An internal object (segment - Index) called I_HH_OBJ#_COL# which has grown
to have really big extents the largest extent size now is over 500MB, and I
believe the next extent request will be for 720MB based on the next_extent
value in dba_segments.
Our DB block size is 8K and the free extents are capped at 520MB because of
this. Next time the object grows and Oracle has to increase extents, it
will look for a 720MB sized extent, but it will only find 520 MB of chunks.
This will result in an error obviously.
This is on the SYSTEM tablespace, which is locally managed and has SYSTEM
allocation type.
I am concerned about two things:
1. Does oracle adjust algorithm to look for smaller extents if a bigger
extent request fails.
2. If it doesn't, what are my options? I see that this is one of the
bootstrap objects in Oracle and can't really be tampered with. The index
was introduced in 7.3.3 so wondering if it's as simple as dropping and
recreating the index.
Any help is appreciated
