On Apr 30, 2011, at 8:22 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> I wonder though if it wouldn't be smarter
> to insist that autovacuum acquire some lock on the main table
> when processing a toast table.
Boy, it sure seems better to avoid the above if we can. So +1 for the other way around - make CLUSTER lock the TOAST table if it cares about a VACUUM happening there.
...Robert