RE: ORA-1578...block corrupted...error is normal...a block...had a NOLOGGING...operation performed against

Someone just replied to me directly and suggested that maybe the index is defined as;

     select OWNER, INDEX_NAME, LOGGING from dba_indexes where LOGGING='NO'

This seems logical except that I have done this many time before...creating indexes nologging and never (maybe incorrectly) changed the to logging after they completed building.

Maybe I need to take back about what I said about being comfortable with NOLOGGING option (I researched)?

*HOWEVER*, I regularly apply redo logs from and active PROD to a database with many indexes defined as LOGGING='NO' to a database I restored and recovered from an RMAN backup. I never have errors on the recovered database???

Chris Marquez
Oracle DBA

-----Original Message-----
From: Riyaj Shamsudeen [mailto:rshamsud_at_jcpenney.com] Sent: Fri 8/19/2005 2:52 PM
To: joelgarry_at_anabolicinc.com
Cc: Marquez, Chris; Thomas.Mercadante_at_labor.state.ny.us; Joel Garry; oracle-l_at_freelists.org Subject: Re: ORA-1578...block corrupted...error is normal...a block...had a NOLOGGING...operation performed against  

NOLOGGING simply marks the range of blocks modified in that nologging operation invalid, by writing redo records with pertinent details. Replaying redo records from this timeframe, will mark the data blocks as invalid. But, If you take a backup after the nologging operation then you should be able to rollforward from that backup, as long as there are no other nologging operations after the backup.

There must have been another nologging operation after the backup performed on this table/index. Are you sure that the table is not inserted with append hint ? What does created column in dba_objects show for this table/index ?

Thanks

Riyaj "Re-yas" Shamsudeen
Certified Oracle DBA (ver 7.0 - 9i)
Allocation & Assortment planning systems JCPenney

Joel Garry wrote:

>>Not really, I still believe it. I know I have restored databases in
>>
>>
>the past
>
>
>>that had very old NOLOGGING operations performance
>>against them and *now* included in a recent backup...not arch logs...
>>in the backup, part of the database.
>>
>>
>
>So that means that the roll forward checking pertains to blocks in the
>data file rather than the logs? A guess, I don't know, doesn't sound
>right. It might make a difference if the blocks had or had not been
>updated in the meantime.
>
>Joel Garry
>http://www.garry.to
>
>--
>http://www.freelists.org/webpage/oracle-l
>
>

--
http://www.freelists.org/webpage/oracle-l