Kubescope on the command line
Kubescope CLI can be used in Kubernetes or locally because it connects to the docker socket to collect stats.
Kubescope needs access to the docker socket on the host, stdin and a tty. It also needs to be running on the same host that you want to monitor. Here's an example pod spec:
apiVersion: v1
kind: Pod
metadata:
name: kubescope-cli-example
spec:
containers:
- image: hharnisc/kube-scope-cli:9c02f636b8cb6d4f251af3e73d19248c2a31b3e1
name: kubescope-cli-example
stdin: true
tty: true
volumeMounts:
- mountPath: /var/run/docker.sock
name: docker-sock-volume
env:
- name: MATCH_NAME
value: ".*my-app.*"
- name: SKIP_NAME
value: ".*POD.*"
# - name: MATCH_IMAGE
# value: ".*my-image.*"
# - name: SKIP_IMAGE
# value: ".*some-image.*"
# - name: DOCKER_SOCKET
# value: /var/run/docker.sock
volumes:
- name: docker-sock-volume
hostPath:
path: /var/run/docker.sock
# pin kubescope to a specific node
# nodeSelector:
# kubernetes.io/hostname: minikube
Attach to the running container:
kubectl attach -it kubescope-cli-example
Observe graphs and the detatch with escape sequence Ctrl+P
followed by Ctrl+Q
Kubescope CLI can be used locally and connects to the docker socket to collect stats
MATCH_NAME=.*my_app.* npm start