/* Copyright 2016 The Kubernetes Authors. Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. */ package bootstrappolicy import ( "k8s.io/apimachinery/pkg/api/meta" metav1 "k8s.io/apimachinery/pkg/apis/meta/v1" "k8s.io/apimachinery/pkg/runtime" "k8s.io/apiserver/pkg/authentication/user" rbac "k8s.io/kubernetes/pkg/apis/rbac" ) var ( ReadWrite = []string{"get", "list", "watch", "create", "update", "patch", "delete", "deletecollection"} Read = []string{"get", "list", "watch"} Label = map[string]string{"kubernetes.io/bootstrapping": "rbac-defaults"} Annotation = map[string]string{rbac.AutoUpdateAnnotationKey: "true"} ) const ( legacyGroup = "" appsGroup = "apps" authenticationGroup = "authentication.k8s.io" authorizationGroup = "authorization.k8s.io" autoscalingGroup = "autoscaling" batchGroup = "batch" certificatesGroup = "certificates.k8s.io" extensionsGroup = "extensions" policyGroup = "policy" rbacGroup = "rbac.authorization.k8s.io" storageGroup = "storage.k8s.io" ) func addDefaultMetadata(obj runtime.Object) { metadata, err := meta.Accessor(obj) if err != nil { // if this happens, then some static code is broken panic(err) } labels := metadata.GetLabels() if labels == nil { labels = map[string]string{} } for k, v := range Label { labels[k] = v } metadata.SetLabels(labels) annotations := metadata.GetAnnotations() if annotations == nil { annotations = map[string]string{} } for k, v := range Annotation { annotations[k] = v } metadata.SetAnnotations(annotations) } func addClusterRoleLabel(roles []rbac.ClusterRole) { for i := range roles { addDefaultMetadata(&roles[i]) } return } func addClusterRoleBindingLabel(rolebindings []rbac.ClusterRoleBinding) { for i := range rolebindings { addDefaultMetadata(&rolebindings[i]) } return } func NodeRules() []rbac.PolicyRule { return []rbac.PolicyRule{ // Needed to check API access. These creates are non-mutating rbac.NewRule("create").Groups(authenticationGroup).Resources("tokenreviews").RuleOrDie(), rbac.NewRule("create").Groups(authorizationGroup).Resources("subjectaccessreviews", "localsubjectaccessreviews").RuleOrDie(), // Needed to build serviceLister, to populate env vars for services rbac.NewRule(Read...).Groups(legacyGroup).Resources("services").RuleOrDie(), // Nodes can register Node API objects and report status. // Use the NodeRestriction admission plugin to limit a node to creating/updating its own API object. rbac.NewRule("create", "get", "list", "watch").Groups(legacyGroup).Resources("nodes").RuleOrDie(), rbac.NewRule("update", "patch").Groups(legacyGroup).Resources("nodes/status").RuleOrDie(), rbac.NewRule("update", "patch", "delete").Groups(legacyGroup).Resources("nodes").RuleOrDie(), // TODO: restrict to the bound node as creator in the NodeRestrictions admission plugin rbac.NewRule("create", "update", "patch").Groups(legacyGroup).Resources("events").RuleOrDie(), // TODO: restrict to pods scheduled on the bound node once field selectors are supported by list/watch authorization rbac.NewRule(Read...).Groups(legacyGroup).Resources("pods").RuleOrDie(), // Needed for the node to create/delete mirror pods. // Use the NodeRestriction admission plugin to limit a node to creating/deleting mirror pods bound to itself. rbac.NewRule("create", "delete").Groups(legacyGroup).Resources("pods").RuleOrDie(), // Needed for the node to report status of pods it is running. // Use the NodeRestriction admission plugin to limit a node to updating status of pods bound to itself. rbac.NewRule("update").Groups(legacyGroup).Resources("pods/status").RuleOrDie(), // Needed for imagepullsecrets, rbd/ceph and secret volumes, and secrets in envs // Needed for configmap volume and envs // Use the NodeRestriction admission plugin to limit a node to get secrets/configmaps referenced by pods bound to itself. rbac.NewRule("get").Groups(legacyGroup).Resources("secrets", "configmaps").RuleOrDie(), // Needed for persistent volumes // Use the NodeRestriction admission plugin to limit a node to get pv/pvc objects referenced by pods bound to itself. rbac.NewRule("get").Groups(legacyGroup).Resources("persistentvolumeclaims", "persistentvolumes").RuleOrDie(), // TODO: add to the Node authorizer and restrict to endpoints referenced by pods or PVs bound to the node // Needed for glusterfs volumes rbac.NewRule("get").Groups(legacyGroup).Resources("endpoints").RuleOrDie(), // Used to create a certificatesigningrequest for a node-specific client certificate, and watch // for it to be signed. This allows the kubelet to rotate it's own certificate. rbac.NewRule("create", "get", "list", "watch").Groups(certificatesGroup).Resources("certificatesigningrequests").RuleOrDie(), } } // ClusterRoles returns the cluster roles to bootstrap an