From the course: Deploying Scalable Machine Learning for Data Science
Join today to access over 24,700 courses taught by industry experts.
Kubernetes monitoring
From the course: Deploying Scalable Machine Learning for Data Science
Kubernetes monitoring
- [Instructor] We also need to monitor Kubernetes when we're deploying scalable machine learning solutions. Now the kinds of information we wanna collect when we're monitoring Kubernetes includes information about the pods that are running our containers, core metrics, and we also wanna detect problems with our nodes and we have a tool called node problem detector which we'll discuss. Now when we're monitoring pods, we can use the Kubectl describe pods command and that gives us a lot of information about the state of the pods, for example whether it's in waiting state, running state, or if it's been terminated. We also get information about whether or not the pod is ready. In that case, it's just a Boolean state. It's either ready or not ready. And then we also get information about recent events. So again, Kubectl describe pods gives us a lot of information both about the general state and readiness as well as some recent events. We also want to measure core metrics. Now the core…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.