Re: Fixing Btrfs Filesystem Full Problems typo?

[Robert White <rwhite@pobox.com>]

On 12/09/2014 02:29 PM, Patrik Lundquist wrote:
> Label: none  uuid: 770fe01d-6a45-42b9-912e-e8f8b413f6a4
>      Total devices 1 FS bytes used 1.35TiB
>      devid    1 size 2.73TiB used 1.36TiB path /dev/sdc1
>
>
> Data, single: total=1.35TiB, used=1.35TiB
> System, single: total=32.00MiB, used=112.00KiB
> Metadata, single: total=3.00GiB, used=1.55GiB
> GlobalReserve, single: total=512.00MiB, used=0.00B

Are you trying to convert a filesystem on a single device/partition to 
RAID 1?

I don't think thats legal. Whithout a second slice to distribute the 
copies of the data onto there is no raiding to be done.

Add the second device with btrfs device add, and _then_ use balance to 
redistribute and copy the data to the second device.

ASIDE: I, personally, think that a single device RAID1 should be legal. 
I also think that it should be possible to tell the system that you want 
N copies if you have N-or-more slices onto which they would spread. 
These would match my expectations from mdadm and several hardware and 
appliance RAID solutions. But my opinions in the matter do _not_ match 
the BTRFS code base. RAID1 means exactly two devices (for any given 
piece of information) [though I don't know whether it always has to be 
the _same_ two devices for two different pieces of information.]

So yea, if that is what you are trying to do, the inability to find a 
second drive on which to allocate the peer-block(s) for an extent would 
produce interesting errors. I cant say for sure that this is the exact 
genesis of your issue, but I've read here in other threads a number of 
comments that would translate as "trying to set RAID1 with on a 
one-slice file system will be full of fail".