Who are your alerts for?

It’s important to get alerts technically correct but too often we forget it’s a human that is receiving the alerts most of the time. We will go over critical questions you should ask and issues you should address to make sure the actual humans responding to your alerts can be effective and happy.

Too often we focus on getting the technicality of the alerts correct or rely on automated tools to receive them that we forget to make the alerts as friendly as possible for its main customer, the human responders. Creating a correct alerting infrastructure is hard enough but making the alerting infrastructure to be human-friendly is possibly the hardest part to get right. The attendees will learn the types of questions that they should ask and answer along with tips on how to address them even before building out any alerting infrastructure to have increased success in designing and implementing one that would be most useful for the actual humans receiving them. We will also go over various real life examples of alerting approaches that are human-focused compared to alerting that is more useful for automation tools to drive the point of what kinds of alerting should be sent to a human versus a machine. This talk is for any operations team, no advanced knowledge necessary.



Ren Lee

Ren is an infrastructure and ops engineer at Arista Networks, busy being a jack-of-all-trades debugging linux servers at datacenters, all the way to building all sorts of automation beasts that ...