[NTLUG:Discuss] strange "No space" issue on /dev/md0

[Fred James]

RH 7.1
"df -k" returns

Filesystem           1k-blocks      Used Available Use% Mounted on
/dev/hda1              3951272    726856   3023700  20% /
/dev/md0             113487584  26951312  80771408  26% /backups

and attempts to ftp files to a directory on /dev/md0 will succeed up to a point and then return errors like "No space left on device" as show in this snippet:

226 Transfer complete.
23225349 bytes received in 2.08 secs (1.1e+04 Kbytes/sec)
local: find_apx01.dbf.gz remote: find_apx01.dbf.gz
227 Entering Passive Mode (63,83,66,43,11,221)
150 Opening BINARY mode data connection for 'find_apx01.dbf.gz' (41915008 bytes).
local: find_apx01.dbf.gz: No space left on device
421 Service not available, remote server has closed connection
local: find_ard01.dbf.gz remote: find_ard01.dbf.gz

The same error occurred whether the the script was run from cron or manually, and when I attempted the ftp itself manually.  During one session, through another terminal window, I deleted the files that had made it up to the point of the error, and the rest of the files in the session made it through.  This seems to lend weight to the idea that /dev/md0 is out of space even though it reports it is not - and that leaves me confused.  Any suggestions on how to check this out, or better yet, how to fix it, would be most appreciated.

Thanks in advance.

-- 
find . -type f | xargs file | grep -i text | cut -f1 -d: | xargs grep "..."