The module generates role-based access model objects based on the standard Kubernetes RBAC mechanism. The module creates a set of cluster roles (ClusterRole) suitable for most user and group access management tasks.

Starting from Deckhouse Kubernetes Platform v1.64, the module features a new role-based access model. The old role-based access model will continue to operate but support for it will be discontinued in the future.

The new role-based access model is incompatible with the old one.

The documentation of the module implies that you use the new role-based model, unless specified otherwise.

The module implements a role-based access model based on the standard RBAC Kubernetes mechanism. It creates a set of cluster roles (ClusterRole) suitable for most user and group access management tasks.

The new role-based model

Unlike the obsolete DKP role-based model, the new role-based one does not use ClusterAuthorizationRule and AuthorizationRule resources. All access rights are configured in the standard Kubernetes RBAC way, i.e., by creating RoleBinding or ClusterRoleBinding resources and specifying one of the roles prepared by the user-authz module in them.

The module creates special aggregated cluster roles (ClusterRole). By using these roles in RoleBinding or ClusterRoleBinding, you can do the following:

  • Manage access to modules of a specific scope.

    For example, you can use the d8:manage:networking:manager role in ClusterRoleBinding to allow a network administrator to configure network modules (such as cni-cilium, ingress-nginx, istio, etc.).

  • Manage access to user resources of modules within the namespace.

    For example, the d8:use:role:manager role in RoleBinding enables deleting/creating/editing the PodLoggingConfig resource in the namespace. At the same time, it does not grant access to the cluster-wide ClusterLoggingConfig and ClusterLogDestination resources of the log-shipper module, nor does it allow configuration of the log-shipper module itself.

The roles created by the module are divided into two classes:

  • Use roles — for assigning rights to users (such as application developers) in a specific namespace.
  • Manage roles — for assigning rights to administrators.

Use roles

The use role can only be used in the RoleBinding resource.

Use roles are intended to assign rights to a user in a specific namespace. Users refer to, for example, developers who use a cluster configured by an administrator to deploy their applications. Such users don’t need to manage DKP modules or a cluster, but they need to be able to, for example, create their Ingress resources, configure application authentication, and collect logs from applications.

The use role defines permissions for accessing namespaced resources of modules and standard namespaced resources of Kubernetes (Pod, Deployment, Secret, ConfigMap, etc.).

The module creates the following use roles:

  • d8:use:role:viewer — allows viewing standard Kubernetes resources in a specific namespace, except for Secrets and RBAC resources, as well as authenticating in the cluster;
  • d8:use:role:user — in addition to the role d8:use:role:viewer it allows viewing secrets and RBAC resources in a specific namespace, connecting to pods, deleting pods (but not creating or modifying them), executing kubectl port-forward and kubectl proxy, as well as changing the number of replicas of controllers;
  • d8:use:role:manager — in addition to the role d8:use:role:user it allows managing module resources (for example, Certificate, PodLoggingConfig, etc.) and standard namespaced Kubernetes resources (Pod, ConfigMap, CronJob, etc.) in a specific namespace;
  • d8:use:role:admin — in addition to the role d8:use:role:manager it allows managing the resources ResourceQuota, ServiceAccount, Role, RoleBinding, NetworkPolicy in a specific namespace.

Manage roles

The manage role does not grant access to the namespace of user applications.

The manage role grants access only to system namespaces (starting with d8- or kube-), and only to those system namespaces where the modules of the corresponding role scope are running.

Manage roles are intended for assigning rights to manage the entire platform or a part of it (the scope), but not the users applications themselves. The manage role, for example, can allow a security administrator to manage security modules (responsible for the security functions of the cluster). Thus, the security administrator will be able to configure authentication, authorization, security policies, etc., but will not be able to manage other cluster functions (such as network and monitoring settings) or change settings in the namespaces of users applications.

The manage role defines access rights:

  • to cluster-wide Kubernetes resources;
  • to manage DKP modules (moduleConfig resource) within the scope of the role, or to all DKP modules for the role d8:manage:all:*;
  • to manage cluster-wide resources of DKP modules within the scope of the role, or to all resources of DKP modules for the role d8:manage:all:*;
  • to system namespaces (starting with d8- or kube-) in which the modules of the scope of the role operate, or to all system namespaces for the role d8:manage:all:*.

The manage role name format is d8:manage:<SCOPE>:<ACCESS_LEVEL>, where:

  • SCOPE is the role’s scope. It can be one of the scope, or all, for access across all scopes;
  • ACCESS_LEVEL is the access level.

    Examples of manage roles:

    • d8:manage:all:viewer — access to view the configuration of all DKP modules (moduleConfig resource), their cluster-wide resources, their namespaced resources, and standard Kubernetes objects (except Secrets and RBAC resources) in all system namespaces (starting with d8- or kube-);
    • d8:manage:all:manager — similar to the role d8:manage:all:viewer, but with admin-level access, i.e., view/create/modify/delete the configuration of all DKP modules (moduleConfig resource), their cluster-wide resources, their namespaced resources, and standard Kubernetes objects in all system namespaces (starting with d8- or kube-);
    • d8:manage:observability:viewer — access to view the configuration of DKP modules (moduleConfig resource) from the observability area, their cluster-wide resources, their namespaced resources, and standard Kubernetes objects (except secrets and RBAC resources) in the system namespaces d8-log-shipper, d8-monitoring, d8-okmeter, d8-operator-prometheus, d8-upmeter, kube-prometheus-pushgateway.

