On Mon, 20 Aug 2007, Kevin Grittner wrote:

The problem went away completely when we used a very aggressive background writer configuration, to put the dirty pages in front of the OS file system right away, so that its algorithms and the controller cache could deal with things before they got out of hand.

Can you share what the aggressive BGW settings you settled on were? Knowing that information would allow giving better suggestions on how to translate what you're currently doing into the very different settings 8.3 uses in this area. Also be helpful to know about how often you have a checkpoint with your current configuration, and how big your shared_buffers is to get a general context for the size/frequency of potential checkpoint problems.

* Greg Smith [EMAIL PROTECTED] http://www.gregsmith.com Baltimore, MD

---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

Reply via email to