Out of space when it's not?

[Piglet]

My 104 has started acting like it's out of space, but by the looks of things it should have almost 100GB free: The dashboard reports "93.86GB free of 5.45TB" and checking via ssh gives the same info:

Filesystem Size Used Avail Use% Mounted on
rootfs 4.0G 590M 3.3G 16% /
tmpfs 10M 4.0K 10M 1% /dev
/dev/md0 4.0G 590M 3.3G 16% /
tmpfs 248M 0 248M 0% /dev/shm
tmpfs 248M 1.3M 247M 1% /run
tmpfs 248M 0 248M 0% /sys/fs/cgroup
tmpfs 248M 0 248M 0% /media
/dev/md127 5.5T 5.4T 94G 99% /data
/dev/md127 5.5T 5.4T 94G 99% /home
/dev/md127 5.5T 5.4T 94G 99% /apps
/dev/md127 5.5T 5.4T 94G 99% /var/ftp/Videos

I first noticed the problem when Transmission began reporting errors: "Unable to save resume file: No space left on device" for my active torrents.

Then later a Time Machine backup failed because "Disk full"… Which is kind of weird because shouldn't Time Machine delete files to make room?

I tried restarting but the problem remains. What could be causing this?

[mdgm-ntgr]

# btrfs fi show
# btrfs fi df /data

[Piglet]

This is after removing 70GB of data to make room, so df shows about 170GB free now (which in reality probably is only 70GB usable)

# df
Filesystem      1K-blocks       Used Available Use% Mounted on
rootfs            4173504     604760   3359236  16% /
tmpfs               10240          4     10236   1% /dev
/dev/md0          4173504     604760   3359236  16% /
tmpfs              253592          0    253592   0% /dev/shm
tmpfs              253592       1292    252300   1% /run
tmpfs              253592          0    253592   0% /sys/fs/cgroup
tmpfs              253592          0    253592   0% /media
/dev/md127     5850829568 5681972788 167181132  98% /data
/dev/md127     5850829568 5681972788 167181132  98% /home
/dev/md127     5850829568 5681972788 167181132  98% /apps
/dev/md127     5850829568 5681972788 167181132  98% /var/ftp/Videos

# btrfs fi show
Label: '0e357b14:data'  uuid: 34bda540-18c4-4437-b708-f7d6d81b53c3
	Total devices 1 FS bytes used 5.29TiB
	devid    1 size 5.45TiB used 5.45TiB path /dev/md127

Btrfs v3.17.3

# btrfs fi df /data
Data, single: total=5.44TiB, used=5.29TiB
System, DUP: total=32.00MiB, used=704.00KiB
System, single: total=4.00MiB, used=0.00B
Metadata, DUP: total=2.55GiB, used=1.79GiB

[mdgm-ntgr]

Do you have a backup?

Do you use snapshots on any of your shares?

The space on your NAS is fully allocated.

[Piglet]

I don't have a backup and I'm not using snapshots.

So what does it mean that the space is fully allocated? Has the NAS been damaged somehow or is this normal? I seems weird that it reports free space that isn't there.

[mdgm-ntgr]

Well the first step would be to make a backup.

It's normal for the space to be fully allocated when it is this full. Space is allocated for Data and for metadata. Now different use cases will lead to differing amounts of metadata. So it's possible too much could be allocated to data or too much to metadata leading to "No space left on device" errors.

Normally when the allocated space is exhausted unallocated space will be allocated to data or metadata as required. However when it needs to allocate more space but there is none available it will fail. A balance can be used to address this.

You can schedule volume maintenance such as balance, defrag etc.