Alerting or Monitoring?An alert that you can't respond to is a wasted alert. But monitoring can cover anything.
When I started talking about deployment monitoring, I didn’t expect so many responses… and they keep coming!
In response to When not to monitor your systems, Mitchell de Rijcke wrote in:
Instead of spamming yourself with alerts you could create a daily alert which only posts the numbers, how many of each, or even better how many % of requests were 500s for every service.
Then you can still see progress in a meaningful way because you can actually understand the volumes :)
A great point… and it makes obvious that I’m a bit guilty of an unintentional conflation of terms.
There’s a big difference between monitoring your system, and alerting about your system. And what Mitchell is advocating for is monitoring.
Whether or not you alert for failed deployments, or 500 errors, or slow SQL queries, or whatever else, it’s often very valuable to monitor these things.
Alerts should all be actionable. Any alert that you can’t (or worse: shouldn’t) respond to immediately, is a wasted alert.
But monitoring can cover all sorts of things. Anything that gives you insight into the health of your system, or customer behavior, is fair game for monitoring.
Adventures in DevOps 129: The Future of Intelligent Monitoring and Alerting with Ava Naeini
Ava Naeini shares her patent-pending tool that uses ML to determin the health and performance of distributed systems.
How serious is a deployment failure?
A deployment failure should go through standard alert channels, but should it page whomever is on call?
Improve your software delivery