Score:0

Better way at scale to pull image uris from all pod specs on k8s cluster

gr flag

Team, I see this [list-all-running-container-image]https://kubernetes.io/docs/tasks/access-application-cluster/list-all-running-container-images/. however, I cannot bank operations on this because it is very heavy operation as I have 25K pods on cluster.

so what is the better performance approach to get all the image uri's from all specs of a pod? or from pod logs when container is pulled.

Goal is count the number of images that are pulled and their frequency. what would be better solution to do this instead of kubectl command?

if this portal is not for such question, please help reroute. Once I have solution, then I will plan on storing it later and how.

Score:1
gr flag

One way I figured out is instead of querying pods, query the node status itself that shows all the images pulled however it still does not show me time it was pulled. if anyone has a better solution, let know.

Kubectl get node node1 -o json | jq .status.images

[
  {
    "names": [
      "nbt.io/tb/gpu-tests@sha256:df989f2a6418bb5fb8af67b20facbd503b92e538489f50ba0a",
      "nbt.io/tb/gpu-tests:20210616-1706"
    ],
    "sizeBytes": 11701534918
  },
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.