RAID-5 implementation questions

RAID-5 implementation questions

[Phil Karn]

Are there any papers documenting the implementation of the Linux RAID
subsystem? I'm interested in some of the details of how RAID-5 works.

I've never seen a virgin disk drive from the factory that wasn't all
0's. Creating a RAID array on a set of such drives triggers an initial
rebuild that simply writes lots zeroes on lots of zeroes. With disks now
pushing past 2 TB, this can easily take half a day.

Except for the admittedly somewhat useful side effect of scanning the
disks for bad sectors, all this activity seems rather unnecessary. Is
there a way to create a RAID-5 (or any other RAID level) array so that
it will immediately come up without an initial rebuild?

File systems generally don't read disk blocks that they haven't already
written. So even when you build a RAID array from drives with old data,
I can't see how skipping the initial rebuild can cause any real harm.
The first write to any block causes the RAID system to initialize the
parity in that stripe, thus making it possible to regenerate that block
in case of a drive failure.

During the initial rebuild of a RAID-5 array, /proc/mdstat suggests that
the array is operating in degraded mode and the last drive in the array
is being rebuilt. Is this true, i.e., are all the rebuild writes going
to that last drive?

How does a rebuilding RAID-5 array handle a read or write operation when
it lands on the "broken" drive? Does it depend on whether the block is
before or after the rebuild pointer?

Thanks much,

Phil

Re: RAID-5 implementation questions

[Mikael Abrahamsson]

On Fri, 3 Dec 2010, Phil Karn wrote:

> Except for the admittedly somewhat useful side effect of scanning the 
> disks for bad sectors, all this activity seems rather unnecessary. Is 
> there a way to create a RAID-5 (or any other RAID level) array so that 
> it will immediately come up without an initial rebuild?

"--assume-clean".

> File systems generally don't read disk blocks that they haven't already
> written. So even when you build a RAID array from drives with old data,
> I can't see how skipping the initial rebuild can cause any real harm.
> The first write to any block causes the RAID system to initialize the
> parity in that stripe, thus making it possible to regenerate that block
> in case of a drive failure.

Some raid implementations won't read/write to all drives, but might 
instead read the block being written to, and the parity block, then write 
the new block and recalculate the parity, thus not read/writing to all 
blocks. If this is the case, if the parity is wrong, it'll still be wrong 
after the operation, thus you don't have any redundancy.

Doing a rebuild when creating the array is something I'd only skip if I 
was doing lab work, never in production. I use raid for redundancy, thus I 
want to make sure everything is ok and it doesn't matter to me if it takes 
half a day.

-- 
Mikael Abrahamsson    email: swmike@swm.pp.se

Re: RAID-5 implementation questions

[Phil Karn]

On 12/3/10 2:02 AM, Mikael Abrahamsson wrote:

> "--assume-clean".

Thanks.

> Some raid implementations won't read/write to all drives, but might
> instead read the block being written to, and the parity block, then
> write the new block and recalculate the parity, thus not read/writing to
> all blocks. If this is the case, if the parity is wrong, it'll still be
> wrong after the operation, thus you don't have any redundancy.

Good point. That had occurred to me too but I didn't know if Linux did
that. I can see how one might dynamically pick one way or the other
depending on how much of the stripe is already in the buffer cache.

> Doing a rebuild when creating the array is something I'd only skip if I
> was doing lab work, never in production. I use raid for redundancy, thus
> I want to make sure everything is ok and it doesn't matter to me if it
> takes half a day.

I hear you. But I think an important special case is when you're
initially loading a new RAID-5 array from an existing (typically
smaller) file system that will then be replaced by the new array.

Why not let the new array work something like a RAID-0, leaving the
parity blocks unwritten until you're finished loading the array? Then
pass through the array writing all the parity blocks with the final
data. If a drive fails in the new array before you're done, you still
have all your original data; you haven't lost anything.

Ultimately, RAID-5 in software is always going to be at least somewhat
vulnerable because of the lack of an atomic (all or none) committed
write of all the blocks in a stripe. This might silently corrupt an old,
stable file in a way that you won't notice until a drive fails and you
don't have the redundancy you thought you had to reconstruct it. can
accept losing whatever files I was writing at the time of a crash, but
silent corruption of an old and stable file seems far more insidious. I
do periodically run checkarray to ensure that the parities are
consistent, but this takes a long time and seems inelegant somehow.
Maybe we need software ECC on all data so that one doesn't have to rely
on the drive itself to detect errors.

Thanks,

Phil

Re: RAID-5 implementation questions

[Neil Brown]

On Fri, 03 Dec 2010 00:49:52 -0800 Phil Karn <karn@ka9q.net> wrote:

> Are there any papers documenting the implementation of the Linux RAID
> subsystem? I'm interested in some of the details of how RAID-5 works.

I would suggest
  man mdadm
and
  man md

That should answer at least some of your questions.

Then try  http://raid.wiki.kernel.org/

If you have further questions after that, please ask.

NeilBrown


> 
> I've never seen a virgin disk drive from the factory that wasn't all
> 0's. Creating a RAID array on a set of such drives triggers an initial
> rebuild that simply writes lots zeroes on lots of zeroes. With disks now
> pushing past 2 TB, this can easily take half a day.
> 
> Except for the admittedly somewhat useful side effect of scanning the
> disks for bad sectors, all this activity seems rather unnecessary. Is
> there a way to create a RAID-5 (or any other RAID level) array so that
> it will immediately come up without an initial rebuild?
> 
> File systems generally don't read disk blocks that they haven't already
> written. So even when you build a RAID array from drives with old data,
> I can't see how skipping the initial rebuild can cause any real harm.
> The first write to any block causes the RAID system to initialize the
> parity in that stripe, thus making it possible to regenerate that block
> in case of a drive failure.
> 
> During the initial rebuild of a RAID-5 array, /proc/mdstat suggests that
> the array is operating in degraded mode and the last drive in the array
> is being rebuilt. Is this true, i.e., are all the rebuild writes going
> to that last drive?
> 
> How does a rebuilding RAID-5 array handle a read or write operation when
> it lands on the "broken" drive? Does it depend on whether the block is
> before or after the rebuild pointer?
> 
> Thanks much,
> 
> Phil
> --
> 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