We assume that
- you have installed the SCONE Operator
- you have an create a Kubernetes secret
sconefor pulling the container images from repo
registry.scontain.com/scone.cloud` - you have created the image pull secret
sconeapps
forregistry.scontain.com/scone.cloud
in the namespace in which you deploy the instances.
You can install SGX Plugin, LAS, and a CAS instance with one command:
kubectl create -f https://raw.githubusercontent.com/scontain/operator-samples/main/base_and_cas.yaml
This deploys a production CAS and all necessary SCONE base services. The CAS database is persistent.
!!! note "No Fail-Over to other nodes" This CAS instance does support fail-over to another Kubernetes node. Such a fail-over requires the backup-controller to be enabled.
You can install SGX Plugin with kubectl
:
kubectl create -f https://raw.githubusercontent.com/scontain/operator-samples/main/base_v1beta1_sgxplugin.yaml
LAS from the included sample manifests:
kubectl create -f https://raw.githubusercontent.com/scontain/operator-samples/main/base_v1beta1_las.yaml
These two manifests have reasonable defaults and should work without any modifications for most deployments. The default Kubernetes namespace for the SGX Plugin and LAS is the namespace of the SCONE Operator. The default Kubernetes namespace of the SCONE Operator is scone
.
In case the defaults do not fit your requirements, you can customize the manifests as described in the SCONE Custom Resource Definitions.
- Install SCONE CAS to your cluster in production mode as follows:
kubectl -f create https://raw.githubusercontent.com/scontain/operator-samples/main/services_v1beta1_cas.yaml
This CAS will use a persistent volume for its database but it does not support a fail-over to a different host.
For more details, see https://sconedocs.github.io/3_resource_definitions/