World map

 

BM: Baca dalam Bahasa Melayu

World K9: Cluster Admin

crictl: containers under the kubelet

You will learn: how to see a node’s pods, containers and logs straight from the stove.

Checked with Kubernetes 1.37 on 1 Nov 2026

The problem

You are here: cluster kind-dev namespace default
The oven pod on dev-worker2 is in CrashLoopBackOff, with 2 restarts. Terminal output: $ kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES oven 0/1 CrashLoopBackOff 2 (13s ago) 26s 10.244.2.3 dev-worker2 <none> <none>

The oven tray keeps crashing on dev-worker2. On a bad night, the office is down too. Then kubectl sees nothing. How do you look anyway?

Try it

Go straight to the stove with crictl: A torch to look straight at the stove: it lists the pods and containers containerd runs on one node.. Type docker exec -it dev-worker2 bash, then crictl pods.

You are here: cluster kind-dev namespace default
Stalls: 3. Trays: 1 (0 ready). Waiting for a stall: 0.
dev-control-planeOpenManager

No trays here.

dev-workerOpen

No trays here.

dev-worker2Open
  • oven↻2CrashLoopBackOff
Time: +0s (sped up)

Try it

crictl ps -a lists containers, stopped ones too. Tap the container that exited.

Tap the container that exited.

0 of 1 found

Try it

Read the oven’s last words. Type crictl ps -a --name oven. Copy the CONTAINER id. Then type crictl logs and paste it.

You are here: cluster kind-dev namespace default
Stalls: 3. Trays: 1 (0 ready). Waiting for a stall: 0.
dev-control-planeOpenManager

No trays here.

dev-workerOpen

No trays here.

dev-worker2Open
  • oven↻2CrashLoopBackOff
Time: +0s (sped up)

You are inside node dev-worker2. Type exit to go back to your laptop.

Predict

The API server is down. Which command still shows the oven’s last words?

Pick one answer.

Check

Which tool fits? Sort each job.

Drag a card into a box. Or tap the card, then tap the box.

      Check

      Put the steps in order: why does a tray crash?

      First step at the top. Put these in order. Use the arrows, or drag.

      • Step inside: ssh dev-worker2
      • Find its stall: kubectl get pods -o wide
      • Find the container: crictl ps -a --name oven
      • Read its last words: crictl logs <id>
      • Leave the stall: exit

      You did it!

      You can now see a node’s containers and logs with crictl.

      Go deeper (optional)

      Use crictl to look. The kubelet runs the containers, so let the kubelet make the changes. crictl images lists what the stove has pulled. crictl inspect <id> shows one container in full.

      Real mission

      1. kubectl config current-context
      2. docker exec dev-worker crictl pods
      3. docker exec dev-worker crictl ps -a
      4. docker exec dev-worker crictl images

      0 of 4 steps done · Your ticks are saved.