I use introscope to monitor the memory usage of an application. For every seven days I see a graphical report generated by introscope and see if there is any memory leak happening on any server. If I suspect a leak I go ahead bouncing that server. Since there are many servers to monitor this becomes a tedious task.

I was thinking if I could automate this somehow. This is what I could come up with, write a shell script to gather the previous sevens days data and see if its touching the threshold level, if it touches generate an email.

Can someone provide me a better idea on how to implement this.

Thanks.

Nagios http://nagios.sourceforge.net/docs/n.../3/en/toc.html is what I use