Backup file size when migrating from raid5 to raid6?

Backup file size when migrating from raid5 to raid6?

[Garðar Arnarsson]

My raid5 array has gotten a bit big, it's containing total 10 drives
right now (I started out with 3 drives). So I am going to convert it
to raid6 before it gets any bigger.

I am doing a test-run on a virtual machine with virtual drives to see
that everything works flawlessly.

When I tried to convert the array to raid6 I got a error message about
a missing backup-file

mdadm --grow /dev/md0 --raid-devices=5 --level=6

mdadm level of /dev/md0 changed to raid6
mdadm: /dev/md0: Cannot grow - need backup-file
mdadm: aborting level change

I added the backup file and was able to convert the array successfully
after that.

My question is, how big is this backup file going to be? My real raid
array consists of 2tb drives, will the backup file be as big as one
drive in the array, or will it just be few megabytes or gigabytes?
I'm asking because I'm wondering if I need to buy an extra hdd for the
backup file or if the backup file can just be on my OS hdd that has
around 100gb free.

Re: Backup file size when migrating from raid5 to raid6?

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 1899 bytes --]

On Sun, 6 May 2012 10:17:52 +0000 Garðar Arnarsson <gardar@giraffi.net> wrote:

> My raid5 array has gotten a bit big, it's containing total 10 drives
> right now (I started out with 3 drives). So I am going to convert it
> to raid6 before it gets any bigger.
> 
> I am doing a test-run on a virtual machine with virtual drives to see
> that everything works flawlessly.
> 
> When I tried to convert the array to raid6 I got a error message about
> a missing backup-file
> 
> mdadm --grow /dev/md0 --raid-devices=5 --level=6
> 
> mdadm level of /dev/md0 changed to raid6
> mdadm: /dev/md0: Cannot grow - need backup-file
> mdadm: aborting level change
> 
> I added the backup file and was able to convert the array successfully
> after that.
> 
> My question is, how big is this backup file going to be? My real raid
> array consists of 2tb drives, will the backup file be as big as one
> drive in the array, or will it just be few megabytes or gigabytes?
> I'm asking because I'm wondering if I need to buy an extra hdd for the
> backup file or if the backup file can just be on my OS hdd that has
> around 100gb free.

The backup file is a few megabytes. Around 16MB I think.

However if you are likely to add another device in the not too distant future
you can save yourself a bit of time.

If you

 mdadm --grow /dev/md0 --level=6 --layout=preserve

It will just make the new few a 'Q-block' device, containing the extra RAID6
'parity' block for each stripe.  This doesn't require any reshape or or any
backup file and is a lot faster.  All it requires is a normal recovery
operation.

Then when you later add another device you can

  mdadm --grow /dev/md0 --raid-devices=N+1 --layout=normalise

This will convert from the Q-on-the-last-device layout to a more normal
rotated-P-and-Q layout at the same time as adding extra space.

NeilBrown

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: Backup file size when migrating from raid5 to raid6?

[Garðar Arnarsson]

That's an excellent idea, I was going to add another disk for extra
space right after migrating to raid6.

Just to be clear, I'll be running the normalize attribute just once to
straighten the array out right? Or will I have to do it for every
extra drive I add in the future?

And what are the N+1 you mention in --raid-devices=N+1

 Thanks.



2012/5/6 NeilBrown <neilb@suse.de>
>
> On Sun, 6 May 2012 10:17:52 +0000 Garðar Arnarsson <gardar@giraffi.net> wrote:
>
> > My raid5 array has gotten a bit big, it's containing total 10 drives
> > right now (I started out with 3 drives). So I am going to convert it
> > to raid6 before it gets any bigger.
> >
> > I am doing a test-run on a virtual machine with virtual drives to see
> > that everything works flawlessly.
> >
> > When I tried to convert the array to raid6 I got a error message about
> > a missing backup-file
> >
> > mdadm --grow /dev/md0 --raid-devices=5 --level=6
> >
> > mdadm level of /dev/md0 changed to raid6
> > mdadm: /dev/md0: Cannot grow - need backup-file
> > mdadm: aborting level change
> >
> > I added the backup file and was able to convert the array successfully
> > after that.
> >
> > My question is, how big is this backup file going to be? My real raid
> > array consists of 2tb drives, will the backup file be as big as one
> > drive in the array, or will it just be few megabytes or gigabytes?
> > I'm asking because I'm wondering if I need to buy an extra hdd for the
> > backup file or if the backup file can just be on my OS hdd that has
> > around 100gb free.
>
> The backup file is a few megabytes. Around 16MB I think.
>
> However if you are likely to add another device in the not too distant future
> you can save yourself a bit of time.
>
> If you
>
>  mdadm --grow /dev/md0 --level=6 --layout=preserve
>
> It will just make the new few a 'Q-block' device, containing the extra RAID6
> 'parity' block for each stripe.  This doesn't require any reshape or or any
> backup file and is a lot faster.  All it requires is a normal recovery
> operation.
>
> Then when you later add another device you can
>
>  mdadm --grow /dev/md0 --raid-devices=N+1 --layout=normalise
>
> This will convert from the Q-on-the-last-device layout to a more normal
> rotated-P-and-Q layout at the same time as adding extra space.
>
> NeilBrown
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Backup file size when migrating from raid5 to raid6?

