ForgeOps

Deploy using Kustomize on GKE, EKS, or AKS

  1. Verify that you have set up your environment and created a Kubernetes cluster as documented in the setup section.

  2. Identify Docker images to deploy:

    • If you want to use custom Docker images for the platform, update the image defaulter file with image names and tags generated by the forgeops build command. The image defaulter file is located at /path/to/forgeops/kustomize/deploy/image-defaulter/kustomization.yaml.

      You can get the image names and tags from the image defaulter file on the system on which the customized Docker images were developed.

    • If you want to use ForgeOps-provided evaluation Docker images for the platform, do not modify the image defaulter file.

  3. Set up your Kubernetes context:

    1. Set the KUBECONFIG environment variable so that your Kubernetes context references the cluster in which you’ll perform the ForgeOps deployment.

    2. Create a Kubernetes namespace in the cluster for the Ping Identity Platform pods:

      $ kubectl create namespace my-namespace

    3. Set the active namespace in your Kubernetes context to the Kubernetes namespace you just created:

      $ kubens my-namespace

  4. Run the forgeops install command. For example, to install a small-sized deployment:

    $ cd /path/to/forgeops/bin
$ ./forgeops install --deployment-size --fqdn forgeops.example.com --namespace my-namespace

    The forgeops install command examines the image defaulter file to determine which Docker images to use.

    For a single-instance deployment, specify --deployment-size as --cdk.

    For small, medium, and large deployments, specify --deployment-size as --small, --medium, or --large.

    For more information about single-instance deployments and deployment sizes, refer to Cluster and deployment sizes.

    If you prefer not to deploy using a single forgeops install command, refer to Alternative deployment techniques when using Kustomize for more information.

    Ping Identity only offers its software or services to legal entities that have entered into a binding license agreement with Ping Identity. When you install Docker images provided by ForgeOps, you agree either that: 1) you are an authorized user of a Ping Identity Platform customer that has entered into a license agreement with Ping Identity governing your use of the Ping Identity software; or 2) your use of the Ping Identity Platform software is subject to the Ping Identity Subscription Agreements.

  5. Check the status of the pods in the namespace in which you deployed the platform until all the pods are ready:

    1. Run the kubectl get pods command.

    2. Review the output. Deployment is complete when:

      • All entries in the STATUS column indicate Running or Completed.

      • The READY column indicates all running containers are available. The entry in the READY column represents [total number of containers/number of available containers].

    3. If necessary, continue to query your deployment’s status until all the pods are ready.

  6. Back up and save the Kubernetes secrets that contain the master and TLS keys:

    1. To avoid accidentally putting the backups under version control, change to a directory that is outside your forgeops repository clone.

    2. The ds-master-keypair secret contains the DS master key. This key is required to decrypt data from a directory backup. Failure to save this key could result in data loss.

      Back up the Kubernetes secret that contains the DS master key:

      $ kubectl get secret ds-master-keypair -o yaml > master-key-pair.yaml

    3. The ds-ssl-keypair secret contains the DS TLS key. This key is needed for cross-environment replication topologies.

      Back up the Kubernetes secret that contains the DS TLS key pair:

      $ kubectl get secret ds-ssl-keypair -o yaml > tls-key-pair.yaml

    4. Save the two backup files.

  7. (Optional) Deploy Prometheus, Grafana, and Alertmanager for monitoring and alerting[1]:

    1. Deploy Prometheus, Grafana, and Alertmanager pods in your ForgeOps deployment:

      $ /path/to/forgeops/bin/prometheus-deploy.sh

**This script requires Helm version 3.04 or later due to changes in the behaviour of 'helm repo add' command.**

namespace/monitoring created
"stable" has been added to your repositories
"prometheus-community" has been added to your repositories
Hang tight while we grab the latest from your chart repositories...
...Successfully got an update from the "ingress-nginx" chart repository
...Successfully got an update from the "codecentric" chart repository
...Successfully got an update from the "prometheus-community" chart repository
...Successfully got an update from the "stable" chart repository
Update Complete. ⎈Happy Helming!⎈
Release "prometheus-operator" does not exist. Installing it now.
NAME: prometheus-operator
LAST DEPLOYED: ...
NAMESPACE: monitoring
STATUS: deployed
REVISION: 1
NOTES:
kube-prometheus-stack has been installed. Check its status by running:
  kubectl --namespace monitoring get pods -l "release=prometheus-operator"

Visit https://github.com/prometheus-operator/kube-prometheus for instructions on how to create & configure Alertmanager and Prometheus instances using the Operator.
...
Release "forgerock-metrics" does not exist. Installing it now.
NAME: forgerock-metrics
LAST DEPLOYED: ...
NAMESPACE: monitoring
STATUS: deployed
REVISION: 1
TEST SUITE: None

    2. Check the status of the pods in the monitoring namespace until all the pods are ready:

      $ kubectl get pods --namespace monitoring
NAME                                                     READY   STATUS    RESTARTS   AGE
alertmanager-prometheus-operator-kube-p-alertmanager-0   2/2     Running   0          119s
prometheus-operator-grafana-95b8f5b7d-nn65h              3/3     Running   0          2m4s
prometheus-operator-kube-p-operator-7d54989595-pdj44     1/1     Running   0          2m4s
prometheus-operator-kube-state-metrics-d95996bc4-wcf7s   1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-67xq4       1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-b4grn       1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-cwhcn       1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-h9brd       1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-q8zrk       1/1     Running   0          2m4s
prometheus-operator-prometheus-node-exporter-vqpt5       1/1     Running   0          2m4s
prometheus-prometheus-operator-kube-p-prometheus-0       2/2     Running   0          119s

  8. (Optional) Install a TLS certificate instead of using the default self-signed certificate in your ForgeOps deployment. Refer to TLS certificate for details.

Alternative deployment techniques when using Kustomize

Staged deployments

If you prefer not to perform a ForgeOps Kustomize deployment using a single forgeops install command, you can deploy the platform in stages, component by component, instead of deploying with a single command. Staging deployments can be useful if you need to troubleshoot a deployment issue.

Generating Kustomize manifests and using kubectl apply -k commands

You can generate Kustomize manifests, and then deploy the platform using the kubectl apply -k command.

The forgeops install command generates Kustomize manifests that let you recreate your ForgeOps deployment. The manifests are written to the /path/to/forgeops/kustomize/deploy directory of your forgeops repository clone. Advanced users who prefer to work directly with Kustomize manifests that describe their ForgeOps deployment can use the generated content in the kustomize/deploy directory as an alternative to using the forgeops command:

  1. Generate an initial set of Kustomize manifests by running the forgeops install command. If you prefer to generate the manifests without installing the platform, you can run the forgeops generate command instead of the forgeops install command.

  2. Run kubectl apply -k commands to deploy and remove platform components. Specify a manifest in the kustomize/deploy directory as an argument when you run kubectl apply -k commands.

    1. Use GitOps to manage configuration changes to the kustomize/deploy directory instead of making changes to files in the kustomize/base and kustomize/overlay directories.

Next step

1. Installing Prometheus, Grafana, and Alertmanager technology in ForgeOps deployments provides an example of how you might set up monitoring and alerting in a Ping Identity Platform deployment in the cloud. Remember, ForgeOps deployments are reference implementations and not for production use. When you create a project plan, you’ll need to determine how to monitor and send alerts in your production deployment.