Tobias Brox <tobixen(at)gmail(dot)com> wrote:
> All while having this problem, there was one heavy query running
> in parallell ... not sure if that's relevant.
Have you turned on checkpoint logging? You might want to see if
these are happening at some particular point in the checkpoint
processing. If so, look through the archives for posts from Greg
Smith on how to tune that -- he's worked out a nice methodology to
iteratively improve your configuration in this regard.
-Kevin