Skip to content
This repository was archived by the owner on Dec 13, 2023. It is now read-only.

README.md

Latest commit

 

History

History
43 lines (29 loc) · 1.59 KB

README.md

File metadata and controls

43 lines (29 loc) · 1.59 KB

Replica reloader

reuse compliant

Introduction

The replica-reloader is a simple controller that watches a Kubernetes Deployment and starts a COMMAND and adds the Deployment replica count as last argument:

$(terminal-1) kubectl create deployment --image=nginx nginx
deployment.apps/nginx created

and running

$(terminal-2) replica-reloader --namespace=default --deployment-name=nginx -- sleep

would start a sleep 1 process. If the watched deployment is scaled, then the controller stops the previous process and starts a new one:

$(terminal-1) kubectl scale deployment my-dep --replicas=10
deployment.apps/my-dep scaled

$(terminal-1) ps | grep sleep
61191 ttys003    0:00.00 sleep 10`,

Why it was created

It was originally created as to be used in situations where you want to run some process with static count of replicas as flag (e.g. apiserver-network-proxy that has a --server-count flag). In some, cases the process must be restarted with the new server count.

Docker images

Docker images for the replica-reloader and replica-reloader bundled with apiserver-network-proxy are available at:

  • eu.gcr.io/gardener-project/gardener/replica-reloader:latest
  • eu.gcr.io/gardener-project/gardener/replica-reloader:v0.2.0
  • eu.gcr.io/gardener-project/gardener/replica-reloader:v0.2.0-konnectivity-server-v0.0.14
  • eu.gcr.io/gardener-project/gardener/replica-reloader:v0.2.0-konnectivity-server-v0.0.15