Squest Operator

Deploy Squest in a controlled way using an Operator.

Note

This Operator is not made by Hewlett Packard but is a community project.

Why an Operator

HP offers an Ansible Playbook to help users deploy HP's Squest on their K8s Clusters. The play by default installs a lot of other tools not necessarily needed to run a Squest instance like Prometheus or CertManager. Also, the whole lifecycle of the Squest instance requires you to manually run Ansible playbooks. With this operator, the upgrade is as simple as changing the image tag in the custom resource Squest, triggering a redeployment and migration.

Getting started

Installation

1. To install the Operator, clone the repo and change directory into the cloned directory.
2. Run make deploy to install all required resources and the operator into your cluster. This requires you to have an existing K8s context on your system.
3. Create new namespace for your Squest deployment to live in: kubectl create ns squest

Note

You can also choose a different name for the namespace. Just make sure to also configure the Custom Resource's namespace accordingly

4. Create a manifest with the Custom Resource so that it fits your needs. The parameters are described here. The operator will be installed in the namespace squest-operator-system
5. Apply your manifest kubectl apply -f <path-to-your-manifest.yaml>

The operator will now attempt to create all needed resources in the squest namespace (unless configured differently in the manifest). Finally, you can login to your squest instance via the configured Ingress hostname.

When deploying the manifest to your Cluster it will deploy:

• A Database
• A Redis Cache
• A RabbitMQ Server (Not clustered like the official K8s deployment)
• A migration job to configure your Squest instance before the initial deployment and for upgrades
• The actual Squest deployment.
• A maintenance deployment (During initial deployment and during upgrades, the ingress - should you have enabled it in the custom resource - will switch over to this deployment's service to display the maintenance page.)
• The celery beat and celery worker deployments
• Persistent Volume Claims, if desired

Upgrade

To upgrade to a newer version of Squest, simply change the image value in the Custom Resource called django_image to a new version and apply the manifest. The operator will automatically create a new migration job and redeploy Squest with the new version.

Uninstall

To remove the operator and the Squest resource and all the created dependencies run make undeploy.

Custom Resource

This custom resource defines your Squest environment and configures all the resources. Copy this sample and configure it to your liking.

apiVersion: web.hewlettpackard.com/v1alpha1
kind: Squest
metadata:
  labels:
    app.kubernetes.io/name: squest
    app.kubernetes.io/instance: squest
    app.kubernetes.io/part-of: squest-operator
    app.kubernetes.io/managed-by: kustomize
    app.kubernetes.io/created-by: squest-operator
  name: squest
  namespace: squest
spec:
  node_selector: ''
  redis_node_selector: ''
  rabbitmq_node_selector: ''
  db_node_selector: ''
  django_node_selector: |
    kubernetes.io/arch: amd64
  maintenance_node_selector: ''

  storage_class: default

  db_persistence_enabled: true
  db_persistence_storage: 2Gi
  db_user: squest
  db_password: toor
  db_root_password: toor

  rabbitmq_persistence_enabled: true
  rabbitmq_persistence_storage: 2Gi

  django_persistence_enabled: true
  django_persistence_storage: 1Gi
  django_ingress_enabled: false
  django_ingress_class: traefik
  django_ingress_hostname: squest.0x01.host
  django_image: quay.io/hewlettpackardenterprise/squest:2.5.1

  redis_password: toor

Configuration

This section describes what each parameter of the Custom Resource spec does and how to configure them.

global parameters

Parameter	Type	Description	Default
storage_class	string	Storage Class for all Persistent Volumes	default
node_selector	string	Multiline string of nodeSelectors to apply to all resources	''

squest

Parameter	Type	Description	Default
django_persistence_enabled	boolean	Enable persistence for the Squest data	true
django_persistence_storage	string	The volume size e.g. 2Gi	1Gi
django_node_selector	string	Multiline string to select nodes for the deployment	''
django_image	string	The image to use for the squest deployment	quay.io/hewlettpackardenterprise/squest:latest
django_ingress_enabled	boolean	Deploy an Ingress	false
django_ingress_hostname	string	The hostname of your Squest web ui	squest.company
django_ingress_class	string	The ingress class to use for the ingress	traefik
django_resource_requirements	map	Define the resource limits and requests according to the official Kubernetes deployment spec	limits:    memory: 512Mi

db

Parameter	Type	Description	Default
db_persistence_enabled	boolean	Enable persistence for the database	true
db_persistence_storage	string	The volume size e.g. 2Gi	2Gi
db_node_selector	string	Multiline string to select nodes for the deployment	''
db_root_password	string	The database root password	toor
db_password	string	The user password	toor
db_user	string	The database user	squest
db_resource_requirements	map	Define the resource limits and requests according to the official Kubernetes deployment spec	limits:    memory: 512Mi

rabbitmq

Parameter	Type	Description	Default
rabbitmq_persistence_enabled	boolean	Enable persistence for rabbitmq	true
rabbitmq_persistence_storage	string	The volume size e.g. 2Gi	2Gi
rabbitmq_node_selector	string	Multiline string to select nodes for the deployment	''
rabbitmq_resource_requirements	map	Define the resource limits and requests according to the official Kubernetes deployment spec	limits:     memory: 256Mi

redis

Parameter	Type	Description	Default
redis_password	string	The redis cache password	toor
redis_node_selector	string	Multiline string to select nodes for the deployment	''
redis_resource_requirements	map	Define the resource limits and requests according to the official Kubernetes deployment spec	limits:     memory: 50Mi

maintenance

Parameter	Type	Description	Default
maintenance_node_selector	string	Multiline string to select nodes for the deployment	''
maintenance_resource_requirements	map	Define the resource limits and requests according to the official Kubernetes deployment spec	limits:     memory: 16Mi