I think I noticed a bug on Metalink about
this some time ago. But I can't guarantee
that my memory is correct.
Next Seminar dates:
(see [url]http://www.jlcomp.demon.co.uk/seminar.html[/url] )
The Co-operative Oracle Users' FAQ
Dave wrote in message .......>I am rebuilding multiple UNUSABLE index partitions on Oracle HP/UXdespite>
>Oracle8i Enterprise Edition Release 220.127.116.11.0 - 64bit Production
>With the Partitioning option
>JServer Release 18.104.22.168.0 - 64bit Production
>The rebuild command ...
>alter index XIE12FCT_ITEM rebuild partition Y2002_M11 compute
>statistics pctfree 0 nologging parallel;
>... which normally completes in less than ten seconds was observed to
>be taking very much longer. OEM showed a "Long Operation" for the
>session, and is showing scans of every partition of the table,>showing the correct SQL.
>We have done this every month for over two years, and suddenly all of
>these rebuilds are full-scanning the source tables. We are not aware
>of any config changes to the DB or server.
>Tables are all range partitioned by DATE with mostly bitmap indexes.
>Does any have any insights here?