The module provides two access level for administrators:

  • viewer — allows viewing standard Kubernetes resources, the configuration of modules (resources moduleConfig), cluster-wide resources of modules, and namespaced resources of modules in the module namespace;
  • manager — in addition to the role viewer it allows managing standard Kubernetes resources, the configuration of modules (resources moduleConfig), cluster-wide resources of modules, and namespaced resources of modules in the module namespace;

Scopes of the role-based model

Each DKP module belongs to a specific scope. For each scope, there is a set of roles with different levels of access. Roles are updated automatically when the module is enabled or disabled.

For example, for the networking scope, there are the following manage roles that can be used in ClusterRoleBinding:

  • d8:manage:networking:viewer
  • d8:manage:networking:manager

The scope of the role restricts its action to all system namespaces of the cluster (scope all) or to those namespaces in which the area modules operate (see the table of area compositions).

Role-based model scopes composition table.

Scope The modules within the scope Namespaces in which modules of a scope operate
all All modules All namespaces
deckhouse
  • deckhouse
  • flant-integration
  • multitenancy-manager
  • d8-flant-integration
  • d8-multitenancy-manager
  • d8-system
infrastructure
  • ceph-csi
  • chrony
  • cloud-provider-aws
  • cloud-provider-azure
  • cloud-provider-gcp
  • cloud-provider-openstack
  • cloud-provider-vcd
  • cloud-provider-vsphere
  • cloud-provider-yandex
  • cloud-provider-zvirt
  • descheduler
  • istio
  • keepalived
  • local-path-provisioner
  • loki
  • node-manager
  • pod-reloader
  • snapshot-controller
  • terraform-manager
  • vertical-pod-autoscaler
  • d8-ceph-csi
  • d8-chrony
  • d8-cloud-instance-manager
  • d8-cloud-provider-aws
  • d8-cloud-provider-azure
  • d8-cloud-provider-gcp
  • d8-cloud-provider-openstack
  • d8-cloud-provider-vcd
  • d8-cloud-provider-vsphere
  • d8-cloud-provider-yandex
  • d8-cloud-provider-zvirt
  • d8-descheduler
  • d8-istio
  • d8-keepalived
  • d8-local-path-provisioner
  • d8-monitoring
  • d8-pod-reloader
  • d8-snapshot-controller
  • d8-system
  • kube-system
kubernetes
  • control-plane-manager
  • flow-schema
  • kube-dns
  • kube-proxy
  • priority-class
  • kube-system
networking
  • cilium-hubble
  • cni-cilium
  • cni-flannel
  • cni-simple-bridge
  • flow-schema
  • ingress-nginx
  • istio
  • kube-dns
  • kube-proxy
  • metallb
  • network-gateway
  • network-policy-engine
  • node-local-dns
  • openvpn
  • static-routing-manager
  • d8-cni-cilium
  • d8-cni-flannel
  • d8-cni-simple-bridge
  • d8-ingress-nginx
  • d8-istio
  • d8-metallb
  • d8-network-gateway
  • d8-openvpn
  • d8-static-routing-manager
  • d8-system
  • kube-system
observability
  • dashboard
  • extended-monitoring
  • log-shipper
  • monitoring-applications
  • monitoring-custom
  • monitoring-deckhouse
  • monitoring-kubernetes
  • monitoring-kubernetes-control-plane
  • monitoring-ping
  • okmeter
  • operator-prometheus
  • prometheus
  • prometheus-metrics-adapter
  • prometheus-pushgateway
  • upmeter
  • d8-dashboard
  • d8-log-shipper
  • d8-monitoring
  • d8-okmeter
  • d8-operator-prometheus
  • d8-upmeter
  • kube-prometheus-pushgateway
security
  • admission-policy-engine
  • cert-manager
  • operator-trivy
  • runtime-audit-engine
  • secret-copier
  • user-authn
  • user-authz
  • d8-admission-policy-engine
  • d8-cert-manager
  • d8-operator-trivy
  • d8-runtime-audit-engine
  • d8-user-authn
storage
  • ceph-csi
  • snapshot-controller
  • d8-ceph-csi
  • d8-snapshot-controller

The obsolete role-based model

Features:

In addition to the RBAC, you can use a set of high-level roles in the module:

