Balancing does not help with "no space left" despite there is capacity available.

I've run into the issue with ReadyNas 4220. Firmware 6.4.2

There were a lot of posts in community for "no space left" error but none worked for me.

# echo Test > /home/admin/test
-bash: /home/admin/test: No space left on device

# btrfs fi show
Label: '43f63420:root' uuid: 701c9fe6-995a-4572-b360-dd7986123e77
Total devices 1 FS bytes used 438.08MiB
devid 1 size 4.00GiB used 1.64GiB path /dev/md0

Label: '43f63420:prod' uuid: 96799cff-1ad8-4deb-a06b-758cacfea9bd
Total devices 1 FS bytes used 5.28TiB
devid 1 size 6.49TiB used 5.81TiB path /dev/md127

# df -ih
Filesystem Inodes IUsed IFree IUse% Mounted on
tmpfs 985K 582 985K 1% /dev
/dev/md0 0 0 0 - /
tmpfs 985K 1 985K 1% /dev/shm
tmpfs 985K 723 985K 1% /run
tmpfs 985K 4 985K 1% /sys/fs/cgroup
tmpfs 985K 1 985K 1% /media
/dev/md127 0 0 0 - /prod
/dev/md127 0 0 0 - /home
/dev/md127 0 0 0 - /apps
/dev/md127 0 0 0 - /run/nfs4/home
/dev/md127 0 0 0 - /run/nfs4/prod/VMStorage
/dev/md127 0 0 0 - /run/nfs4/prod/ISO

# df -h
Filesystem Size Used Avail Use% Mounted on
tmpfs 10M 4.0K 10M 1% /dev
/dev/md0 4.0G 467M 3.2G 13% /
tmpfs 3.9G 0 3.9G 0% /dev/shm
tmpfs 3.9G 13M 3.9G 1% /run
tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup
tmpfs 3.9G 0 3.9G 0% /media
/dev/md127 6.5T 5.3T 1.2T 82% /prod
/dev/md127 6.5T 5.3T 1.2T 82% /home
/dev/md127 6.5T 5.3T 1.2T 82% /apps
/dev/md127 6.5T 5.3T 1.2T 82% /run/nfs4/home
/dev/md127 6.5T 5.3T 1.2T 82% /run/nfs4/prod/VMStorage
/dev/md127 6.5T 5.3T 1.2T 82% /run/nfs4/prod/ISO

# btrfs fi df /
Data, single: total=1.21GiB, used=425.52MiB
System, DUP: total=8.00MiB, used=16.00KiB
System, single: total=4.00MiB, used=0.00B
Metadata, DUP: total=204.56MiB, used=12.53MiB
Metadata, single: total=8.00MiB, used=0.00B
GlobalReserve, single: total=16.00MiB, used=0.00B

# btrfs fi df /prod
Data, single: total=5.80TiB, used=5.27TiB
System, DUP: total=32.00MiB, used=672.00KiB
Metadata, DUP: total=9.00GiB, used=2.65GiB
GlobalReserve, single: total=512.00MiB, used=0.00B

# btrfs filesystem usage /prod
Overall:
Device size: 6.49TiB
Device allocated: 5.83TiB
Device unallocated: 675.58GiB
Device missing: 0.00B
Used: 5.28TiB
Free (estimated): 1.18TiB (min: 871.52GiB)
Data ratio: 1.00
Metadata ratio: 2.00
Global reserve: 512.00MiB (used: 0.00B)

Data,single: Size:5.79TiB, Used:5.27TiB
/dev/md127 5.79TiB

Metadata,DUP: Size:17.50GiB, Used:2.65GiB
/dev/md127 35.00GiB

System,DUP: Size:64.00MiB, Used:672.00KiB
/dev/md127 128.00MiB

Unallocated:
/dev/md127 675.58GiB

BTRFS info (device md127): 634 enospc errors during balance
md: requested-resync of RAID array md127
md: minimum _guaranteed_ speed: 30000 KB/sec/disk.
md: using maximum available idle IO bandwidth (but not more than 200000 KB/sec) for requested-resync.
md: using 128k window, over a total of 995354752k.
md: md127: requested-resync done.
BTRFS info (device md127): resizing devid 1

After balancing start from UI:

# dmesg | tail
BTRFS info (device md127): relocating block group 5189960466432 flags 1
BTRFS info (device md127): found 1307 extents
BTRFS info (device md127): found 1307 extents
BTRFS info (device md127): relocating block group 5165264404480 flags 1
BTRFS info (device md127): found 91 extents
BTRFS info (device md127): found 91 extents
BTRFS info (device md127): relocating block group 4394317774848 flags 1
BTRFS info (device md127): relocating block group 3251856474112 flags 1
BTRFS info (device md127): 329 enospc errors during balance
BTRFS info (device md127): 1 enospc errors during balance

I tried to free up to 20% on /prod and run balancing again. It made progress to 75% and dropped off again with same error.

After balancing attempt /prod usage raised from 80% to 83%.

Playing with balancing filted did not help as well

# btrfs balance start -v -dusage=0 /prod
Dumping filters: flags 0x1, state 0x0, force is off
DATA (flags 0x2): balancing, usage=0
Done, had to relocate 0 out of 5844 chunks
# btrfs balance start -v -dusage=10 /prod
Dumping filters: flags 0x1, state 0x0, force is off
DATA (flags 0x2): balancing, usage=10
Done, had to relocate 0 out of 5844 chunks
# btrfs balance start -v -dusage=20 /prod
Dumping filters: flags 0x1, state 0x0, force is off
DATA (flags 0x2): balancing, usage=20
ERROR: error during balancing '/prod' - No space left on device
There may be more info in syslog - try dmesg | tail

I also tried scrubling and defragmentation.

Would you please suggest what else I should try to fix the issue?