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 3088 bytes --]

On Mon, 7 May 2012 00:32:35 +0000 Garðar Arnarsson <gardar@giraffi.net> wrote:

> That's an excellent idea, I was going to add another disk for extra space
> right after migrating to raid6.
> 
> Just to be clear, I'll be running the normalize attribute just once to
> straighten the array out right? Or will I have to do it for every extra
> drive I add in the future?

Just once.

> 
> And what are the N+1 you mention in --raid-devices=N+1

By "N+1" I just meant "1 more than the number of devices currently in the
array".

If you have both new devices ready to go, you just do a single reshape
operation that converts to RAID6 and adds more space.  This does not need a
backup file and is probably the best approach.

If you currently have a 10-drive RAID5 and want a 12-drive RAID6, then

 mdadm --grow /dev/md0 --raid-devices=12 --level=6

is what you want.

NeilBrown


> 
> Thanks.
> 
> 
> 2012/5/6 NeilBrown <neilb@suse.de>
> 
> > On Sun, 6 May 2012 10:17:52 +0000 Garðar Arnarsson <gardar@giraffi.net>
> > wrote:
> >
> > > My raid5 array has gotten a bit big, it's containing total 10 drives
> > > right now (I started out with 3 drives). So I am going to convert it
> > > to raid6 before it gets any bigger.
> > >
> > > I am doing a test-run on a virtual machine with virtual drives to see
> > > that everything works flawlessly.
> > >
> > > When I tried to convert the array to raid6 I got a error message about
> > > a missing backup-file
> > >
> > > mdadm --grow /dev/md0 --raid-devices=5 --level=6
> > >
> > > mdadm level of /dev/md0 changed to raid6
> > > mdadm: /dev/md0: Cannot grow - need backup-file
> > > mdadm: aborting level change
> > >
> > > I added the backup file and was able to convert the array successfully
> > > after that.
> > >
> > > My question is, how big is this backup file going to be? My real raid
> > > array consists of 2tb drives, will the backup file be as big as one
> > > drive in the array, or will it just be few megabytes or gigabytes?
> > > I'm asking because I'm wondering if I need to buy an extra hdd for the
> > > backup file or if the backup file can just be on my OS hdd that has
> > > around 100gb free.
> >
> > The backup file is a few megabytes. Around 16MB I think.
> >
> > However if you are likely to add another device in the not too distant
> > future
> > you can save yourself a bit of time.
> >
> > If you
> >
> >  mdadm --grow /dev/md0 --level=6 --layout=preserve
> >
> > It will just make the new few a 'Q-block' device, containing the extra
> > RAID6
> > 'parity' block for each stripe.  This doesn't require any reshape or or any
> > backup file and is a lot faster.  All it requires is a normal recovery
> > operation.
> >
> > Then when you later add another device you can
> >
> >  mdadm --grow /dev/md0 --raid-devices=N+1 --layout=normalise
> >
> > This will convert from the Q-on-the-last-device layout to a more normal
> > rotated-P-and-Q layout at the same time as adding extra space.
> >
> > NeilBrown
> >
> 
> 
> 


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: Backup file size when migrating from raid5 to raid6?

[Thomas Fjellstrom]

On Sun May 6, 2012, NeilBrown wrote:
> On Mon, 7 May 2012 00:32:35 +0000 Garðar Arnarsson <gardar@giraffi.net> 
wrote:
> 
> > That's an excellent idea, I was going to add another disk for extra space
> > right after migrating to raid6.
> > 
> > Just to be clear, I'll be running the normalize attribute just once to
> > straighten the array out right? Or will I have to do it for every extra
> > drive I add in the future?
> 
> Just once.
> 
> > 
> > And what are the N+1 you mention in --raid-devices=N+1
> 
> By "N+1" I just meant "1 more than the number of devices currently in the
> array".
> 
> If you have both new devices ready to go, you just do a single reshape
> operation that converts to RAID6 and adds more space.  This does not need a
> backup file and is probably the best approach.
> 
> If you currently have a 10-drive RAID5 and want a 12-drive RAID6, then
> 
>  mdadm --grow /dev/md0 --raid-devices=12 --level=6
> 
> is what you want.

