Strange ROWID problem

Hello out there,

today I encountered a very strange behaviour of Oracle 7.3.3.6.0 when I tried to eliminate duplicates from a small table. The problem can be reduced to the following SQL:

select a.rowid, b.rowid
from mytable a, mytable b
where a.rowid<b.rowid
and a.rowid=b.rowid;

To my great surprise, this query resulted in 20 rows on one specific table!
With any other table I tried, I got the expected empty result set.

Can anyone put some light on that?

Thanks in advance,
Lothar

--

Lothar Armbrüster       | lothar.armbruester_at_rheingau.netsurf.de
Schulstr. 12            | lothar.armbruester_at_t-online.de
D-65375 Oestrich-Winkel |