From: Patrick Plenefisch <simonpatp@gmail.com>
To: linux-lvm@lists.linux.dev
Subject: error while submitting device barriers: BTRFS+LVM read-only on newer kernels
Date: Tue, 20 Feb 2024 22:13:09 -0500 [thread overview]
Message-ID: <CAOCpoWdaXbNw8OhW5vqN2HThzwd+sV-HJ8CgGZqyWGBY+gVinA@mail.gmail.com> (raw)
This is a very weird error I'm getting. I have a system that I'm
migrating from Debian 11 (5.10 kernel) to Debian 12 (6.1 kernel). As a
part of this process, I'm dual booting both systems, with a LVM pool
that contains three relevant LV's with three BTRFS partitions. Two of
these BTRFS partitions (single) are working just fine. The other one
(also single) only works on 5.10 kernels. Whenever I boot it on 6.1
kernels, I can mount it just fine until I trigger a write, either direct, or
`atime`. Then the BTRFS partition is locked `ro` until
a reboot with this in dmesg:
BTRFS info (device dm-75): first mount of filesystem
00d78dac-3576-435d-b575-61c3d951ff66
BTRFS info (device dm-75): using crc32c (crc32c-intel) checksum algorithm
BTRFS info (device dm-75): disk space caching is enabled
BTRFS info: devid 1 device path /dev/mapper/lvm-brokenDisk changed to
/dev/dm-75 scanned by (udev-worker) (11442)
BTRFS info: devid 1 device path /dev/dm-75 changed to
/dev/mapper/lvm-brokenDisk scanned by (udev-worker) (11442)
BTRFS error (device dm-75): bdev /dev/mapper/lvm-brokenDisk errs: wr
0, rd 0, flush 1, corrupt 0, gen 0
BTRFS warning (device dm-75): chunk 13631488 missing 1 devices, max
tolerance is 0 for writable mount
BTRFS: error (device dm-75) in write_all_supers:4379: errno=-5 IO
failure (errors while submitting device barriers.)
BTRFS info (device dm-75: state E): forced readonly
BTRFS warning (device dm-75: state E): Skipping commit of aborted transaction.
BTRFS: error (device dm-75: state EA) in cleanup_transaction:1992:
errno=-5 IO failure
The only thing I can seem to find online is posts warning that you
should backup your data. But that clearly isn't the case here, because
none of the devices in the lvm pool are giving smart warnings or
reallocation counts. Further, reading is fine, I can read the whole
disk. Finally, I can just reboot back to the 5.10 kernel and it just
works again. 6.5bpo kernel also exhibits 6.1 behavior
I had originally thought this was a btrfs issue, but peeling back the
layers with people in btrfs IRC, I reduced it to be a LVM/DM error, I
think.
LVM setup is 4 HDD in a LVM VG. A Raid1 LV is created, and a nested pv
is created a second VG is created using that LV/PV, and a final LV is
created on the upper VG, That LV has the btrfs filesystem. However,
there is something else going on, because I have 2 other partitions in
the exact same layout, also with btrfs that aren't exhibiting this
behavior. What could be going on here?
Patrick
next reply other threads:[~2024-02-21 3:13 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-21 3:13 Patrick Plenefisch [this message]
2024-02-25 3:59 ` error while submitting device barriers: BTRFS+LVM read-only on newer kernels Patrick Plenefisch
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=CAOCpoWdaXbNw8OhW5vqN2HThzwd+sV-HJ8CgGZqyWGBY+gVinA@mail.gmail.com \
--to=simonpatp@gmail.com \
--cc=linux-lvm@lists.linux.dev \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).