Re: No space left on device when doing "mkdir"

2017-05-01 Thread Gerard Saraber
It did it again: shrapnel share # touch test.txt touch: cannot touch 'test.txt': No space left on device shrapnel share # df -h Filesystem Size Used Avail Use% Mounted on /dev/root35G 19G 15G 56% / devtmpfs 10M 0 10M 0% /dev tmpfs 3.2G 1.2M 3.2G 1% /

Re: No space left on device when doing "mkdir"

2017-04-28 Thread Gerard Saraber
Dmarc is off, here's the output of the allocations: it's working correctly right now, I'll update when it does it again. /sys/fs/btrfs/7af2e65c-3935-4e0d-aa63-9ef6be991cb9/allocation/system/flags:2 /sys/fs/btrfs/7af2e65c-3935-4e0d-aa63-9ef6be991cb9/allocation/system/raid1/used_bytes:3948544 /sys/f

Re: No space left on device when doing "mkdir"

2017-04-27 Thread Chris Murphy
On Thu, Apr 27, 2017 at 10:46 AM, Gerard Saraber wrote: > After a reboot, I found this in the logs: > [ 322.510152] BTRFS info (device sdm): The free space cache file > (36114966511616) is invalid. skip it > [ 488.702570] btrfs_printk: 847 callbacks suppressed > > > > On Thu, Apr 27, 2017 at 10:

Re: No space left on device when doing "mkdir"

2017-04-27 Thread Gerard Saraber
After a reboot, I found this in the logs: [ 322.510152] BTRFS info (device sdm): The free space cache file (36114966511616) is invalid. skip it [ 488.702570] btrfs_printk: 847 callbacks suppressed On Thu, Apr 27, 2017 at 10:18 AM, Gerard Saraber wrote: > no snapshots and no qgroups, just a st

Re: No space left on device when doing "mkdir"

2017-04-27 Thread Gerard Saraber
no snapshots and no qgroups, just a straight up large volume. shrapnel gerard-store # btrfs fi df /home/exports Data, RAID1: total=20.93TiB, used=20.86TiB System, RAID1: total=32.00MiB, used=3.73MiB Metadata, RAID1: total=79.00GiB, used=61.10GiB GlobalReserve, single: total=512.00MiB, used=544.00K

Re: No space left on device when doing "mkdir"

2017-04-27 Thread Roman Mamedov
On Thu, 27 Apr 2017 08:52:30 -0500 Gerard Saraber wrote: > I could just reboot the system and be fine for a week or so, but is > there any way to diagnose this? `btrfs fi df` for a start. Also obligatory questions: do you have a lot of snapshots, and do you use qgroups? -- With respect, Roman