On 2/16/15 2:41 AM, Andres Freund wrote:
Since you actually don't want cancellations for the long running reporting queries it very much might be sensible to switch to a more complicated version of HeapTupleSatisfiesVacuum() if there's longrunning queries. One that can recognize if rows are actually visible to any backend, or if there are just snapshots that see older rows. I've previously wondered how hard this would be, but I don't think it's *that* hard. And it'd be a much more general solution.
Josh Berkus mentioned on IRC that they're working on improving vacuum, and I think it was something along these lines.
-- Jim Nasby, Data Architect, Blue Treble Consulting Data in Trouble? Get it in Treble! http://BlueTreble.com -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
