When I see these things I always wonder, what is the best way to integrate these monitoring stuff into your existing stack. So server restarts and failures are

When I see these things I always wonder, what is the best way to integrate these monitoring stuff into your existing stack. So server restarts and failures are just okay. Anyone used similar monitoring utilities in production?