworks.com
Subject: RE: I have a volume without a spare...


Art,

Issue the following at the command prompt:

fcadmin device_map

If you see an entry with xxx instead of a number, that is the missing drive.
You should also review the message files - you will probably see a line
about a missing disk, finding a spare and reconstructing. The "missing"
disk is spun down, but not marked as failed. If you pull out the drive you
find that the netapp thinks is missing, and put in a new spare, you will
first see a "failed" message. Fun stuff!

--sam

-----Original Message-----
From: Art Hebert [mailto:art@arzoon.com]
Sent: Friday, August 09, 2002 1:49 PM
To: 'toasters@mathworks.com'
Subject: I have a volume without a spare...




I was just reading about how a disk can drop out and the spare gets used
but its not marked as failed.
So I thought I would check my filer and sure enough I have a volume that
has the following:

1 parity drive
6 data drives

RAID group 0

RAID Disk HA.ID HA SHELF BAY CHAN Used (MB/blks) Phys
(MB/blks)
--------- ----- ------------ ---- --------------
--------------
parity 4.2 4 0 2 FC:A 34500/70656000
35003/71687368
data 4.4 4 0 4 FC:A 34500/70656000
35003/71687368
data 4.1 4 0 1 FC:A 34500/70656000
35003/71687368
data 4.6 4 0 6 FC:A 34500/70656000
35003/71687368
data 4.0 4 0 0 FC:A 34500/70656000
35003/71687368
data 4.3 4 0 3 FC:A 34500/70656000
35003/71687368
data 4.5 4 0 5 FC:A 34500/70656000
35003/71687368

I always have a spare on a filer so I'm not sure if it has been used for a
failed disk (I go to our hosting once
a week to check all of the machines physically and didn't see any red
lights).

How can I tell if I have a failed disk? I do a df of the volume and it
shows 6x36gb disks.

Filesystem kbytes used avail capacity Mounted on
/vol/vol0/ 190657844 52445200 138212644 28% /vol/vol0/
/vol/vol0/.snapshot 0 3108820 0 ---%
/vol/vol0/.snapshot

My other conclusion is I screwed up when I created the volume, if so how
can I turn one of the data disks
into a spare?

Thanks

art

On Fri, Aug 09, 2002 at 02:19:14PM -0700, Art Hebert wrote:
>
> Sam -- thanks, everything has numbers assigned. I must of screwed up when I
> created the volume.
>
> Any way to make one of the 6 data disks a spare?

Nope, you would need to rebuild it.

--
Mike Horwath Admin & Manager @ VISI.com WORK: drechsau@visi.com
IRC: Drechsau http://www.visi.com/ HOME: drechsau@geeks.org
The only Minnesota ISP with public statistics: http://noc.visi.com/
Garbage In -- Gospel Out. - berkeley fortune(6)

art@arzoon.com (Art Hebert) writes:
>
> Sam -- thanks, everything has numbers assigned. I must of screwed up when I
> created the volume.

It certainly looked that way. A volume won't magically get bigger because
a disc has gone missing, and I assume the 7 x 36GB shelf is all you've got.

> Any way to make one of the 6 data disks a spare?

Sure, backup the whole filing system to tape and rebuild it from scratch!
Or add another disk shelf ...

I see that we had a toasters thread back in August 2000 ("Removing disks
from a volume") speculating on what would be involved technically in
shrinking a volume to recover disks. I'm still sure it would be possible
in principle, but if NetApp are about to add such a facility to ONTAP
they are certainly keeping quiet about it ...

Chris Thompson
Email: cet1@cam.ac.uk