It's human error. Simple as that. Happens to absolutly everyone even the brightest individuals. I've seen in "the livesite" that this kind of spike problems are more common than you think, and the ideal solution is to automate an alarm that tells you both, when a spike is happening, and when X throughput passes over a threshold. This kind of automation is easy and spares lots of trouble.Could be as simple as scheduled task parsing some logs. Some more fancy option is always good though :)
Comments
It's human error. Simple as that. Happens to absolutly everyone even the brightest individuals. I've seen in "the livesite" that this kind of spike problems are more common than you think, and the ideal solution is to automate an alarm that tells you both, when a spike is happening, and when X throughput passes over a threshold. This kind of automation is easy and spares lots of trouble.Could be as simple as scheduled task parsing some logs. Some more fancy option is always good though :)