Skip to main content

Expose your local TCP server as Kubernetes service in remote cluster.

Project description


as in ngrok

Expose your local TCP server as Kubernetes service in remote cluster.


Giving we want to create service in namespace with name service_name on port service_port forwarding all traffic from it to local_host (usually and local_port.

  1. Install krok client: pip install krok

  2. Install krok server: kubectl -n <namespace> run --image=smpio/krok-server krok

  3. Run krok: krok -n <namespace> -l <local_host> -p <local_port> <service_name> <service_port>

Note: local_host can be any host reachable from your machine, but usually is the most useful.

How it works

We have running krok server in the pod krok_pod in namespace. The server is simple OpenSSH server that allows to forward incoming connections.

Krok client is simple script that automates the following steps:

  1. Run kubectl port-forward <krok_pod> :22. It will listen on random local_ssh_port forwarding all connections to krok's OpenSSH server on krok_pod.

  2. Run ssh -N -R *:0:<local_host>:<local_port> -p <local_ssh_port> krok@localhost. This causes krok's OpenSSH server on krok_pod to listen on random pod_port forwarding all connections to local_host:local_port using SSH tunnel.

  3. Create or update service service_name with port service_port and targetPort pod_port, with selector matching krok_pod.

In the result, all connections to service_name:service_port will be forwarded to krok_pod, then to your local machine via SSH tunnel forwarded by kubectl port-forward, and then forwarded by SSH client to local_host:local_port.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

krok-1.0.0.tar.gz (4.4 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page