Hi,
with 3.14-rc4 my box does not finish initializing the LVM array. It
hangs within the call to rw_header() on line 632 of dm-log.c. See
backtrace at end. Please excuse typos within - had to type it from a
screen shot.
The problem seems to be introduced by the commit
003b5c5719f159f4f4bf97511c4702a0638313dd (block: Convert drivers to
immutable biovecs).
Michael
INFO: task lvm:976 blocked for more than 120 seconds.
Not tainted 3.14.0-rc4+ #5
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
lvm D ffff8802110c0e48 0 976 947 0x00000000
ffff8802110c0a30 0000000000000082 0000000000014300 0000000000014300
ffff880212f7bfd8 ffff8802110c0a30 ffff88021dcd4b88 ffff880212f7bbc0
0000000000000000 0000000000000001 ffffffffa016da40 0000000000000131
Call Trace:
[<ffffffffa016da40>] ? table_load+0x330/0x330 [dm_mod]
[<ffffffff814b0844>] ? io_schedule+0x94/0x130
[<ffffffffa016f3c5>] ? sync_io+0x95/0x120 [dm_mod]
[<ffffffff810b43d0>] ? dyntick_save_progress_counter+0x20/0x20
[<ffffffff814b0900>] ? _cond_resched+0x20/0x40
[<ffffffff814b0e5f>] ? wait_for_completion+0x1f/0x110
[<ffffffffa016f62e>] ? dm_io+0x1de/0x210 [dm_mod]
[<ffffffffa016ef50>] ? bio_next_page+0x30/0x30 [dm_mod]
[<ffffffffa016edc0>] ? bio_get_page+0x40/0x40 [dm_mod]
[<ffffffffa016da40>] ? table_load+0x330/0x330 [dm_mod]
[<ffffffffa01f884a>] ? disk_resume+0x14a/0x2d0 [dm_log]
[<ffffffff8105f03e>] ? put_online_cpus+0x1e/0x80
[<ffffffffa0168bad>] ? dm_swap_table+0x3d/0x300 [dm_mod]
[<ffffffffa01fe3e9>] ? mirror_resume+0x29/0x50 [dm_mirror]
[<ffffffffa016b365>] ? dm_table_resume_targets+0x85/0xd0 [dm_mod]
[<ffffffffa0169082>] ? dm_resume+0x32/0xe0 [dm_mod]
[<ffffffffa016db63>] ? dev_suspend+0x123/0x220 [dm_mod]
[<ffffffffa016e38d>] ? ctl_ioctl+0x1ed/0x420 [dm_mod]
[<ffffffffa016e5ca>] ? dm_ctl_ioctl+0xa/0x10 [dm_mod]
[<ffffffff8119366f>] ? do_vfs_ioctl+0x2cf/0x4b0
[<ffffffff811f9878>] ? ipcget+0x138/0x1c0
[<ffffffff811938d0>] ? SyS_ioctl+0x80/0xa0
[<ffffffff811fd9e1>] ? SyS_semctl+0x1e1/0x270
[<ffffffff814bb6b9>] ? system_call_fastpath+0x16/0x1b
On Tue, Feb 25 2014 at 7:12pm -0500,
Michael Mueller <[email protected]> wrote:
> Hi,
>
> with 3.14-rc4 my box does not finish initializing the LVM array. It
> hangs within the call to rw_header() on line 632 of dm-log.c. See
> backtrace at end. Please excuse typos within - had to type it from a
> screen shot.
>
> The problem seems to be introduced by the commit
> 003b5c5719f159f4f4bf97511c4702a0638313dd (block: Convert drivers to
> immutable biovecs).
You need these two 3.14-rc fixes that are staged in linux-next (via
linux-dm.git's 'for-next' branch):
https://git.kernel.org/cgit/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=for-next&id=d73f9907294c670da14baea3fb31be27879e818e
https://git.kernel.org/cgit/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=for-next&id=f3a44fe0608eb628e1f8f6e3540462e6d171a745
I wanted to send these to Linus last week but I was waiting for some
dm-thin fixes that aren't quite ready. I'll send whatever I have staged
to Linus by this Friday so that 3.14-rc5 will have these fixes no matter
what.
Sorry for the inconvenience.
Mike
Hi Mike,
you wrote:
> You need these two 3.14-rc fixes that are staged in linux-next (via
> linux-dm.git's 'for-next' branch):
Thanks for the fast reply. Did pull in the whole branch 'linux-next'
from the repo now. It is back to working.
Michael