A configuration for Camunda Platform 8 that uses NGINX as an Ingress Controller.
This folder contains a Helm values file for installing the Camunda Platform Helm Chart on an existing GKE cluster (if you don't have one yet, see Kubernetes Installation for Camunda 8 on GCP). A Makefile is provided to automate the installation process.
If this is your first time here, make sure you have installed the prerequisites.
After you've installed the prerequisites, follow these steps:
Open a Terminal and cd into this directory
Edit the Makefile found in this directory and set the following bash variables so that they are appropriate for your specific environment.
If you don't have a Kubernetes cluster, the values provided will be used to create a new cluster. Otherwise, the values are used to connect and manage an existing cluster.
project ?= camunda-researchanddevelopment
region ?= us-east1-b # see: https://cloud.withgoogle.com/region-picker/
clusterName ?= CLUSTER_NAME
# Azure provides temporary dns names such as: MY_DOMAIN_NAME.region.cloudapp.azure.com
# However GCP does not. DNS names will be configured using https://nip.io
# dnsLabel ?= MY_DOMAIN_NAME
machineType ?= n1-standard-16
minSize ?= 1
maxSize ?= 6
certEmail ?= [email protected]
If you need to create a new GKE Cluster, run make kube.
Once you have a GKE Cluster, run make to do the following:
- Set up a Kubernetes letsencrypt certificate manager
- Install a Kubernetes Nginx Ingress Controller. A corresponding GCP Load Balancer is provisioned automatically
- Attempt to find the ip address of the Load Balancer. This ip address is then used generate a
camunda-values.yamlfile. - Helm is used to install Camunda 8 using the
camunda-values.yamlfile with the Load Balancer IP Address - The ingress controller is annotated so that letsencrypt tls certificates are provisioned.
You can re-install this profile easily. First run make clean to remove all kubernetes objects created by make. Then, re-run make to re-install.
WARNING!!! This will completely destroy your cluster and everything inside of it!!! To completely delete your cluster, run make clean-kube.
See the main README for Troubleshooting, Tips, and Tricks