I apologize for bringing back a long dead thread, but I've been wondering if 
mdadm does the grow op in this case, in one step? Or does it internally do 
each step separately, doing a reshape with each one?

I've currently got a 7x1TB disk raid5, and have a couple more disks to add and 
I was planning on moving to raid6. I'm hoping to reduce the amount of time the 
array is "reshaping" because I'm a bit paranoid that my bad luck with hard 
drives will decide to hit right then and there.

> NeilBrown
> 
> 
> > 
> > Thanks.
> > 
> > 
> > 2012/5/6 NeilBrown <neilb@suse.de>
> > 
> > > On Sun, 6 May 2012 10:17:52 +0000 Garðar Arnarsson <gardar@giraffi.net>
> > > wrote:
> > >
> > > > My raid5 array has gotten a bit big, it's containing total 10 drives
> > > > right now (I started out with 3 drives). So I am going to convert it
> > > > to raid6 before it gets any bigger.
> > > >
> > > > I am doing a test-run on a virtual machine with virtual drives to see
> > > > that everything works flawlessly.
> > > >
> > > > When I tried to convert the array to raid6 I got a error message about
> > > > a missing backup-file
> > > >
> > > > mdadm --grow /dev/md0 --raid-devices=5 --level=6
> > > >
> > > > mdadm level of /dev/md0 changed to raid6
> > > > mdadm: /dev/md0: Cannot grow - need backup-file
> > > > mdadm: aborting level change
> > > >
> > > > I added the backup file and was able to convert the array successfully
> > > > after that.
> > > >
> > > > My question is, how big is this backup file going to be? My real raid
> > > > array consists of 2tb drives, will the backup file be as big as one
> > > > drive in the array, or will it just be few megabytes or gigabytes?
> > > > I'm asking because I'm wondering if I need to buy an extra hdd for the
> > > > backup file or if the backup file can just be on my OS hdd that has
> > > > around 100gb free.
> > >
> > > The backup file is a few megabytes. Around 16MB I think.
> > >
> > > However if you are likely to add another device in the not too distant
> > > future
> > > you can save yourself a bit of time.
> > >
> > > If you
> > >
> > >  mdadm --grow /dev/md0 --level=6 --layout=preserve
> > >
> > > It will just make the new few a 'Q-block' device, containing the extra
> > > RAID6
> > > 'parity' block for each stripe.  This doesn't require any reshape or or 
any
> > > backup file and is a lot faster.  All it requires is a normal recovery
> > > operation.
> > >
> > > Then when you later add another device you can
> > >
> > >  mdadm --grow /dev/md0 --raid-devices=N+1 --layout=normalise
> > >
> > > This will convert from the Q-on-the-last-device layout to a more normal
> > > rotated-P-and-Q layout at the same time as adding extra space.
> > >
> > > NeilBrown
> > >
> > 
> > 
> > 
> 
> 


-- 
Thomas Fjellstrom
thomas@fjellstrom.ca
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Backup file size when migrating from raid5 to raid6?

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 1924 bytes --]

On Thu, 16 Aug 2012 17:28:13 -0600 Thomas Fjellstrom <thomas@fjellstrom.ca>
wrote:

> On Sun May 6, 2012, NeilBrown wrote:
> > On Mon, 7 May 2012 00:32:35 +0000 Garðar Arnarsson <gardar@giraffi.net> 
> wrote:
> > 
> > > That's an excellent idea, I was going to add another disk for extra space
> > > right after migrating to raid6.
> > > 
> > > Just to be clear, I'll be running the normalize attribute just once to
> > > straighten the array out right? Or will I have to do it for every extra
> > > drive I add in the future?
> > 
> > Just once.
> > 
> > > 
> > > And what are the N+1 you mention in --raid-devices=N+1
> > 
> > By "N+1" I just meant "1 more than the number of devices currently in the
> > array".
> > 
> > If you have both new devices ready to go, you just do a single reshape
> > operation that converts to RAID6 and adds more space.  This does not need a
> > backup file and is probably the best approach.
> > 
> > If you currently have a 10-drive RAID5 and want a 12-drive RAID6, then
> > 
> >  mdadm --grow /dev/md0 --raid-devices=12 --level=6
> > 
> > is what you want.
> 
> I apologize for bringing back a long dead thread, but I've been wondering if 
> mdadm does the grow op in this case, in one step? Or does it internally do 
> each step separately, doing a reshape with each one?
> 
> I've currently got a 7x1TB disk raid5, and have a couple more disks to add and 
> I was planning on moving to raid6. I'm hoping to reduce the amount of time the 
> array is "reshaping" because I'm a bit paranoid that my bad luck with hard 
> drives will decide to hit right then and there.
> 

If you change the level and the number of devices in the one command, and
there are already enough spares present to satisfy the requirements of the
new size, then it will only perform a single reshape operation which will use
the minimal amount of time.

