On 2017-01-18 09:21, Steven Hum wrote:
Added 2 drives to my RAID10, then ran btrfs balance. The system appears
to have crashed after several hours (I was ssh'd in at the time on my
local network). When I reboot the Arch system, I ran btrfs check and no
errors were reported.
However, attempting to mount the vol hangs the system. Trying to mount
the system in recovery mode also hangs the system.
───── uname -a ~
Linux luna 4.8.13-1-ARCH #1 SMP PREEMPT Fri Dec 9 07:24:34 CET 2016 x86_64 GNU/Linux
───── btrfs --version ~
btrfs-progs v4.9
───── btrfs fi show ~
Label: 'Archlinux' uuid: e5ee12ac-7fb3-4e73-b9b2-a5cfb35145ae
Total devices 1 FS bytes used 66.41GiB
devid 1 size 111.24GiB used 111.24GiB path /dev/sda2
Label: 'Share' uuid: 9102bb33-49ad-4baf-a88d-1c7622bff938
Total devices 6 FS bytes used 4.20TiB
devid 1 size 2.73TiB used 1.90TiB path /dev/sdc
devid 2 size 2.73TiB used 1.90TiB path /dev/sdd
devid 3 size 2.73TiB used 1.90TiB path /dev/sde
devid 4 size 2.73TiB used 1.90TiB path /dev/sdb
devid 5 size 2.73TiB used 440.34GiB path /dev/sdf
devid 6 size 2.73TiB used 440.34GiB path /dev/sdg
───── btrfs fi df /home # Replace /home with the mount point of your btrfs-filesystem ~
Data, single: total=109.22GiB, used=64.92GiB
System, single: total=4.00MiB, used=16.00KiB
Metadata, single: total=2.02GiB, used=1.48GiB
GlobalReserve, single: total=179.64MiB, used=0.00B
The dmesg log when attempting to mount the vol can be found at url
http://cwillu.com:8080/108.63.177.210/1
Any ideas what can be done to be able to mount the volume and complete
the balancing?
Based on the dmesg, something is screwed up either with your filesystem,
your hardware, or your kernel. Based on past experience, I'm inclined
to blame things in that order when BTRFS is involved. Other than the
OOPS (which probably was the ultimate cause of the other BUG splats), I
don't really see anything particularly worrying in the log (the only
significant errors I saw were ACPI related, but I see errors like those
on lots of systems with no ill effects). The first thing I'd suggest
trying is mounting with 'skip_balance' in the mount options. This will
prevent BTRFS from trying to start the balance operation again, and if
it's the balance itself that's choking, should make the filesystem
mountable (as long as you mount read-write, you only need this option
once, but I would generally suggest just having it as the default for
exactly the type of situation you've run into). If that works, and you
get another crash when trying to balance again, then it's probably a
filesystem issue or kernel bug, unless you are seeing other odd storage
related issues, in which case it's probably a hardware problem. If the
new balance doesn't crash, then it was probably a hardware error (most
likely transient).
NOTE: the wiki link for mailing list subscribe instructions links to
a page indicating web subscriptions are no longer handled with no
instructions how do subscribe manually -- presumably by sending
a subscribe email request. Would appreciate a reply or cc to this email
(and instructions how to subscribe to the mailing list).
Subscriptions are handled just like LKML. Click the subscribe link
here: http://vger.kernel.org/vger-lists.html#linux-btrfs and it should
open your e-mail client with the message pre-loaded and ready to send.
If things go well, you'll get a message back to confirm the subscription
with instructions in it on how to do so.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
