Kubernetes Deployment

We deploy with kubernetes. In order to deploy your own network you have to install kubectl and get a kubernetes cluster.

We have tested two different kubernetes providers: Minikube and Digital Ocean.

Minikube

There are many Kubernetes providers, but if you're just getting started, Minikube is a tool that you can use to get your feet wet.

Install Minikube

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.

Follow the installation instruction below. If all the pods and services have settled and everything looks green in your minikube dashboard, expose the nitro-web service on your host system with:

$ minikube service nitro-web --namespace=human-connection

Digital Ocean

  1. At first, create a cluster on Digital Ocean.

  2. Download the config.yaml if the process has finished.

  3. Put the config file where you can find it later (preferable in your home directory under ~/.kube/)

  4. In the open terminal you can set the current config for the active session: export KUBECONFIG=~/.kube/THE-NAME-OF-YOUR-CLUSTER-kubeconfig.yaml. You could make this change permanent by adding the line to your .bashrc or ~/.config/fish/config.fish depending on your shell.

    Otherwise you would have to always add --kubeconfig ~/.kube/THE-NAME-OF-YOUR-CLUSTER-kubeconfig.yaml on every kubectl command that you are running.

  5. Now check if you can connect to the cluster and if its your newly created one by running: kubectl get nodes

If you got the steps right above and see your nodes you can continue.

First, install kubernetes dashboard:

$ kubectl apply -f dashboard/
$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/master/aio/deploy/recommended/kubernetes-dashboard.yaml

Get your token on the command line:

$ kubectl -n kube-system describe secret $(kubectl -n kube-system get secret | grep admin-user | awk '{print $1}')

It should print something like:

Name:         admin-user-token-6gl6l
Namespace:    kube-system
Labels:       <none>
Annotations:  kubernetes.io/service-account.name=admin-user
              kubernetes.io/service-account.uid=b16afba9-dfec-11e7-bbb9-901b0e532516

Type:  kubernetes.io/service-account-token

Data
====
ca.crt:     1025 bytes
namespace:  11 bytes
token:      eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJrdWJlcm5ldGVzL3NlcnZpY2VhY2NvdW50Iiwia3ViZXJuZXRlcy5pby9zZXJ2aWNlYWNjb3VudC9uYW1lc3BhY2UiOiJrdWJlLXN5c3RlbSIsImt1YmVybmV0ZXMuaW8vc2VydmljZWFjY291bnQvc2VjcmV0Lm5hbWUiOiJhZG1pbi11c2VyLXRva2VuLTZnbDZsIiwia3ViZXJuZXRlcy5pby9zZXJ2aWNlYWNjb3VudC9zZXJ2aWNlLWFjY291bnQubmFtZSI6ImFkbWluLXVzZXIiLCJrdWJlcm5ldGVzLmlvL3NlcnZpY2VhY2NvdW50L3NlcnZpY2UtYWNjb3VudC51aWQiOiJiMTZhZmJhOS1kZmVjLTExZTctYmJiOS05MDFiMGU1MzI1MTYiLCJzdWIiOiJzeXN0ZW06c2VydmljZWFjY291bnQ6a3ViZS1zeXN0ZW06YWRtaW4tdXNlciJ9.M70CU3lbu3PP4OjhFms8PVL5pQKj-jj4RNSLA4YmQfTXpPUuxqXjiTf094_Rzr0fgN_IVX6gC4fiNUL5ynx9KU-lkPfk0HnX8scxfJNzypL039mpGt0bbe1IXKSIRaq_9VW59Xz-yBUhycYcKPO9RM2Qa1Ax29nqNVko4vLn1_1wPqJ6XSq3GYI8anTzV8Fku4jasUwjrws6Cn6_sPEGmL54sq5R4Z5afUtv-mItTmqZZdxnkRqcJLlg2Y8WbCPogErbsaCDJoABQ7ppaqHetwfM_0yMun6ABOQbIwwl8pspJhpplKwyo700OSpvTT9zlBsu-b35lzXGBRHzv5g_RA

Proxy localhost to the remote kubernetes dashboard:

$ kubectl proxy

Grab the token from above and paste it into the login screen at http://localhost:8001/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/

Installation with kubernetes

You have to do some prerequisites e.g. change some secrets according to your own setup.

