-
Book Overview & Buying
-
Table Of Contents
-
Feedback & Rating
Cracking the Kubernetes Interview
By :
A running, connected application can still fail under load if it doesn’t have the resources it needs to perform well. This brings us to our final, and often most complex, area of troubleshooting: performance. Here, we move from asking “Is it working?” to “Is it working efficiently?” We’ll investigate what happens when your application runs out of memory or CPU and how to ensure it scales to meet demand.
This is a critical SRE and DevOps topic. Questions about resource management test your understanding of how to build stable, predictable systems. An interviewer wants to know that you can configure appropriate resource limits to prevent a single memory-hungry application from destabilizing an entire node. Your ability to diagnose an OOMKilled event shows you can connect the dots between Kubernetes-level metrics and application-level behavior.