Application startup
- Clone repo:
git clone https://github.com/setevoy2/keygrav.git - Start minikube:
minikube start - Create deployment:
kubectl create -f ci/keygrav.yml - Start service:
kubectl expose deployment keygrav --type=NodePort - Check it with curl:
curl $(minikube service keygrav --url)
Project description
All build files located in the ci directory:
-
keygrav.yml - pods deploy config
-
jenkins.groovy - Jenkins pipeline plugin's build script (functions only)
-
keybuild.groovy - Jenkins build script (build&&push image to the DockerHub)
-
Dockerfile in the root - to build images.
Links
DockerHub repo: https://hub.docker.com/r/setevoy/keygrav/
Jenkins CI: http://keyjenkins.setevoy.org.ua/job/KeyGravPoC/ (In order to access Jenkin's host - please, provide me your IP (to add it to the Jenkin's EC2 security group rules) and use credentials from an email)
Questions
how you would handle data persistence of the grav container in the future.
If Grav have ability to use persistent storage (and it have, judging the [Docker-Grav image's documentation] (https://hub.docker.com/r/garywiz/docker-grav/)) - then an AWS EFS directory (from an AWS EBS volume) can be mounted to the Kubernetes cluster and then PersistentVolumes configured to attach it to the containers.
I also added few points related to the deployment in the kuberDeploy() function of the jenkins.groovy build script.
But again - not sure how "elegant" they are, as this is my first Kubernetes experience.
I also described my build process in my blog, if no objections (in rus), it's available here>>>.