From: Brian Rogers <brian@xyzw.org>
To: The development of BTRFS <linux-btrfs@vger.kernel.org>
Subject: RAID0 limiting disk utilization
Date: Wed, 27 Oct 2010 23:55:22 -0700 [thread overview]
Message-ID: <4CC91E5A.7080503@xyzw.org> (raw)
I noticed that if I have single-device allocation for data in a
multi-device btrfs filesystem, a balance operation will convert the data
to RAID0. This is true even if '-d single' is specified explicitly when
creating the filesystem. Then it wants to continue using RAID0 for
future data allocations, and I run out of space once there's no longer
two drives with space available.
I hit this problem when adding to an existing single-device filesystem.
I ran a balance to mirror my metadata on multiple drives, and that had
the side-effect of not letting me use all my space, since one device is
considerably bigger than the other two combined.
$ sudo btrfs f show
Label: 'Aux' uuid: 35a08a97-4f96-4549-b7c0-18380bbf00e0
Total devices 3 FS bytes used 510.78GB
devid 3 size 133.09GB used 131.01GB path /dev/sda4
devid 1 size 465.76GB used 310.01GB path /dev/sdb1
devid 2 size 74.52GB used 73.00GB path /dev/sdc1
Btrfs Btrfs v0.19
$ btrfs f df /aux
Data, RAID0: total=357.00GB, used=357.00GB
Data: total=153.00GB, used=152.91GB
System, RAID1: total=8.00MB, used=56.00KB
System: total=4.00MB, used=0.00
Metadata, RAID1: total=2.00GB, used=896.57MB
I think btrfs should fall back to single-device allocation when striped
allocation fails. Does that sound like a reasonable policy? Without such
a change, there can be a lot of reported free space that can't be used.
reply other threads:[~2010-10-28 6:55 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=4CC91E5A.7080503@xyzw.org \
--to=brian@xyzw.org \
--cc=linux-btrfs@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).