NeilBrown


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: Backup file size when migrating from raid5 to raid6?

[Thomas Fjellstrom]

On Thu Aug 16, 2012, NeilBrown wrote:
> On Thu, 16 Aug 2012 17:28:13 -0600 Thomas Fjellstrom <thomas@fjellstrom.ca>
> 
> wrote:
> > On Sun May 6, 2012, NeilBrown wrote:
> > > On Mon, 7 May 2012 00:32:35 +0000 Garðar Arnarsson <gardar@giraffi.net>
> > 
> > wrote:
> > > > That's an excellent idea, I was going to add another disk for extra
> > > > space right after migrating to raid6.
> > > > 
> > > > Just to be clear, I'll be running the normalize attribute just once
> > > > to straighten the array out right? Or will I have to do it for every
> > > > extra drive I add in the future?
> > > 
> > > Just once.
> > > 
> > > > And what are the N+1 you mention in --raid-devices=N+1
> > > 
> > > By "N+1" I just meant "1 more than the number of devices currently in
> > > the array".
> > > 
> > > If you have both new devices ready to go, you just do a single reshape
> > > operation that converts to RAID6 and adds more space.  This does not
> > > need a backup file and is probably the best approach.
> > > 
> > > If you currently have a 10-drive RAID5 and want a 12-drive RAID6, then
> > > 
> > >  mdadm --grow /dev/md0 --raid-devices=12 --level=6
> > > 
> > > is what you want.
> > 
> > I apologize for bringing back a long dead thread, but I've been wondering
> > if mdadm does the grow op in this case, in one step? Or does it
> > internally do each step separately, doing a reshape with each one?
> > 
> > I've currently got a 7x1TB disk raid5, and have a couple more disks to
> > add and I was planning on moving to raid6. I'm hoping to reduce the
> > amount of time the array is "reshaping" because I'm a bit paranoid that
> > my bad luck with hard drives will decide to hit right then and there.
> 
> If you change the level and the number of devices in the one command, and
> there are already enough spares present to satisfy the requirements of the
> new size, then it will only perform a single reshape operation which will
> use the minimal amount of time.

Ah. Thank you for clarifying that for me :) just running the two new drives 
through a bit of a stress test. Then I'll get to expanding the array.

Thanks.

> NeilBrown


-- 
Thomas Fjellstrom
thomas@fjellstrom.ca
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Backup file size when migrating from raid5 to raid6?

[John Robinson]

On 17/08/2012 00:28, Thomas Fjellstrom wrote:
[...]
> I apologize for bringing back a long dead thread, but I've been wondering if
> mdadm does the grow op in this case, in one step? Or does it internally do
> each step separately, doing a reshape with each one?
>
> I've currently got a 7x1TB disk raid5, and have a couple more disks to add and
> I was planning on moving to raid6. I'm hoping to reduce the amount of time the
> array is "reshaping" because I'm a bit paranoid that my bad luck with hard
> drives will decide to hit right then and there.

If you have a non-degraded 7-disc array, and add two more discs as 
spares, then issue a `mdadm --grow /dev/mdN -l 6 -n 9,` then yes I 
believe it will do it in one pass. There are circumstances where you 
might want to `echo frozen > /sys/block/md/mdN/sync_action` before 
issuing a series of mdadm commands, but I don't think this is one of them.

But if you're worried about your drives, make sure they're clean before 
you start. If you already run weekly scrubs and SMART tests, and they 
come back OK, you should be fine. If you don't, look at your SMART stats 
to see if any of the drives are already in trouble, and rather than 
running a scrub or long SMART tests, just `dd if=/dev/sdX of=/dev/null 
bs=65536` all your drives (though maybe not all at once depending on the 
controllers). Doing that won't kick them out of your raid5 unless a 
drive dies or otherwise totally locks up, and will give you a heads up 
if there may be trouble ahead.

Cheers,

John.

Re: Backup file size when migrating from raid5 to raid6?

[Boyan Alexiev]

If You have array with 9 devices and add another one You may at same time
grow array and change raid level. In this case You not need backup file.
Please try this:
mdadm --grow /dev/md0 --raid-devices=10 --level=6 

Re: Backup file size when migrating from raid5 to raid6?

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 548 bytes --]

On Wed, 17 Jul 2013 13:01:15 +0000 (UTC) Boyan Alexiev <alexiev@resonance.bg>
wrote:

> If You have array with 9 devices and add another one You may at same time
> grow array and change raid level. In this case You not need backup file.
> Please try this:
> mdadm --grow /dev/md0 --raid-devices=10 --level=6 
> 

You can only skip the backup file if you  are increasing the number of *data*
devices.  I don't think your example does that.

(However with mdadm-3.3 this problem goes away if you have a new enough
kernel).

NeilBrown

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]