I'm having difficulty getting a profile conversion from raid0 ->
single to finish. Is there a way to tell the filesystem to stop all
new allocations for the raid0 profile, before the full balance
finishes? Otherwise I feel like I'm in a sisyphian push of blocks into
higher block numbers.
Since the profile conversion seems to fail frequently with ENOSPC even
though space isn't an issue:
$ (btrfs file show && btrfs file df /mirror && df -h /mirror)
Label: isibackup uuid: 550c0f77-8f75-40f3-a64e-42c87a0c8e8d
Total devices 3 FS bytes used 61.35TiB
devid 1 size 30.01TiB used 20.16TiB path /dev/sdb
devid 2 size 30.01TiB used 20.16TiB path /dev/sdc
devid 3 size 40.02TiB used 29.15TiB path /dev/sdd
Btrfs v3.12-dirty
Data, RAID0: total=11.58TiB, used=11.57TiB
Data, single: total=57.64TiB, used=49.68TiB
System, RAID1: total=40.00MiB, used=7.06MiB
System, single: total=4.00MiB, used=0.00
Metadata, RAID1: total=134.00GiB, used=90.15GiB
Metadata, single: total=8.00MiB, used=0.00
Filesystem Size Used Avail Use% Mounted on
/dev/sdb 101T 62T 38T 63% /mirror
I wrote a little script to try the profile conversion, and when it
fails try a regular balance which always seems to work fine, ala:
e=550000000000000
i=479300000000000
step=100000000000
btrfs balance start -dusage=0,vrange=0..$e /mirror
while [ $i -lt $e ]
do
h=$((i-step))
if ! btrfs balance start
-dprofiles=raid0,convert=single,vrange=0..$i /mirror
then
btrfs balance start -dvrange=$h..$i /mirror
fi
i=$((i+step))
done
However while this is ongoing, new data written to the filesystem
sometimes gets written to the raid0 profile. So making progress
appears quite slow, maybe 0.5 TB a week from data raid0 -> data
single. Thus the question can I tell it to stop allocating new raid0
space some how, and only use single for new data?
--
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
