Kubernetes Deployment

Helm chart

There’s a Helm chart inside the helm/ folder, in each project repository.

Scoold
Scoold Pro

The Helm chart bootstraps a Scoold instance on a Kubernetes cluster using the Helm package manager.

This chart does not install Para next to Scoold in the K8s cluster. There is a separate Helm chart for Para. We also offer a fully managed Para service at ParaIO.com

Prerequisites

Para backend service (latest version recommended; Helm chart)
Helm 3.0+
Kubernetes 1.21+ (for the optional CronJob helper)

Installation

Clone the repo and go inside the ./helm/ directory :

git clone https://github.com/Erudika/scoold
cd scoold/helm
# edit values.yaml as needed
helm install scoold ./scoold

The command deploys Scoold on the Kubernetes cluster in the default configuration. The configuration section lists the parameters that can be configured during installation.

If you want to auto-initialize Scoold from scratch, deploy the Para Helm chart first, then take note of the root app secret key. The root app secret key is stored in the para-application.conf file located inside /para/application.conf inside the Para pod. Then modify values.yaml and add this Scoold configuration property:

scoold.autoinit.root_app_secret_key="{secret key for root Para app}"

Finally, run helm upgrade --install scoold ./scoold.

Uninstalling the Chart

To uninstall/delete the scoold deployment:

helm uninstall scoold

The command removes all the Kubernetes components associated with the chart and deletes the release.

Configuration properties

The following table lists the configurable parameters inside values.yaml and their default values.

Parameter	Default value
`image.repository` Scoold image name	`erudikaltd/scoold`
`image.tag` Scoold image tag	`1.65.0`
`image.pullPolicy` Image pull policy	`IfNotPresent`
`image.pullSecrets` References to image pull secrets	`[]`
`service.type` Kubernetes Service type	`ClusterIP`
`service.port` Service HTTP port	`8000`
`service.name` Service port name	`http`
`applicationConf` Scoold configuration	Sample block in `values.yaml`
`javaOpts` `JAVA_OPTS` JVM arguments	`-Xmx512m -Xms512m -Dconfig.file=/scoold/config/application.conf`
`podAnnotations` Pod annotations	`{}`
`extraEnvs` Extra environment variables	`[]`
`updateStrategy` Deployment update strategy	`RollingUpdate`
`ingress.enabled` Create Ingress	`false`
`ingress.className` Ingress class name	`""`
`ingress.hosts[0].host` Hostname for the Ingress	`scoold.local`
`ingress.hosts[0].paths[0].path` HTTP path served by the Ingress	`/`
`ingress.tls` TLS configuration	`[]`
`resources` CPU/Memory resource requests/limits	`{}`
`nodeSelector` Node selector	`{}`
`tolerations` Tolerations	`[]`
`affinity` Affinity rules	`{}`

A YAML file that specifies the values for the above parameters can be provided while installing the chart. For example,

helm install scoold ./scoold -f values.yaml

Or, when updating the chart with new configuration:

helm upgrade --install scoold ./scoold -f values.yaml

The Helm chart bootstraps a Scoold Pro deployment on a Kubernetes cluster using the Helm package manager.

This chart does not install Para next to Scoold in the K8s cluster. There is a separate Helm chart for Para. We also offer a fully managed Para service at ParaIO.com

Prerequisites

Para backend service (latest version recommended; Helm chart)
Access credentials for the private Scoold Pro Docker registry on ECR
Helm 3.0+
Kubernetes 1.21+ (for the optional CronJob helper)

Installation

First of all, make sure you have followed the guide to get access to the private Docker registry on ECR. Access to the private Docker registry can be requested by emailing support, after you purchase a Scoold Pro license.

Then clone the Helm chart locally:

git clone https://github.com/Erudika/scoold-pro
cd scoold-pro/helm

Inside the ./helm/ directory execute the following console command, while replacing the values {ecr_key_id} and {ecr_secret} with your ECR credentials:

helm install scooldpro ./scoold-pro \
    --set ecrCredentials.accessKeyId={ecr_key_id},ecrCredentials.accessSecretKey={ecr_secret}

The command deploys Scoold on the Kubernetes cluster in the default configuration. The configuration section lists the parameters that can be configured during installation.

scoold.autoinit.root_app_secret_key="{secret key for root Para app}"

Finally, run helm upgrade --install scooldpro ./scoold-pro.

