Re: SKIP LOCKED DATA

On Mon, Jan 16, 2012 at 5:33 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Thomas Munro <munro(at)ip9(dot)org> writes:
>> I am wondering out loud whether I am brave enough to try to propose
>> SKIP LOCKED DATA support and would be grateful for any feedback and/or
>> {en|dis}couragement.  I don't see it on the todo list, and didn't find
>> signs of others working on this (did I miss something?), but there are
>> examples of users asking for this feature (by various names) on the
>> mailing lists.  Has the idea already been rejected, is it
>> fundamentally infeasible for some glaring reason, or far too
>> complicated for new players?
>
> It sounds to me like "silently give the wrong answers".  Are you sure
> there are not better, more deterministic ways to solve your problem?

The name is misleading.

It means "open a cursor on a query, when you fetch if you see a locked
row return quickly from the fetch".

The idea is that if its locked it is still being written and therefore
not interesting (yet).

Sounds reasonable request.

--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services