I have some strange entries in my slow logs whose timestamps corrolate to an event we are investigating:
Query_time in the 2-4 range Lock_time: 0 for ALL entries Rows_sent in the single to low double-digits Rows_examined in the low to mid hundreds the question is whether these are cause or effect? I am thinking the this reflects the clients becoming unresponsive and MySQL is saying "this is how long it took to get it off the dock" as opposed to "this is how long it took to get it to shipping". is this correst? if not what are plausible explanations for a short burst of these queries which execute hundreds of thousands of time/day to suddeny slow log (w/such low Rows_% #s) across multiple MySQL nodes then just as suddenly stop?
