[arch-general] [signoff] btrfs-progs

Jackson Alley toomanymirrors at gmail.com
Sat Aug 4 08:53:47 EDT 2012


On 08/03/2012 09:11 PM, C Anthony Risinger wrote:
> On Aug 3, 2012 7:56 PM, "Jackson Alley" <toomanymirrors at gmail.com> wrote:
>> On 08/03/2012 08:50 PM, C Anthony Risinger wrote:
>>> On Aug 3, 2012 5:32 PM, "Jackson Alley" <toomanymirrors at gmail.com>
> wrote:
>>>> On 08/03/2012 06:20 PM, Tom Gundersen wrote:
>>>>> On Sat, Aug 4, 2012 at 12:04 AM, Jackson Alley <
> toomanymirrors at gmail.com>
>>> wrote:
>>>>>> Ok, looks like I've got an unusual setup and am having issues at
> boot.
>>>>>> I've got multi-device btrfs raid and am still using the old rc/init
>>>>>> layout.
>>>>> Could you give more info. What exactly fails? What is the output?
>>>>>
>>>>>> I found this, but haven't verified if it works yet:
>>>>>>
> http://help.lockergnome.com/linux/Bug-634658-BTRFS-raid-configurations-work-box--ftopict558957.html
>>>>> I didn't get that. You haven't verified that what works?
>>>>>
>>>>> -t
>>>> Sorry, I'm just getting back from vacation and am having trouble
> finding
>>>> the time to reboot right now playing catchup. At boot it gives an error
>>>> about wrong fs type and doesn't mount any of the btrfs volumes. Also I
>>>> get an error with mkinitcpio trying to include a btrfs hook, says no
>>>> hook found.
>>> I suppose if you want a concrete solution you'll need to produce an
>>> accurate account of the symptoms ... you're configs mean little sans the
>>> necessary context to evaluate them.
>>>
>>> So what [exactly] are the symptoms here?  Capture + paste the boot logs.
>>>
>> That's hard to do considering /var is btrfs and networking fails without
>> it. The symptom is quite simple, during boot I get an error that the
>> wrong file system type was found on one of the /dev/sd's and to check
>> dmesg |tail. Logging in as root and running btrfs device scan then
>> remounting works so I just need to ensure that gets done before the
>> fstab is read. It used to be that having USEBTRFS="yes" in rc.conf did
>> that for me, but not anymore.
> If btrfs initramfs hook should only be needed for a root btrfs AFAIK, but
> if the initramfs is probing before the btrfs module is loaded/available or
> devices scanned that would produce such error.
>
> Is it actually failing in initramfs, or after?
>
> Paste output of failing `mkinitcpio` command.
>
> I am mobile ATM so I can only read check so much, but verbatim output is
> required to debug further.
>

I've rebooted and confirmed the udev rule I found does not resolve the
issue. I don't believe the problem lies with mkinitcpio as I'm not using
a root btrfs system but rather in the init scripts
(https://projects.archlinux.org/initscripts.git/commit/?id=13ca7f028ac775773a6685966d0faeee5cc55a9d).
I've attached the boot log and dmesg output.
dmesg: http://pastebin.com/PxfnYgAm
boot log: http://pastebin.com/Pg9xY4u5
Jackson



More information about the arch-general mailing list