Understand the Kubernetes Cluster Access provided to Run:ai¶
Run:ai has the ability to work under restrictive Kubernetes environments. Namely:
- Kubernetes PodSecurityPolicy
You can enable these restricted environment by setting the
openshift configuration flags in the Helm values file before installing the Run:ai cluster.
Other configuration flags are controlling specific behavioral aspects of Run:ai. Specifically, those which require additional permissions. Such as automatic namespace/project creation, secret propagation, and more.
The purpose of this document is to provide security officers with the ability to review what cluster-wide access Run:ai requires, and verify that it is in line with organizational policy, before installing the Run:ai cluster.
Review Cluster Access Roles¶
helm repo add runai https://run-ai-charts.storage.googleapis.com helm repo update helm install runai-cluster runai/runai-cluster -n runai -f runai-<cluster-name>.yaml \ --dry-run > cluster-all.yaml
cluster-all.yaml can be then be reviewed. You can use the internal filenames (provided in comments within the file) to gain more understanding according to the table below:
| || ||Mandatory Kubernetes Cluster Roles and Cluster Role Bindings|
| || ||Automatic Project Creation and Maintenance. Provides Run:ai with the ability to create Kubernetes namespaces when the Run:ai administrator creates new Projects. Can be turned on/off via flag|
| || ||Automatically assign Users to Projects. Can be turned on/off via flag|
| || ||Allow the propagation of Secrets. See Secrets in Jobs. Can be turned on/off via flag|
| || ||Disables the usage of the Kubernetes Limit Range feature. Can be turned on/off via flag|
| || ||OpenShift-specific Security Contexts|
| ||4 files||Folder contains a list of Priority Classes used by Run:ai|
| || ||A subset of the Kubernetes baseline PodSecurityPolicy (here)|
| || ||Required for NVIDIA components|
| || ||Required for Run:ai GPU Fractions technology. Can be turned on/off via flag.|
| || ||Required for User Workloads. Extends the Kubernetes baseline PodSecurityPolicy for Run:ai GPU Fractions technology. Can be turned on/off via flag.|
| || ||priviliged Kubernetes PodSecurityPolicy. Used only for installing Node Feature Discovery and NVIDIA GPU Feature Discovery third parties. Can be turned on/off via flag|