Mailing List Archive

Raid Group question
I have a current volume (vol0) that has a raidgroup size of 8. I had 3
spares that weren't being used and
I wanted to add them to the volume, but like an idiot didn't check the
raidgroup size. Thus when
I added the 3 disks to the volume (vol0) 1 went to raidgroup 0 and the
other 2 went to raidgroup 1 as a parity
drive and a data drive.

This will create a performance problem from what I can tell not mention
another parity drive being used.

I'd like to get these two disks back under raid group 0 if I can.

Suggestions on the best way to do this would be appreciated.

I'm thinking the safest way is to back it up and recreate the volumes and
restore.

Thanks

art hebert
RE: Raid Group question [ In reply to ]
Art,

There is no dynamic way to rebuild raid group sizing, or allocate from one raid group to another. Your statement about the safest way is also the only way that I have ever heard of (well, I guess volcopy would work, copying to another volume, destroying and rebuilding the source, and volcopy back).

--sam

-----Original Message-----
From: Art Hebert [mailto:art@arzoon.com]
Sent: Monday, December 09, 2002 4:42 PM
To: 'toasters@mathworks.com'
Subject: Raid Group question




I have a current volume (vol0) that has a raidgroup size of 8. I had 3
spares that weren't being used and
I wanted to add them to the volume, but like an idiot didn't check the
raidgroup size. Thus when
I added the 3 disks to the volume (vol0) 1 went to raidgroup 0 and the
other 2 went to raidgroup 1 as a parity
drive and a data drive.

This will create a performance problem from what I can tell not mention
another parity drive being used.

I'd like to get these two disks back under raid group 0 if I can.

Suggestions on the best way to do this would be appreciated.

I'm thinking the safest way is to back it up and recreate the volumes and
restore.

Thanks

art hebert
Re: Raid Group question [ In reply to ]
If vol0 = root volume make sure you also copy the /etc/ dir to another volume
and change that volume to be your root volume if you want to use the volcopy
method. If I recall correctly you copy /etc/ to another volume. Change
the new volume to be the root volume. Reboot with this new volume. Volcopy
vol0 data somewhere else. Destroy vol0. Rebuild vol0. Reverse the process
and then reboot.

Been a while since I had to ask a vol other than 0 to be my root volume so
check the procedures with NetApp before following this method. Would hate to
point you down the path of no return.
C-


On Mon, Dec 09, 2002 at 04:59:20PM -0800, Sam Schorr wrote:
> Art,
>
> There is no dynamic way to rebuild raid group sizing, or allocate from one raid group to another. Your statement about the safest way is also the only way that I have ever heard of (well, I guess volcopy would work, copying to another volume, destroying and rebuilding the source, and volcopy back).
>
> --sam
>
> -----Original Message-----
> From: Art Hebert [mailto:art@arzoon.com]
> Sent: Monday, December 09, 2002 4:42 PM
> To: 'toasters@mathworks.com'
> Subject: Raid Group question
>
>
>
>
> I have a current volume (vol0) that has a raidgroup size of 8. I had 3
> spares that weren't being used and
> I wanted to add them to the volume, but like an idiot didn't check the
> raidgroup size. Thus when
> I added the 3 disks to the volume (vol0) 1 went to raidgroup 0 and the
> other 2 went to raidgroup 1 as a parity
> drive and a data drive.
>
> This will create a performance problem from what I can tell not mention
> another parity drive being used.
>
> I'd like to get these two disks back under raid group 0 if I can.
>
> Suggestions on the best way to do this would be appreciated.
>
> I'm thinking the safest way is to back it up and recreate the volumes and
> restore.
>
> Thanks
>
> art hebert

--
------------------------------------------------------------------------------
* Chris Blackmor _______ | "I said it too many times *
* Advanced Micro Devices \____ | | And I still stand firm *
* Phone: (512) 602-1608 /| | | | You get what you put in *
* Fax: (512) 602-5155 | |___| | |And people get what they deserve" *
* Email: chris.blackmor@amd.com |____/ \| | Lyrics from "Only God Knows Why" *
* | By Kid Rock *
------------------------------------------------------------------------------
* My comments are mine, and mine alone. *
------------------------------------------------------------------------------
Re: Raid Group question [ In reply to ]
On Mon, Dec 09, 2002 at 10:49:13PM -0600, Chris Blackmor wrote:
> If vol0 = root volume make sure you also copy the /etc/ dir to another volume
> and change that volume to be your root volume if you want to use the volcopy
> method. If I recall correctly you copy /etc/ to another volume. Change
> the new volume to be the root volume. Reboot with this new volume. Volcopy
> vol0 data somewhere else. Destroy vol0. Rebuild vol0. Reverse the process
> and then reboot.
>
This method works great. I did it last Aug to shrink the size of our
root volume on an F740.

http://now.netapp.com/Knowledgebase/solutionarea.asp?id=3%2E0%2E975846%2E2725449

--
Jeff Bryer bryer@sfu.ca
Systems Administrator (604) 291-4935
Academic Computing, Simon Fraser University
RE: Raid Group question [ In reply to ]
Hi Art,

Firstly it won't create a _serious_ performance problem. WAFL stripes across
all the disks in the volume, regardsless of the number of RAID groups. Yes,
you have another parity disk but you won't have "single drive seek" issues
on the raid group with 1 data drive in it. Remember, we're dealing with a
RAID-aware filesystem here, not dumb block-level RAID.

However, the space consumption will be an issue for you, at least until you
add more drives. Most of my customers run 14 disk RAID groups (MTTDL=~95000
years for the old 18GB drive, I can't imagine it would be worse for the
newer drives).

Yep, you need to backup, recreate volume and restore. Don't forget to
recreate your qtrees manually before you restore.

OR, if you can temporarily borrowe another shelf from your FLNR (friendly
local NetApp reseller), you can create a new volume on the other shelf and
volcopy the data to and from it. Much quicker :o)



-----Original Message-----
From: Art Hebert [mailto:art@arzoon.com]
Sent: Tuesday, 10 December 2002 11:42 AM
To: 'toasters@mathworks.com'
Subject: Raid Group question




I have a current volume (vol0) that has a raidgroup size of 8. I had 3
spares that weren't being used and
I wanted to add them to the volume, but like an idiot didn't check the
raidgroup size. Thus when
I added the 3 disks to the volume (vol0) 1 went to raidgroup 0 and the
other 2 went to raidgroup 1 as a parity
drive and a data drive.

This will create a performance problem from what I can tell not mention
another parity drive being used.

I'd like to get these two disks back under raid group 0 if I can.

Suggestions on the best way to do this would be appreciated.

I'm thinking the safest way is to back it up and recreate the volumes and
restore.

Thanks

art hebert


**** ASI Solutions Disclaimer ****
The material transmitted may contain confidential and/or privileged
material and is intended only for the addressee. If you receive this in
error, please notify the sender and destroy any copies of the material
immediately. ASI will protect your Privacy according to the 10 Privacy
Principles outlined under the new Privacy Act, Dec 2001.

This email is also subject to copyright. Any use of or reliance upon this
material by persons or entities other than the addressee is prohibited.

E-mails may be interfered with, may contain computer viruses or other
defects. Under no circumstances do we accept liability for any loss or
damage which may result from your receipt of this message or any
attachments.
**** END OF MESSAGE ****