testing entire filesystem

HI,

I've made a RAID1 over two sata drives using MD and one partition on
each device. I use etx3 on the resulting md device (/dev/md0).

I used a tool to size the partitions to avoid having to guess the
maximum sizes, but I wanted to make sure I could write to the entire
filesystem and so I used dd to make a big file :

sudo dd if=/dev/zero of=bigfile bs=1024K count=500M

This resulted in a somewhat predictable :

dd: writing `bigfile': No space left on device
472433+0 records in
472432+0 records out
495381721088 bytes (495 GB) copied, 8204.09 s, 60.4 MB/s

However, it also resulted in :

Jul 25 17:37:42 jeeves mdadm: Rebuild60 event detected on md device /dev/md0

and md is dutifully rebuilding it.

Clearly, this concerns me.

Could someone enlighten me as to what went wrong (if anything)?

Is there a better way to test the file system and partition/drive are
writable in their entirety?

Thanks,

Max.

Using ubuntu 8.04, btw.