6fed4797ed
The recommended way to update a kubernetes deployment to a new image is to change the image tag. However, our build server is configured to push every commit of the `master` branch to docker hub to the respective repository and tag `:latest`. So adding some configuration that can be changed seems to be a trick to re-deploy the `:latest` image. See here: https://stackoverflow.com/a/51835397
Human-Connection Nitro | Deployment Configuration
Currently the deployment is not primetime ready as you still have to do some manual work. That we need to change, the following list gives some glimpse of the missing steps.
Todo`s
- check labels and selectors if they all are correct
- configure NGINX from yml
- configure Let's Encrypt cert-manager from yml
- configure ingress form yml
- configure persistent & shared storage between nodes
- reproduce setup locally
The dummy directory has some lb configurations that did not work properly on Digital Ocean but could be used as a starting point for getting it right
Install Minikube, kubectl
There are many Kubernetes distributions, but if you're just getting started, Minikube is a tool that you can use to get your feet wet.
Open minikube dashboard
$ minikube dashboard
This will give you an overview. Some of the steps below need some timing to make ressources available to other dependent deployments. Keeping an eye on the dashboard is a great way to check that.
Create a namespace locally
$ kubectl create -f namespace-staging.yml
Switch to the namespace staging in your kubernetes dashboard.
Setup config maps
$ cp db-migration-worker.template.yml config/db-migration-worker.yml
# edit all variables according to the setup of the remote legacy server
$ kubectl apply -f config/
Setup secrets and deploy themn
If you want to edit secrets, you have to base64 encode them. See kubernetes
documentation.
# example how to base64 a string:
$ echo -n 'admin' | base64
YWRtaW4=
$ cp secrets.yml.template secrets.yml
# change all variables as needed and deploy them
$ kubectl apply -f secrets.yml
Create volumes
$ kubectl apply -f volumes/
Expose the services
$ kubectl apply -f services/
Wait until persistent volumes and services become available.
Create deployments
$ kubectl apply -f deployments/
This can take a while because kubernetes will download the docker images.
Sit back and relax and have a look into your kubernetes dashboard.
Wait until all pods turn green and they don't show a warning
Waiting: ContainerCreating anymore.
Access the services
$ minikube service nitro-web --namespace=staging
Provision db-migration-worker
Copy your private ssh key and the .known-hosts file of your remote legacy server.
# check the corresponding db-migration-worker pod
$ kubectl --namespace=staging get pods
# change <POD_ID> below
$ kubectl cp path/to/your/ssh/keys/.ssh staging/nitro-db-migration-worker-<POD_ID>:/root/
Run the migration:
# change <POD_IDs> below
$ kubectl --namespace=staging exec -it nitro-db-migration-worker-<POD_ID> ./import.sh
$ kubectl --namespace=staging exec -it nitro-neo4j-<POD_ID> ./import/import.sh