API server with func ClusterRoles() []rbac.ClusterRole { roles := []rbac.ClusterRole{ { // a "root" role which can do absolutely anything ObjectMeta: metav1.ObjectMeta{Name: "cluster-admin"}, Rules: []rbac.PolicyRule{ rbac.NewRule("*").Groups("*").Resources("*").RuleOrDie(), rbac.NewRule("*").URLs("*").RuleOrDie(), }, }, { // a role which provides just enough power to determine if the server is ready and discover API versions for negotiation ObjectMeta: metav1.ObjectMeta{Name: "system:discovery"}, Rules: []rbac.PolicyRule{ rbac.NewRule("get").URLs("/healthz", "/version", "/swaggerapi", "/swaggerapi/*", "/api", "/api/*", "/apis", "/apis/*").RuleOrDie(), }, }, { // a role which provides minimal resource access to allow a "normal" user to learn information about themselves ObjectMeta: metav1.ObjectMeta{Name: "system:basic-user"}, Rules: []rbac.PolicyRule{ // TODO add future selfsubjectrulesreview, project request APIs, project listing APIs rbac.NewRule("create").Groups(authorizationGroup).Resources("selfsubjectaccessreviews").RuleOrDie(), }, }, { // a role for a namespace level admin. It is `edit` plus the power to grant permissions to other users. ObjectMeta: metav1.ObjectMeta{Name: "admin"}, Rules: []rbac.PolicyRule{ rbac.NewRule(ReadWrite...).Groups(legacyGroup).Resources("pods", "pods/attach", "pods/proxy", "pods/exec", "pods/portforward").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(legacyGroup).Resources("replicationcontrollers", "replicationcontrollers/scale", "serviceaccounts", "services", "services/proxy", "endpoints", "persistentvolumeclaims", "configmaps", "secrets").RuleOrDie(), rbac.NewRule(Read...).Groups(legacyGroup).Resources("limitranges", "resourcequotas", "bindings", "events", "pods/status", "resourcequotas/status", "namespaces/status", "replicationcontrollers/status", "pods/log").RuleOrDie(), // read access to namespaces at the namespace scope means you can read *this* namespace. This can be used as an // indicator of which namespaces you have access to. rbac.NewRule(Read...).Groups(legacyGroup).Resources("namespaces").RuleOrDie(), rbac.NewRule("impersonate").Groups(legacyGroup).Resources("serviceaccounts").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(appsGroup).Resources("statefulsets", "deployments", "deployments/scale", "deployments/rollback").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(autoscalingGroup).Resources("horizontalpodautoscalers").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(batchGroup).Resources("jobs", "cronjobs", "scheduledjobs").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(extensionsGroup).Resources("daemonsets", "deployments", "deployments/scale", "deployments/rollback", "ingresses", "replicasets", "replicasets/scale", "replicationcontrollers/scale").RuleOrDie(), // additional admin powers rbac.NewRule("create").Groups(authorizationGroup).Resources("localsubjectaccessreviews").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(rbacGroup).Resources("roles", "rolebindings").RuleOrDie(), }, }, { // a role for a namespace level editor. It grants access to all user level actions in a namespace. // It does not grant powers for "privileged" resources which are domain of the system: `/status` // subresources or `quota`/`limits` which are used to control namespaces ObjectMeta: metav1.ObjectMeta{Name: "edit"}, Rules: []rbac.PolicyRule{ rbac.NewRule(ReadWrite...).Groups(legacyGroup).Resources("pods", "pods/attach", "pods/proxy", "pods/exec", "pods/portforward").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(legacyGroup).Resources("replicationcontrollers", "replicationcontrollers/scale", "serviceaccounts", "services", "services/proxy", "endpoints", "persistentvolumeclaims", "configmaps", "secrets").RuleOrDie(), rbac.NewRule(Read...).Groups(legacyGroup).Resources("limitranges", "resourcequotas", "bindings", "events", "pods/status", "resourcequotas/status", "namespaces/status", "replicationcontrollers/status", "pods/log").RuleOrDie(), // read access to namespaces at the namespace scope means you can read *this* namespace. This can be used as an // indicator of which namespaces you have access to. rbac.NewRule(Read...).Groups(legacyGroup).Resources("namespaces").RuleOrDie(), rbac.NewRule("impersonate").Groups(legacyGroup).Resources("serviceaccounts").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(appsGroup).Resources("statefulsets", "deployments", "deployments/scale", "deployments/rollback").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(autoscalingGroup).Resources("horizontalpodautoscalers").