Re: Documentation bug: reference to checkpoint activity in bgwriter

On 9 January 2012 18:47, Robert Haas <robertmhaas@gmail.com> wrote:
> On Sat, Dec 31, 2011 at 8:54 PM, Peter Geoghegan <peter@2ndquadrant.com> =
wrote:
>> ISTM that the following reference, at config.sgml line 1345, ought to
>> be adjusted due to the introduction of the new checkpointer process:
>>
>> =A0Setting this to zero disables
>> =A0 =A0 =A0 =A0 background writing (except for checkpoint activity).
>
> Hmm, so what should we adjust it *to*?

How about "Setting this to zero disables background writing. Note that
checkpoints, which are managed by a separate, dedicated auxiliary
process, are unaffected."

If this was the first release of Postgres, I'd suggest that we remove
any references to checkpoints, which are sort of orthogonal to the
stated main function of the bgwriter in past releases, which is to
clean dirty buffers so that backends don't have to. However, people
may still expect the bgwriter to manage checkpointing, particularly if
they're working off old sources of information, so it's useful to set
them straight.

--=20
Peter Geoghegan =A0 =A0 =A0 http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services