Decrease the polling frequnecy on the nodes in question. What else is ocurring on the DB at this time? Anymore details about the timeouts?
I get those too. I get an alert for a CISCO core switch going down but when i run the uptime command, it shows that it never went down. Weird. Goes down and comes back up in the same minute. 2minute polling intervals.