Kubernetes
You're viewing a free preview. Sign in free to read every script.
Why this exists Every image a node ever pulled stays in its local cache until the kubelet's own garbage collector decides to remove it (triggered by disk pressure, not by "is anyone using this"), so a cluster that has…
Cluster Image Inventory: What's Actually Running, Across Every Node
A report of every image cached on every node, cross-referenced with what a currently running Pod actually uses, with no SSH and no node access — run for real on a three-node cluster carrying a full session's worth of accumulated images, correctly separating in-use system images from about a gigabyte of cached-but-unused leftovers.
Sign in free to read this script, along with the rest of the library.
Sign in free