Re: pg_autovacuum: short, wide tables

>>I'm not proposing it for 8.1 though ...
>>
>>
>
>Well, why not?  Arguably Mark's problem is a bug, and it's not too late
>to address bugs.
>
>
FWIW, I'd be satisfied with a warning in the autovacuum docs about this
corner case.

>I'm not sure about the idea of not vacuuming the toast table when we
>decide to vacuum the main table.
>
>
If you promote toast tables to autovacuum candidates that can be
vacuumed independently, I think autovacuum doesn't need to do both when
it does the main table.  This would potentially improve performance by
minimizing the amount of work that needs to be done when a vacuum is
necessary.  OTOH, you can't stop regular vacuum from including the toast
table otherwise way more people would start bugging you :)

-Mark.