Subscribe to our channel: https://youtube.pl/c/DevoxxPoland?sub_confirmation=1
Metrics and monitoring systems are 1 of the fundamental parts of distributed systems. Sometimes they wake us up at night thanks to automatic alerts :wink: , but they’re here on purpose: to give us better overview of situation in case of failures. But what do you actually know about YOUR metrics? Has this always happened to you that you faced an actual failure and realized that you don’t realize what you’re looking at? Or spent hours debugging any weird problem only to discover that your metrics lied to you for all this time?
In this presentation, I’d like to rise awareness of certain class of problems related to bad metrics and deficiency of experience in measuring things. any of you may have known about it for years. At the same minute quite a few (even senior) developers inactive hadn’t heard of them, making the same mistakes... or possibly didn’t have a chance to work on a right task to gain this experience? I’ll besides share, how conventions can aid your squad making better metrics... and yet better alerts detecting actual failures. And show you any specified conventions - possibly they will work for you?
This is not a presentation about tools or frameworks. It’s about your skills to diagnose problems and data to assist you in that. We’ll just stick to examples from the domain of distributed systems moving on JVM.
Recorded at Devoxx Poland 2024
Twitter: https://twitter.com/DevoxxPL
Instagram: https://www.instagram.com/DevoxxPL
Join us besides here:
Devflix: https://devflix.pl
#Devoxx #DevoxxPoland #IT #Development #SoftwareDevelopment