Skip to content

Repo containing an example set of manifests and docs to deploy Netbox (https://github.com/netbox-community/netbox) on Kubernetes

License

Notifications You must be signed in to change notification settings

sdktr/netbox-kubernetes

 
 

Repository files navigation

NetBox

Alternative installation instructions for NetBox using Kubernetes. The main value is the netbox configurations but postgresql and redis instructions briefly provided via helm charts.

Notes

  • The netbox deployment will work with existing redis and postgresql installations, just ensure that the configurations are changed in netbox-configmap.yaml and netbox-deployment.yaml
  • Relevant places that will need to be changed are indicated by the inline comment #changeme
  • The manifests were tested against the following:
    • NetBox: 2.6.3
    • Postgresql: 11.4.0
    • Redis: 5.0.5
    • Kubernetes: 1.13.2

Installation

To get NetBox up and running on a Kubernetes cluster:

  1. Deploy netbox-namespace.yaml
    • If the namespace is changed, ensure all manifests are updated accordingly
  2. Deploy Postgres. One way to do this is via a helm chart. Make sure to change the password! helm install --name netbox-community --namespace netbox-community stable/postgresql --set postgresqlPassword=changeme,postgresqlDatabase=netbox,postgresqlUsername=netbox,persistence.enabled=true,persistence.storageClass=rook-ceph-block,persistence.size=10Gi
  3. Deploy Redis. One way to do this is via a helm chart: helm install --name netbox-redis --namespace netbox-community stable/redis -f redis-values.yaml
  4. Change the values you need in the netbox-configMap.yaml and netbox-secrets.yaml
    • This would include any host, DB name, user changes that would be needed
    • The AUTH_LDAP_BIND_PASSWORD password would also need to change depending on the AUTH_LDAP_BIND_DN chosen
    • The AUTH_LDAP_BIND_DN is the Distinguished Name (DN) of the service account that will be making queries when a user requests login
    • Update the AUTH_LDAP_SERVER_URI to point to your LDAP server
    • If the DNS resolution for the LDAP server doesn't work use the IP address of your LDAP server
    • Any password changes will need to be reflected in netbox-secrets.yaml
    • Change appropriate values for DB, Email, and Redis to reflect your environment
  5. Edit the volumeMount for the media-files. To allow for HA across your pods, ensure that the mount allows for multiple pods to attach to it, such as through the use of a shared filesystem kubectl apply -f netbox-deployment.yaml
    • NetBox will initialize the needed tables into the DB
    • The readiness probe will check and ensure that the service is up and running
  6. kubectl apply -f netbox-service.yaml
    • Usual deployment is via an Ingress but if needed, change the service to use NodePort in order to access it from outside the cluster

Verification

  1. A netbox instance is up and running with multiple pods
  2. A PostgreSQL DB for netbox is up and running with all the necessary schema
  3. Redis caching enabled and working

Deploying an Ingress (Optional)

This isn't necessary for just testing. If you don't want to deploy the ingress resource, you should switch the service to use a NodePort so you can access it. nginx-ingress was used as the Ingress Controller in this example

  1. Change the netbox-ingress.yaml file to your subdomain and TLS certificate for https
  2. kubectl apply -f netbox-ingress.yaml

Verification

  1. You can navigate to your netbox instance via a FQDN
  2. https enabled with the certificate configured

Deploying Metrics (Optional)

The manifests were generated against a prometheus/grafana deployment deployed via the kube-prometheus project. Consider checking them out at https://github.com/coreos/kube-prometheus

  1. kubectl apply -f netbox-monitoring-rbac.yaml
    • Necessary if RBAC is enabled on the cluster
  2. kubectl apply -f netbox-monitoring.yaml
    • This will create a ServiceMonitor object that you can see when navigating to Prometheus/Targets
  3. Grafana dashboard # 9528 can be used to display the exposed Django metrics: https://grafana.com/grafana/dashboards/9528

Verification

  1. Prometheus target exists and is reporting ready for all pods deployed
  2. Dashboard should report information regarding the metrics from Django backend

References

About

Repo containing an example set of manifests and docs to deploy Netbox (https://github.com/netbox-community/netbox) on Kubernetes

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published