There seems to be a lot of this needed lately. I suspect the issue is 5.2 -> 5.3 transition.

Putting something like this in /etc/cron.daily/ when near a major update *might* make the transition times between point releases easier -

#!/bin/bash

RANGE=120 number=$RANDOM let "number %= $RANGE" delay=`/usr/bin/expr $number * 60` sleep $delay /usr/bin/yum clean all > /dev/null 2>&1 /usr/bin/yum makecache > /dev/null 2>&1

-=- Point of the random delay is to not have lots of people hit all the servers at once.