  • User — has access to information about all objects (including viewing pod logs) but cannot exec into containers, read secrets, and perform port-forwarding;
  • PrivilegedUser — the same as User + can exec into containers, read secrets, and delete pods (and thus, restart them);
  • Editor — is the same as PrivilegedUser + can create and edit all objects that are usually required for application tasks.
  • Admin — the same as Editor + can delete service objects (auxiliary resources such as ReplicaSet, certmanager.k8s.io/challenges and certmanager.k8s.io/orders);
  • ClusterEditor — the same as Editor + can manage a limited set of cluster-wide objects that can be used in application tasks (ClusterXXXMetric, KeepalivedInstance, DaemonSet, etc.). This role is best suited for cluster operators.
  • ClusterAdmin — the same as both ClusterEditor and Admin + can manage cluster-wide service objects (e.g., MachineSets, Machines, OpenstackInstanceClasses…, as well as ClusterAuthorizationRule, ClusterRoleBindings and ClusterRole). This role is best suited for cluster administrators. Note that since ClusterAdmin can edit ClusterRoleBindings, he can broaden his privileges within the cluster;
  • SuperAdmin — can perform any actions with any objects (note that namespaceSelector and limitNamespaces restrictions remain valid).

Currently, the multi-tenancy mode (namespace-based authorization) is implemented according to a temporary scheme and isn’t guaranteed to be entirely safe and secure!

If a ClusterAuthorizationRule Custom Resource contains the namespaceSelector field, neither limitNamespaces nor allowAccessToSystemNamespacesare taken into consideration.

The allowAccessToSystemNamespaces, namespaceSelector and limitNamespaces options in the custom resource will no longer be applied if the authorization system’s webhook is unavailable for some reason. As a result, users will have access to all namespaces. After the webhook availability is restored, the options will become relevant again.

Default access list for each role

verbs aliases:

  • read - get, list, watch
  • read-write - get, list, watch, create, delete, deletecollection, patch, update
  • write - create, delete, deletecollection, patch, update

Role User:

read:
    - apiextensions.k8s.io/customresourcedefinitions
    - apps/daemonsets
    - apps/deployments
    - apps/replicasets
    - apps/statefulsets
    - autoscaling.k8s.io/verticalpodautoscalers
    - autoscaling/horizontalpodautoscalers
    - batch/cronjobs
    - batch/jobs
    - configmaps
    - discovery.k8s.io/endpointslices
    - endpoints
    - events
    - events.k8s.io/events
    - extensions/daemonsets
    - extensions/deployments
    - extensions/ingresses
    - extensions/replicasets
    - extensions/replicationcontrollers
    - limitranges
    - metrics.k8s.io/nodes
    - metrics.k8s.io/pods
    - namespaces
    - networking.k8s.io/ingresses
    - networking.k8s.io/networkpolicies
    - nodes
    - persistentvolumeclaims
    - persistentvolumes
    - pods
    - pods/log
    - policy/poddisruptionbudgets
    - rbac.authorization.k8s.io/rolebindings
    - rbac.authorization.k8s.io/roles
    - replicationcontrollers
    - resourcequotas
    - serviceaccounts
    - services
    - storage.k8s.io/storageclasses

Role PrivilegedUser (includes all rules from the role User):

create:
    - pods/eviction
create,get:
    - pods/attach
    - pods/exec
delete,deletecollection:
    - pods
read:
    - secrets

Role Editor (includes all rules from the role User, PrivilegedUser):

read-write:
    - apps/deployments
    - apps/statefulsets
    - autoscaling.k8s.io/verticalpodautoscalers
    - autoscaling/horizontalpodautoscalers
    - batch/cronjobs
    - batch/jobs
    - configmaps
    - discovery.k8s.io/endpointslices
    - endpoints
    - extensions/deployments
    - extensions/ingresses
    - networking.k8s.io/ingresses
    - persistentvolumeclaims
    - policy/poddisruptionbudgets
    - serviceaccounts
    - services
write:
    - secrets

Role Admin (includes all rules from the role User, PrivilegedUser, Editor):

create,patch,update:
    - pods
delete,deletecollection:
    - apps/replicasets
    - extensions/replicasets

Role ClusterEditor (includes all rules from the role User, PrivilegedUser, Editor):

read:
    - rbac.authorization.k8s.io/clusterrolebindings
    - rbac.authorization.k8s.io/clusterroles
write:
    - apiextensions.k8s.io/customresourcedefinitions
    - apps/daemonsets
    - extensions/daemonsets
    - storage.k8s.io/storageclasses

Role ClusterAdmin (includes all rules from the role User, PrivilegedUser, Editor, Admin, ClusterEditor):

read-write:
    - deckhouse.io/clusterauthorizationrules
write:
    - limitranges
    - namespaces
    - networking.k8s.io/networkpolicies
    - rbac.authorization.k8s.io/clusterrolebindings
    - rbac.authorization.k8s.io/clusterroles
    - rbac.authorization.k8s.io/rolebindings
    - rbac.authorization.k8s.io/roles
    - resourcequotas

You can get additional list of access rules for module role from cluster (existing user defined rules and non-default rules from other deckhouse modules):

D8_ROLE_NAME=Editor
kubectl get clusterrole -A -o jsonpath="{range .items[?(@.metadata.annotations.user-authz\.deckhouse\.io/access-level=='$D8_ROLE_NAME')]}{.rules}{'\n'}{end}" | jq -s add