From: Zdenek Kaspar <zkaspar82@gmail.com>
To: Jessie Evangelista <jessie.evangelista@gmail.com>
Cc: linux-raid@vger.kernel.org
Subject: Re: raid10n2/xfs setup guidance on write-cache/barrier
Date: Sat, 17 Mar 2012 23:10:43 +0100 [thread overview]
Message-ID: <4F650BE3.5010305@gmail.com> (raw)
In-Reply-To: <CAA8mOyDKrWg0QUEHxcD4ocXXD42nJu0TG+sXjC4j2RsigHTcmw@mail.gmail.com>
Dne 15.3.2012 1:30, Jessie Evangelista napsal(a):
> I want to create a raid10,n2 using 3 1TB SATA drives.
> I want to create an xfs filesystem on top of it.
> The filesystem will be used as NFS/Samba storage.
>
> mdadm --zero /dev/sdb1 /dev/sdc1 /dev/sdd1
> mdadm -v --create /dev/md0 --metadata=1.2 --assume-clean
> --level=raid10 --chunk 256 --raid-devices=3 /dev/sdb1 /dev/sdc1
> /dev/sdd1
> mkfs -t xfs -l lazy-count=1,size=128m -f /dev/md0
> mount -t xfs -o barrier=1,logbsize=256k,logbufs=8,noatime /dev/md0
> /mnt/raid10xfs
>
> Will my files be safe even on sudden power loss? Is barrier=1 enough?
> Do i need to disable the write cache?
> with: hdparm -W0 /dev/sdb /dev/sdc /dev/sdd
>
> I tried it but performance is horrendous.
>
> Am I better of with ext4? Data safety/integrity is the priority and
> optimization affecting it is not acceptable.
>
> Thanks and any advice/guidance would be appreciated
> --
> 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
I think today you're most safe with old ext3. Maybe data=journal is not
best idea, because it has small user-base. The more aggressive caching
and features leading to awesome performance will bite you harder by
power loss or software bugs.
Limit power loss with UPS unit. You need to cleanly shutdown the system
when UPS reaches it's low level capacity. That's mandatory IMO and still
inexpensive. Next you can think about multiple PSU/UPS units...
But just think in a way that you will never make it 100% safe, so use
damn good backups!
HTH, Z.
prev parent reply other threads:[~2012-03-17 22:10 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-15 0:30 raid10n2/xfs setup guidance on write-cache/barrier Jessie Evangelista
2012-03-15 5:38 ` Stan Hoeppner
2012-03-15 12:06 ` Jessie Evangelista
2012-03-15 14:07 ` Peter Grandi
2012-03-15 14:07 ` Peter Grandi
2012-03-15 15:25 ` keld
2012-03-15 15:25 ` keld
2012-03-15 16:52 ` Jessie Evangelista
2012-03-15 16:52 ` Jessie Evangelista
2012-03-15 17:15 ` keld
2012-03-15 17:15 ` keld
2012-03-15 17:40 ` keld
2012-03-15 17:40 ` keld
2012-03-15 16:18 ` Jessie Evangelista
2012-03-15 16:18 ` Jessie Evangelista
2012-03-15 23:00 ` Peter Grandi
2012-03-15 23:00 ` Peter Grandi
2012-03-16 3:36 ` Jessie Evangelista
2012-03-16 3:36 ` Jessie Evangelista
2012-03-16 11:06 ` Michael Monnerie
2012-03-16 11:06 ` Michael Monnerie
2012-03-16 12:21 ` Peter Grandi
2012-03-16 12:21 ` Peter Grandi
2012-03-16 17:15 ` Brian Candler
2012-03-16 17:15 ` Brian Candler
2012-03-17 15:35 ` Peter Grandi
2012-03-17 15:35 ` Peter Grandi
2012-03-17 21:39 ` raid10n2/xfs setup guidance on write-cache/barrier (GiB alignment) Zdenek Kaspar
2012-03-17 21:39 ` Zdenek Kaspar
2012-03-18 0:08 ` Peter Grandi
2012-03-18 0:08 ` Peter Grandi
2012-03-26 19:50 ` raid10n2/xfs setup guidance on write-cache/barrier Martin Steigerwald
2012-03-17 4:21 ` NOW:Peter goading Dave over delaylog - WAS: " Stan Hoeppner
2012-03-17 22:34 ` Dave Chinner
2012-03-18 2:09 ` Peter Grandi
2012-03-18 2:09 ` Peter Grandi
2012-03-18 11:25 ` Peter Grandi
2012-03-18 11:25 ` Peter Grandi
2012-03-18 14:00 ` Christoph Hellwig
2012-03-18 14:00 ` Christoph Hellwig
2012-03-18 19:17 ` Peter Grandi
2012-03-18 19:17 ` Peter Grandi
2012-03-19 9:07 ` Stan Hoeppner
2012-03-19 9:07 ` Stan Hoeppner
2012-03-20 12:34 ` Jessie Evangelista
2012-03-20 12:34 ` Jessie Evangelista
2012-03-18 18:08 ` Stan Hoeppner
2012-03-18 18:08 ` Stan Hoeppner
2012-03-22 21:26 ` Peter Grandi
2012-03-22 21:26 ` Peter Grandi
2012-03-23 5:10 ` Stan Hoeppner
2012-03-23 5:10 ` Stan Hoeppner
2012-03-23 22:48 ` Martin Steigerwald
2012-03-24 1:27 ` Peter Grandi
2012-03-24 16:27 ` GNU 'tar', Schilling's 'tar', write-cache/barrier Peter Grandi
2012-03-24 17:11 ` Brian Candler
2012-03-24 18:35 ` Peter Grandi
2012-03-16 12:25 ` raid10n2/xfs setup guidance on write-cache/barrier Stan Hoeppner
2012-03-16 18:01 ` Jon Nelson
2012-03-16 18:03 ` Jon Nelson
2012-03-16 19:28 ` Peter Grandi
2012-03-16 19:28 ` Peter Grandi
2012-03-17 0:02 ` Stan Hoeppner
2012-03-17 0:02 ` Stan Hoeppner
2012-03-17 22:10 ` Zdenek Kaspar [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4F650BE3.5010305@gmail.com \
--to=zkaspar82@gmail.com \
--cc=jessie.evangelista@gmail.com \
--cc=linux-raid@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.