There's an Open Item for 8.4 about documenting the fact that the
bgwriter is now active during archive recovery. So down to me.
We don't currently document which processes are active at any given
time. i.e. we didn't document that the bgwriter *wasn't* active during
recovery, so there's nothing to change.
Thoughts?
-- Simon Riggs www.2ndQuadrant.comPostgreSQL Training, Services and Support