Re: reindexdb program error under PG 8.1.3

[David Wall <d.wall@xxxxxxxxxxxx>]

Rats!  I found nothing....  But just for grins, I ran the reindexdb program a second time, and this time it reported a different OID, higher than the previous.  I know we didn't change the template1 database other than using it when creating new databases, creating users, tablespaces, etc.

My guess is this is somehow related to the reindexdb program.  Do you know if there's any advantage to using the reindexdb program versus running the psql commands REINDEX DATABASE and REINDEX SYSTEM? 

When reading the docs, I'm beginning to wonder if it's even necessary to do the reindexing.  I think it may be a holdover from earlier thinking that doesn't apply anymore.  I believe the idea is that as rows are inserted, updated and deleted, the index can get rather scrambled, and that reindexing often sorts the values and rewrites that data so that the indexes take less space and work faster.  Is that not the case here?

Thanks,
David



Tom Lane wrote:

David Wall <d.wall@xxxxxxxxxxxx> writes:
  

Thanks, but I clearly lack that expertise.  How do I find "a table with 
that OID"?
    

Try "select * from pg_class where oid = 41675" and
"select * from pg_attribute where attrelid = 41675"
to see if you find any rows.

			regards, tom lane