On Mon, 2007-02-05 at 23:25 +0000, Gregory Stark wrote:
> "Gregory Stark" <stark@enterprisedb.com> writes:
>
> > "Bruce Momjian" <bruce@momjian.us> writes:
> >
> >> OK, please propose some wording so at least we can get agreement on
> >> that.
> >
> > How about something open-ended like "arrange for updates that do not update
> > columns referenced by foreign keys from other tables to avoid being blocked by
> > locks from concurrent RI checks"
>
> Hum. Reading back in the thread it seems what I wrote is basically equivalent
> to the wording Simon originally proposed.
I like your wording. It's clearer and includes Stephan's clarification.
Some minor mods...
TODO
"avoid blocking of updates because of concurrent RI checks when those
updates do not alter columns referenced by foreign keys from other
tables"
-- Simon Riggs EnterpriseDB http://www.enterprisedb.com