Re: Logical block corruption
Date: Wed, 1 Apr 2009 13:15:35 -0700 (PDT)
Message-ID: <ff6b24c9-ee9f-46ad-9472-7ed89200a71f_at_z23g2000prd.googlegroups.com>
On Apr 1, 11:13 am, Mladen Gogala <gogala.mla..._at_gmail.com> wrote:
> On Wed, 01 Apr 2009 10:04:30 -0700, jgar the jorrible wrote:
> > Wouldn't that be using a corrupt index to make a new corrupt index? Or
> > did I miss a joke?
>
> Nope, it would not. The "rebuild" option essentially drops & re-creates
> an index. There is another option, "coalesce", which re-balances it, sort
> of. Coalescing an index does not change the level of the index being
> coalesced, so even if you deleted most of the records, the level of the
> index will remain the same after coalescing. It will not remain the same
> after a rebuild.
>
> --http://mgogala.free hostia.com
So am I misreading http://download.oracle.com/docs/cd/B19306_01/server.102/b14231/indexes.htm#sthref2553 ?
"When you rebuild an index, you use an existing index as the data source. Creating an index in this manner enables you to change storage characteristics or move to a new tablespace. Rebuilding an index based on an existing data source removes intra-block fragmentation. Compared to dropping the index and using the CREATE INDEX statement, recreating an existing index offers better performance."
Wish I had time to play, I would expect a corrupt index to fail the rebuild. But it wouldn't be the first time the docs were wrong. But seeing your DBV response, I could envision that if it isn't really corrupt, rebuilding might be the answer. But since the OP specifies it failed with RMAN, I'd think the first thing to do at this point is a regular create index.
On this particular day of the year, I find myself wondering if anything is a joke. One I really had to think about was the sciam.com article about vinyl flooring causing autism. I just want to bang my head on the floor...
jg
-- _at_home.com is bogus. http://www3.signonsandiego.com/stories/2009/apr/01/1m1ucsd011031-ucsd-e-mail-mistakenly-sent-denied-a/?uniontribReceived on Wed Apr 01 2009 - 15:15:35 CDT