Re: BUG #7521: Cannot disable WAL log while using pg_dump

Excerpts from Tom Lane's message of jue sep 06 17:23:07 -0300 2012:
> Heikki Linnakangas <hlinnaka(at)iki(dot)fi> writes:
> > On 06.09.2012 13:07, Robert Haas wrote:
> >> On Thu, Sep 6, 2012 at 3:55 PM, Tom Lane<tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> >>> Doesn't hint-bit setting cause WAL traffic these days?
>
> >> I sure as heck don't think so.
>
> > It does not. HOT page pruning does, however. It could be that..
>
> Sorry, I was thinking of the freeze case, which is also somewhat
> unlikely --- but if it were happening, it could easily happen on
> most/all pages of a table.

How about secondary effects -- say pg_dump reads a lot of data into
buffers that were previously dirty and those get evicted. Could page
eviction cause lots of WAL to be emitted?

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