4.1.6 gentoo-hardened: Hang during rename

4.1.6 gentoo-hardened: Hang during rename

[Kenneth Lakin]

[-- Attachment #1: Type: text/plain, Size: 1155 bytes --]

Hey. It looks like I'm being bitten by:
http://article.gmane.org/gmane.comp.file-systems.btrfs/44987/ except I'm
hitting it during rename, rather than unlink. Dmesg spew here:
http://pastebin.ca/3137198 (That pastebin post includes the -unrelated-
source of the W taint.)

I'm running on Gentoo hardened-sources 4.1.6 on x86 (rather than amd64).
The affected volume is mounted with
"rw,relatime,ssd,discard,space_cache,autodefrag" and -AFAIK- has no
subvolumes.

emerge is the blocked program and is the Gentoo package manager. This
lockup happens occasionally during the "Updating Portage cache" step of
the emerge sync operation. The emerge task is unkillable and is blocked
in D status. Attempts to re-run the cache update task have it hang at
what appears to be the same place. Only a reboot can "fix" the problem.
The rest of the system works just fine; the only thing that appears to
be blocked is the Portage cache update.

I couldn't find an open bug for this, but my Google-fu may be weak
today. Did I miss the bug report, or is this a currently unknown or
unconfirmed issue? Let me know if I can help with diagnostics.



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: 4.1.6 gentoo-hardened: Hang during rename

[Kenneth Lakin]

[-- Attachment #1: Type: text/plain, Size: 844 bytes --]

Apologies, here is the rest of the info I should have included in my
first post:

# uname -a
Linux campstove 4.1.6-hardened #2 SMP PREEMPT Thu Aug 27 08:08:50 PDT
2015 i686 Intel(R) Core(TM) Duo CPU L2400 @ 1.66GHz GenuineIntel GNU/Linux
# btrfs fi show
Label: 'root'  uuid: 6635e5f8-550d-4a62-8448-a68f6b93598d
  Total devices 1 FS bytes used 23.48GiB
  devid    1 size 37.17GiB used 35.18GiB path /dev/mapper/campstovevg-root

Label: none  uuid: 58acea0c-bf02-456f-b068-6f8f6bd0ca9f
  Total devices 1 FS bytes used 36.49GiB
  devid    1 size 41.49GiB used 38.02GiB path /dev/mapper/_dev_dm_1

btrfs-progs v4.1.2
# btrfs fi df /
Data, single: total=32.40GiB, used=22.03GiB
System, single: total=32.00MiB, used=16.00KiB
Metadata, single: total=2.75GiB, used=1.46GiB
GlobalReserve, single: total=512.00MiB, used=0.00B




[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: 4.1.6 gentoo-hardened: Hang during rename

[Filipe David Manana]

On Sat, Aug 29, 2015 at 9:23 PM, Kenneth Lakin <kennethlakin@gmail.com> wrote:
> Hey. It looks like I'm being bitten by:
> http://article.gmane.org/gmane.comp.file-systems.btrfs/44987/ except I'm
> hitting it during rename, rather than unlink. Dmesg spew here:
> http://pastebin.ca/3137198 (That pastebin post includes the -unrelated-
> source of the W taint.)
>
> I'm running on Gentoo hardened-sources 4.1.6 on x86 (rather than amd64).
> The affected volume is mounted with
> "rw,relatime,ssd,discard,space_cache,autodefrag" and -AFAIK- has no
> subvolumes.
>
> emerge is the blocked program and is the Gentoo package manager. This
> lockup happens occasionally during the "Updating Portage cache" step of
> the emerge sync operation. The emerge task is unkillable and is blocked
> in D status. Attempts to re-run the cache update task have it hang at
> what appears to be the same place. Only a reboot can "fix" the problem.
> The rest of the system works just fine; the only thing that appears to
> be blocked is the Portage cache update.
>
> I couldn't find an open bug for this, but my Google-fu may be weak
> today. Did I miss the bug report, or is this a currently unknown or
> unconfirmed issue? Let me know if I can help with diagnostics.

The fix landed in 4.2-rc:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=6ca0709756710c47ec604dd08b9fc45929d36390

And it's not in 4.1 nor any other older releases.
>
>



-- 
Filipe David Manana,

"Reasonable men adapt themselves to the world.
 Unreasonable men adapt the world to themselves.
 That's why all progress depends on unreasonable men."