Kubernetes Manifest Check
By Yuki Tanaka · Published · Updated
Resource limits, probes, and the liveness config that restarts healthy pods.
Tags: Engineering, Technical, Coding
Prompt template
``` [textarea: Paste deployment, service, and anything related] ``` What the app does: [textarea: Including startup time and traffic pattern] Cluster: [textarea: Managed service, node types, size] Symptoms: [textarea: Restarts, OOM kills, slow rollouts, or nothing yet] Environment: [dropdown: Development, Staging, Production] Check requests and limits against how the app actually behaves, since wrong values cause both throttling and eviction and the symptoms look nothing alike. Look hard at the probes. A liveness probe configured like a readiness probe will restart a healthy pod that is merely slow, and that failure mode is miserable to debug. Then rolling update settings, disruption budgets, anti-affinity, and the security context. Give me the corrected manifests.