Skip to content

Part 2: Installing NMaaS

Once a working Kubernetes cluster has been deployed, we are ready to proceed to the next step - installing NMaaS.

All the necessary components will be installed in a single namespace – nmaas-system. If this namespace has not been created so far, execute:

kubectl create namespace nmaas-system

GitLab Installation

The first NMaaS dependency that we will set up is GitLab, a self-hosted web based Git repository hosting service. Many applications that are deployed by NMaaS users store their configuration data in a Git repository, allowing easier editing, and version management.

GitLab has an official Helm chart, and we will use it to create a basic GitLab installation locally. Some parameters must be customized in the values.yaml file before deployment:

  • global.hosts.domain – should be set to the domain that will be allocated to GitLab. Note that the final hostname where GitLab will be reachable will have a gitlab prepended to it. If nmaas.example.local is set as the global.hosts.domain parameter, then GitLab will be available on gitlab.nmaas.example.local.
  • global.hosts.ssh – in order for users to be able to interact with their GitLab repositories via SSH, the value of global.hosts.ssh should be set to the MetalLB IP that will be assigned to this new service (usually the next available one) for the gitlab-shell component. If the IP is not known at the time of deployment, then after the initial deployment, once the LoadBalancer service is created and the IP is allocated, a chart upgrade can be performed, where the global.hosts.ssh parameter will be set to the appropriate value.
  • global.ingress.tls.secretName – an existing Kubernetes TLS secret where the TLS certificate to be used is stored.
  • global.ingress.annotations.kubernetes.io/ingress.class – should be set to the ingress class used by the deployed ingress-nginx instance. In case of MicroK8s this should be set to public. In case of K3s, it should be set to nginx.
  • optionally, if an email server is available, the global.smtp section can be edited with the appropriate parameters so that outbound email is enabled.
gitlab-values.yaml
certmanager:
  install: false
nginx-ingress:
  enabled: false
prometheus:
  install: false
redis:
  install: true
postgresql:
  postgresqlUsername: gitlab
  install: true
  postgresqlDatabase: gitlabhq_production
  usePasswordFile: false
  existingSecret: 'gitlab-postgresql'
  master:
    extraVolumeMounts:
      - name: custom-init-scripts
        mountPath: /docker-entrypoint-preinitdb.d/init_revision.sh
        subPath: init_revision.sh
    podAnnotations:
      postgresql.gitlab/init-revision: "1"
  metrics:
    enabled: false
gitlab-runner:
  install: false
gitlab-shell:
  service:
    type: LoadBalancer
global:
  edition: ce
  hosts:
    domain: nmaas.<INGRESS_IP>.nip.io
    https: true
    ssh: <LB_SSH_IP>.nip.io
  ingress:
    enabled: true
    configureCertmanager: false
    tls:
      enabled: true
      # secretName: <EXISTING_OR_DUMMY_TLS_SECRET_NAME> # can be left empty, self-signed certificates will be generated
    path: /
    annotations:
      kubernetes.io/ingress.class: "public"
  initialRootPassword:
    secret: gitlab-root-password
    key: password
  appConfig:
    defaultProjectFeatures:
      builds: false
  time_zone: Europe/Warsaw
  smtp:
    enabled: false
    address: smtp.mailgun.org
    port: 2525
    user_name: ""
    ## doc/installation/secrets.md#smtp-password
    password:
      secret: "my-smtp-secret"
      key: password
    # domain:
    authentication: "plain"
    starttls_auto: false
    openssl_verify_mode: "peer"
## doc/installation/deployment.md#outgoing-email
## Email persona used in email sent by GitLab
  email:
    from: ''
    display_name: GitLab
    reply_to: ''
    subject_suffix: ''
    smime:
      enabled: false
      secretName: ""
      keyName: "tls.key"
      certName: "tls.crt"

GitLab requires the deployment of a PostgreSQL instance. The necessary secrets containing the PostgreSQL passwords need to be created, as well as the secret containing the initial root GitLab password:

kubectl create secret generic -n <NAMESPACE> gitlab-postgresql --from-literal=postgresql-password=<POSTGRESQL_USER_PASSWORD> --from-literal=postgresql-postgres-password=<POSTGRESQL_ROOT_PASSWORD>
kubectl create secret generic -n <NAMESPACE> gitlab-root-password --from-literal=password=<GITLAB_ROOT_PASSWORD>

The root GitLab password will be used for login to the GitLab web interface.

