Re: Proposal: Log inability to lock pages during vacuum

Heikki Linnakangas wrote:
> On 12/18/2014 09:41 PM, Alvaro Herrera wrote:

> >Here's my proposal.  Instead of punting, I split the message in
> >separately translatable units, and emit only the ones that apply.  The
> >code is messier this way, but I think we can live with that.
> 
> Works for me.

Great, thanks, pushed.  I tweaked it a bit more, so that it would say
either "skipped N pages" or "waited N pins" in both autovacuum and
vacuum verbose cases, but only if N > 0.

-- 
Álvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services