linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrei Borzenkov <arvidjaar@gmail.com>
To: Pete <pete@petezilla.co.uk>
Cc: Linux BTRFS Mailinglist <linux-btrfs@vger.kernel.org>
Subject: Re: Balance conversion to metadata RAID1, data RAID1 leaves some metadata as DUP
Date: Mon, 9 Sep 2019 11:28:00 +0300	[thread overview]
Message-ID: <CAA91j0W1V0oRTKNy8BYDBbUbZEfbTZaxpqQnUm2tZBGxV1nV+Q@mail.gmail.com> (raw)
In-Reply-To: <eda2fb2a-c657-5e4a-8d08-1bf57b57dd3b@petezilla.co.uk>

On Mon, Sep 9, 2019 at 11:03 AM Pete <pete@petezilla.co.uk> wrote:
>
> I recently recovered created a fresh filesystem on one disk and
> recovered from backups with data as SINGLE and metadata as DUP.  I added
> a second disk yesterday and ran a balance with -dconvert=raid1
> -mconvert=raid1.  I did reboot during the process for a couple of
> reasons, putting the sides on the PC case, putting it back under the
> desk and I updated the kernel from 5.3.9 to 5.2.13 at some point during
> this process. Balance resumed as one would expect.  Balance has now
> completed:
>
> root@phoenix:~# btrfs balance status /home_data
> No balance found on '/home_data'
>
> However, some metadata remains as DUP which does not seem right:
>
> root@phoenix:~# btrfs fi usage  /home_data/
> Overall:
>     Device size:                  10.92TiB
>     Device allocated:              4.69TiB
>     Device unallocated:            6.23TiB
>     Device missing:                  0.00B
>     Used:                          4.61TiB
>     Free (estimated):              3.15TiB      (min: 3.15TiB)
>     Data ratio:                       2.00
>     Metadata ratio:                   2.00
>     Global reserve:              512.00MiB      (used: 0.00B)
>
> Data,RAID1: Size:2.34TiB, Used:2.30TiB
>    /dev/mapper/data_disk_ESFH      2.34TiB
>    /dev/mapper/data_disk_EVPC      2.34TiB
>
> Metadata,RAID1: Size:7.00GiB, Used:4.48GiB
>    /dev/mapper/data_disk_ESFH      7.00GiB
>    /dev/mapper/data_disk_EVPC      7.00GiB
>
> Metadata,DUP: Size:1.00GiB, Used:257.22MiB
>    /dev/mapper/data_disk_ESFH      2.00GiB
>
> System,RAID1: Size:32.00MiB, Used:368.00KiB
>    /dev/mapper/data_disk_ESFH     32.00MiB
>    /dev/mapper/data_disk_EVPC     32.00MiB
>
> Unallocated:
>    /dev/mapper/data_disk_ESFH      3.11TiB
>    /dev/mapper/data_disk_EVPC      3.11TiB
> root@phoenix:~#
>
> root@phoenix:~# btrfs --version
> btrfs-progs v5.1
>
>
> I presume running another balance will fix this, but surely all metadata
> should have been converted?  Is there a way to only balance the DUP
> metadata?


btrfs balance start -m convert=raid1,soft

should do it.

      parent reply	other threads:[~2019-09-09  8:28 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-08  7:09 Balance conversion to metadata RAID1, data RAID1 leaves some metadata as DUP Pete
2019-09-08  7:57 ` Holger Hoffstätte
2019-09-08  9:18   ` Pete
2019-09-08  9:48     ` Holger Hoffstätte
2019-09-09  8:28 ` Andrei Borzenkov [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=CAA91j0W1V0oRTKNy8BYDBbUbZEfbTZaxpqQnUm2tZBGxV1nV+Q@mail.gmail.com \
    --to=arvidjaar@gmail.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=pete@petezilla.co.uk \
    /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).