We are ready to add the GitLab Helm repository and install the 4.X version of GitLab:

1
2
3
helm repo add gitlab https://charts.gitlab.io
helm repo update
helm install -f gitlab.yaml --namespace nmaas-system nmaas-gitlab --version 4.12.13 gitlab/gitlab

Once GitLab has been deployed, it should be possible to navigate to the login page using a web browser. After logging in, users are advised to configure the following settings:

  • disable new user registrations (Admin Area -> Settings -> General -> Sign-up restrictions)
    • Sign-up enabled should be unchecked
    • Require admin approval for new sign-ups should be unchecked
  • enable webhooks to local addresses (Admin Area -> Settings -> Network -> Outbound requests)
    • Allow requests to the local network from web hooks and services = checked
    • Allow requests to the local network from system hooks = checked

The final step before installing NMaaS itself is to generate a GitLab personal access token which will allow NMaaS to connect to the GitLab API. This can be done from the User Profile page:

1
2
- Click on the user avatar in the right-hand corner of the screen, Edit Profile. Select Access Tokens from the left-hand navigation menu. Give a new name for the authentication token, as well as an optional expiry date. Check all scopes.
- Store the token until the next section, where we will create a new secret containing it.

NMaaS Installation

The final step is to install NMaaS. NMaaS uses SSH communication to connect between components, so we need to create an SSH key pair and store it in a Kubernetes secret. This can be done by executing the following commands:

1
2
3
4
5
6
7
8
9
#!/bin/bash
tmpdir=$(mktemp -d)
ssh-keygen -f $tmpdir/key -N ""

# nmaas-helm-key-private should be replaced with {{ .Values.global.helmAccessKeyPrivate }}
kubectl create secret generic nmaas-helm-key-private -n <NMAAS_NAMESPACE> --from-file=id_rsa=$tmpdir/key

# nmaas-helm-key-private should be replaced with {{ .Values.global.helmAccessKeyPublic }}
kubectl create secret generic nmaas-helm-key-public -n <NMAAS_NAMESPACE> --from-file=helm=$tmpdir/key.pub

A few parameters need to be customized in the values.yaml file, to reflect the environment where NMaaS is deployed.

  • global.wildcardCertificateName – the name of the secret containing the TLS certificate to be used to secure the HTTP communication
  • global.nmaasDomain – the hostname where NMaaS will be accessible.
  • platform.properties.adminEmail – the email address which will receive various notifications such as new user sign-up, deployment errors, new application versions...
  • platform.adminPassword.literal – the password used to login as the admin user in the NMaaS Portal.
  • platform.properties.k8s.ingress.certificate.issuerOrWildcardName – the name of the wilcard certificate to be used for customer deployed applications, or the name of the cert-manager issuer to use if certificates are issued ad-hoc.
  • platform.properties.k8s.ingress.controller.ingressClass – the ingress class to be used for deployed applications. Should be set to nginx in the case of K3s and public in the case of MicroK8s.
  • platform.properties.k8s.ingress.controller.publicIngressClass – the ingress class to be used for applications where the users have explicitly selected to enable public access (e.g. without a VPN). Since this is a local deployment, the value of this parameter should equal the value set in platform.properties.k8s.ingress.controller.ingressClass.
  • publicServiceDomain, externalServiceDomain – for a local deployment this parameter should be set to the same value as global.nmaasDomain.
  • janitor.properties.gitlabToken.literal – the value of the personal access token created in GitLab, previously.
nmaas-values.yaml
global:
  acmeIssuer: false
  wildcardCertificateName: <EXISTING_OR_DUMMY_TLS_SECRET_NAME>
  ingressName: public
  nmaasDomain: nmaas.<INGRESS_IP>.nip.io
  demoDeployment: true

platform:
  image:
    tag: 1.4.4-SNAPSHOT
  adminPassword:
    literal: saamn
  apiSecret:
    literal: saamn
  initscripts:
    image:
      tag: 1.4.4
  properties:
    gitlab:
      host: nmaas-gitlab-webservice-default
    sso:
      encrpytionSecret:
        literal: saamn
    adminEmail: noreply@nmaas.local
    k8s:
      ingress:
        certificate:
          issuerOrWildcardName: <EXISTING_OR_DUMMY_TLS_SECRET_NAME>
        controller:
          ingressClass: public
          publicIngresClass: public
          publicServiceDomain: nmaas.<INGRESS_IP>.nip.io
          externalServiceDomain: nmaas.<INGRESS_IP>.nip.io

portal:
  image:
    tag: 1.4.4-SNAPSHOT

janitor:
  image:
    tag: 1.4.4
  properties:
    gitlabToken:
      literal: <GITLAB_ACCESS_TOKEN>
    gitlabApiUrl: http://nmaas-gitlab-webservice-default:8181/api/v4

Once the values.yaml file has been customized, NMaaS can be deployed by executing:

helm repo add nmaas https://artifactory.software.geant.org/artifactory/nmaas-helm
helm install -f values.yaml --namespace nmaas-system nmaas --version 1.1.2 nmaas/nmaas

NMaaS also requires an the stakater autoreloader component, which can simply be installed using the commands below. This component takes care of restarting the affected pods whenever a configuration change is submitted via GitLab.

1
2
3
helm repo add stakater https://stakater.github.io/stakater-charts
helm repo update
helm install config-reload --namespace nmaas-system stakater/reloader

After the installation, login as the admin user should be possible with the configured password.

Preview of the NMaaS catalog of applications