Edit secrets

$ cp secrets.template.yaml human-connection/secrets.yaml

Change all secrets as needed.

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=

Those secrets get base64 decoded in a kubernetes pod.

Create a namespace

$ kubectl apply -f namespace-human-connection.yaml

Switch to the namespace human-connection in your kubernetes dashboard.

Run the configuration

$ kubectl apply -f human-connection/

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.

Setup Ingress and HTTPS

Follow this quick start guide and install certmanager via helm and tiller:

$ kubectl create serviceaccount tiller --namespace=kube-system
$ kubectl create clusterrolebinding tiller-admin --serviceaccount=kube-system:tiller --clusterrole=cluster-admin
$ helm init --service-account=tiller
$ helm repo update
$ helm install stable/nginx-ingress
$ kubectl apply -f https://raw.githubusercontent.com/jetstack/cert-manager/release-0.6/deploy/manifests/00-crds.yaml
$ helm install --name cert-manager --namespace cert-manager stable/cert-manager

Create letsencrypt issuers. Change the email address in these files before running this command.

$ kubectl apply -f human-connection/https/

Create an ingress service in namespace human-connection. Change the domain name according to your needs:

$ kubectl apply -f human-connection/ingress/

Check the ingress server is working correctly:

$ curl -kivL -H 'Host: <DOMAIN_NAME>' 'https://<IP_ADDRESS>'

If the response looks good, configure your domain registrar for the new IP address and the domain.

Now let's get a valid HTTPS certificate. According to the tutorial above, check your tls certificate for staging:

$ kubectl describe --namespace=human-connection certificate tls
$ kubectl describe --namespace=human-connection secret tls

If everything looks good, update the issuer of your ingress. Change the annotation certmanager.k8s.io/issuer from letsencrypt-staging to letsencrypt-prod in your ingress configuration in human-connection/ingress/ingress.yaml.

$ kubectl apply -f human-connection/ingress/ingress.yaml

Delete the former secret to force a refresh:

$ kubectl  --namespace=human-connection delete secret tls

Now, HTTPS should be configured on your domain. Congrats.

Legacy data migration

This setup is completely optional and only required if you have data on a server which is running our legacy code and you want to import that data. It will import the uploads folder and migrate a dump of mongodb into neo4j.

Prepare migration of Human Connection legacy server

Create a configmap with the specific connection data of your legacy server:

$ kubectl create configmap maintenance-worker          \
  --namespace=human-connection                          \
  --from-literal=SSH_USERNAME=someuser                  \
  --from-literal=SSH_HOST=yourhost                      \
  --from-literal=MONGODB_USERNAME=hc-api                \
  --from-literal=MONGODB_PASSWORD=secretpassword        \
  --from-literal=MONGODB_AUTH_DB=hc_api                 \
  --from-literal=MONGODB_DATABASE=hc_api                \
  --from-literal=UPLOADS_DIRECTORY=/var/www/api/uploads \
  --from-literal=NEO4J_URI=bolt://localhost:7687

Create a secret with your public and private ssh keys. As the kubernetes documentation points out, you should be careful with your ssh keys. Anyone with access to your cluster will have access to your ssh keys. Better create a new pair with ssh-keygen and copy the public key to your legacy server with ssh-copy-id:

$ kubectl create secret generic ssh-keys          \
  --namespace=human-connection                    \
  --from-file=id_rsa=/path/to/.ssh/id_rsa         \
  --from-file=id_rsa.pub=/path/to/.ssh/id_rsa.pub \
  --from-file=known_hosts=/path/to/.ssh/known_hosts

Migrate legacy database

Patch the existing deployments to use a multi-container setup:

cd legacy-migration
kubectl apply -f volume-claim-mongo-export.yaml
kubectl patch --namespace=human-connection deployment nitro-backend --patch "$(cat deployment-backend.yaml)"
kubectl patch --namespace=human-connection deployment nitro-neo4j   --patch "$(cat deployment-neo4j.yaml)"
cd ..

Run the migration:

$ kubectl --namespace=human-connection get pods
# change <POD_IDs> below
$ kubectl --namespace=human-connection exec -it nitro-neo4j-65bbdb597c-nc2lv migrate
$ kubectl --namespace=human-connection exec -it nitro-backend-c6cc5ff69-8h96z sync_uploads

Last updated