* Re: N-Way (traditional) RAID-1 development status
2015-10-20 3:13 N-Way (traditional) RAID-1 development status james harvey
@ 2015-10-20 12:45 ` Austin S Hemmelgarn
0 siblings, 0 replies; 2+ messages in thread
From: Austin S Hemmelgarn @ 2015-10-20 12:45 UTC (permalink / raw)
To: james harvey, linux-btrfs
[-- Attachment #1: Type: text/plain, Size: 1476 bytes --]
On 2015-10-19 23:13, james harvey wrote:
> Wanted to see if there's active development on N-Way (traditional) RAID-1.
>
> By this, I mean that RAID-1 across "n" disks traditionally means "n"
> copies of data, but btrfs currently implements RAID-1 as "2" copies of
> data. So, unlike traditional RAID-1, losing 2 drives in a many drive
> array might cause data loss. It's been mentioned before that N-Way
> (traditional) RAID-1 should eventually be implemented through btrfs.
Before I go any further, I would like to point out that most traditional
hardware RAID controllers only support two disks for RAID-1
configurations, and the only one I've seen that doesn't does RAID-1 like
BTRFS (two copies total of each block). The whole 'RAID-1 on n disks is
n copies' thing actually started with Linux's MD-RAID subsystem, and
just happens to have been popular enough that other software RAID
implementations adopted those semantics as well.
OK, with that historical rant out of the way, back to your actual
question: Based on what has been said in the past, N-way replication is
planned for development after the raid56 code is considered stable.
Based on that and past experience, I'd say to expect it no sooner than
4.8, or if we're _really_ lucky, 4.7. On top of that, based on past
experience with BTRFS\, I would not suggest actually trying it for
anything other than testing until at least one release after the initial
release.
[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 3019 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread