Seems that stopping inetd solved the problem 
, temporarily stop it with:
/etc/init.d/S60inetd stop
or permanently disable it by e.g.
mv /etc/init.d/S60inetd /root/
     (enable again by mv /root/S60inetd /etc/init.d/)
my uptime is 10 days and I copied ~ 1 TB, I’ve never reached this limit before…
UPDATE : crashed again 
after 10 days - this thing is so buggy - i updated to firmware 1.02.06 , stopped MioNet, Twonky and inetd again (firmware update removes all system modifications!) and keep on trying … Consider rebooting your Windows machines (shared folders might not be accessible from Windows after the update!)
UPDATE : uptime 5 days now, ~ 3GB transferred since last reboot …
UPDATE : uptime 7 days, switched NFS to RW access
UPDATE : uptime 12 days …
UPDATE : uptime 18 days, I guess the firmware upgrade and deactivation of services has improved this poor device significantly and WDC support suggests changing hardware (!!!) A friend of mine bought the same box in the meantime. He is very upset because he did not ask me before and - surprise - his brand new MBWE II suffers from the exact same error !
UPDATE : uptime 43 days, I’m beginning to trust this box now 
