Watching the system as vacuum is running, I can see that we are encountering
the kswapd/kscand problem in the 2.4.20 kernel. This could very well
account for the non-linear increase in vacuum time.
This problem is fixed in the 2.6 kernel, but we can't upgrade because DELL
is dragging their feet in releasing hardware monitoring compatible with 2.6
kernel. So, we're going to try a 2.4.29 kernel and hope that the problem is
fixed there. With any luck, by Friday I'll know if the kswapd problem is
fixed in 2.4.29 and if that solves the excessive vacuum times.
Wes
---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