Uninstalling the Chart

To uninstall/delete the scoold deployment:

helm uninstall scooldpro

The command removes all the Kubernetes components associated with the chart and deletes the release.

Configuration properties

The following table lists the configurable parameters inside [values.yaml](values.yaml) and their default values.

Parameter	Default value
`image.repository` Scoold image name	`374874639893.dkr.ecr.eu-west-1.amazonaws.com/scoold-pro`
`image.tag` Scoold image tag	`1.65.0`
`image.pullPolicy` Image pull policy	`IfNotPresent`
`image.pullSecrets` References to extra image pull secrets	`[]`
`ecrCredentials.registry` Registry host used inside the generated Secret	`374874639893.dkr.ecr.eu-west-1.amazonaws.com`
`ecrCredentials.region` AWS region passed to the helper job	`eu-west-1`
`ecrCredentials.accessKeyId` AWS access key ID stored in a Secret	`""`
`ecrCredentials.accessSecretKey` AWS secret access key stored in a Secret	`""`
`ecrCredentials.secretName` Override for the generated docker-registry Secret name	`""`
`ecrCredentials.refreshSchedule` Cron expression for refreshing the pull secret	`"0 /6 * *"`
`ecrCredentials.helper.image` Container image used for the helper Job/CronJob	`public.ecr.aws/aws-cli/aws-cli:2.15.30`
`ecrCredentials.helper.pullPolicy` Image pull policy for the helper Job/CronJob	`IfNotPresent`
`ecrCredentials.helper.kubectlVersion` Version of kubectl downloaded inside the helper Pods	`v1.29.2`
`ecrCredentials.helper.pullPolicy` Image pull policy for the helper Job/CronJob	`IfNotPresent`
`service.type` Kubernetes Service type	`ClusterIP`
`service.port` Service HTTP port	`8000`
`service.name` Service port name	`http`
`applicationConf` Scoold configuration	Sample block in `values.yaml`
`persistentVolumes.uploads.size` Requested capacity for `/scoold/uploads` PVC	`2Gi`
`persistentVolumes.uploads.accessModes` Access modes for `/scoold/uploads` PVC	`[ReadWriteOnce]`
`persistentVolumes.uploads.storageClassName` StorageClass for `/scoold/uploads` PVC	`""`
`javaOpts` `JAVA_OPTS` JVM arguments	`-Xmx512m -Xms512m -Dconfig.file=/scoold/config/application.conf`
`podAnnotations` Pod annotations	`{}`
`extraEnvs` Extra environment variables	`[]`
`updateStrategy` Deployment update strategy	`RollingUpdate`
`ingress.enabled` Create Ingress	`false`
`ingress.className` Ingress class name	`""`
`ingress.hosts[0].host` Hostname for the Ingress	`scooldpro.local`
`ingress.hosts[0].paths[0].path` HTTP path served by the Ingress	`/`
`ingress.tls` TLS configuration	`[]`
`resources` CPU/Memory resource requests/limits	`{}`
`nodeSelector` Node selector	`{}`
`tolerations` Tolerations	`[]`
`affinity` Affinity rules	`{}`

When working with the private AWS ECR registry, provide an AWS key pair (ecrCredentials.accessKeyId / ecrCredentials.accessSecretKey). The chart stores the credentials in a Secret and deploys a helper Job plus CronJob that run aws ecr get-login-password inside the cluster. The generated token is continuously applied to the docker-registry Secret referenced by the Deployment, so Kubernetes keeps pulling the private Scoold Pro image without any manual steps. Adjust ecrCredentials.refreshSchedule (default every 6 hours) or the helper image/kubectl version if you need different behavior.

A YAML file that specifies the values for the above parameters can be provided while installing the chart. For example,

$ helm install scooldpro ./scoold-pro -f values.yaml

Or, when updating the chart with new configuration:

$ helm upgrade --install scooldpro ./scoold-pro -f values.yaml

Configuration Reference For the complete list of configuration options, visit the configuration reference page.