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(batchGroup).Resources("jobs", "cronjobs", "scheduledjobs").RuleOrDie(), rbac.NewRule(ReadWrite...).Groups(extensionsGroup).Resources("daemonsets", "deployments", "deployments/scale", "deployments/rollback", "ingresses", "replicasets", "replicasets/scale", "replicationcontrollers/scale").RuleOrDie(), }, }, { // a role for namespace level viewing. It grants Read-only access to non-escalating resources in // a namespace. ObjectMeta: metav1.ObjectMeta{Name: "view"}, Rules: []rbac.PolicyRule{ rbac.NewRule(Read...).Groups(legacyGroup).Resources("pods", "replicationcontrollers", "replicationcontrollers/scale", "serviceaccounts", "services", "endpoints", "persistentvolumeclaims", "configmaps").RuleOrDie(), rbac.NewRule(Read...).Groups(legacyGroup).Resources("limitranges", "resourcequotas", "bindings", "events", "pods/status", "resourcequotas/status", "namespaces/status", "replicationcontrollers/status", "pods/log").RuleOrDie(), // read access to namespaces at the namespace scope means you can read *this* namespace. This can be used as an // indicator of which namespaces you have access to. rbac.NewRule(Read...).Groups(legacyGroup).Resources("namespaces").RuleOrDie(), rbac.NewRule(Read...).Groups(appsGroup).Resources("statefulsets", "deployments", "deployments/scale").RuleOrDie(), rbac.NewRule(Read...).Groups(autoscalingGroup).Resources("horizontalpodautoscalers").RuleOrDie(), rbac.NewRule(Read...).Groups(batchGroup).Resources("jobs", "cronjobs", "scheduledjobs").RuleOrDie(), rbac.NewRule(Read...).Groups(extensionsGroup).Resources("daemonsets", "deployments", "deployments/scale", "ingresses", "replicasets", "replicasets/scale", "replicationcontrollers/scale").RuleOrDie(), }, }, { // a role to use for heapster's connections back to the API server ObjectMeta: metav1.ObjectMeta{Name: "system:heapster"}, Rules: []rbac.PolicyRule{ rbac.NewRule(Read...).Groups(legacyGroup).Resources("events", "pods", "nodes", "namespaces").RuleOrDie(), }, }, { // a role for nodes to use to have the access they need for running pods ObjectMeta: metav1.ObjectMeta{Name: "system:node"}, Rules: NodeRules(), }, { // a role to use for node-problem-detector access. It does not get bound to default location since // deployment locations can reasonably vary. ObjectMeta: metav1.ObjectMeta{Name: "system:node-problem-detector"}, Rules: []rbac.PolicyRule{ rbac.NewRule("get").Groups(legacyGroup).Resources("nodes").RuleOrDie(), rbac.NewRule("patch").Groups(legacyGroup).Resources("nodes/status").RuleOrDie(), eventsRule(), }, }, { // a role to use for setting up a proxy ObjectMeta: metav1.ObjectMeta{Name: "system:node-proxier"}, Rules: []rbac.PolicyRule{ // Used to build serviceLister rbac.NewRule("list", "watch").Groups(legacyGroup).Resources("services", "endpoints").RuleOrDie(), rbac.NewRule("get").Groups(legacyGroup).Resources("nodes").RuleOrDie(), eventsRule(), }, }, { // a role to use for bootstrapping a node's client certificates ObjectMeta: metav1.ObjectMeta{Name: "system:node-bootstrapper"}, Rules: []rbac.PolicyRule{ // used to create a certificatesigningrequest for a node-specific client certificate, and watch for it to be signed rbac.NewRule("create", "get", "list", "watch").Groups(certificatesGroup).Resources("certificatesigningrequests").RuleOrDie(), }, }, { // a role to use for allowing authentication and authorization delegation ObjectMeta: metav1.ObjectMeta{Name: "system:auth-delegator"}, Rules: []rbac.PolicyRule{ // These creates are non-mutating rbac.NewRule("create").Groups(authenticationGroup).Resources("tokenreviews").RuleOrDie(), rbac.NewRule("create").Groups(authorizationGroup).Resources("subjectaccessreviews").RuleOrDie(), }, }, { // a role to use for the API registry, summarization, and proxy handling ObjectMeta: metav1.ObjectMeta{Name: "system:kube-aggregator"}, Rules: []rbac.PolicyRule{ // it needs to see all services so that it knows whether the ones it points to exist or not rbac.NewRule(Read...).Groups(legacyGroup).Resources("services", "endpoints").RuleOrDie(), }, }, { // a role to use for bootstrapping the kube-controller-manager so it can create the shared informers // service accounts, and secrets that we need to create separate identities for other controllers ObjectMeta: metav1.ObjectMeta{Name: "system:kube-controller-manager"}, Rules: []rbac.PolicyRule{ eventsRule(), rbac.NewRule("create").Groups(legacyGroup).Resources("endpoints", "secrets", "serviceaccounts").RuleOrDie(), rbac.NewRule("delete").Groups(legacyGroup).Resources("secrets").RuleOrDie(), rbac.NewRule("get").Groups(legacyGroup).Resources("endpoints", "namespaces", "secrets", "serviceaccounts").RuleOrDie(), rbac.NewRule("update").Groups(legacyGroup).Resources("endpoints", "secrets", "serviceaccounts").RuleOrDie(), // Needed to check API access. These creates are non-mutating rbac.NewRule("create").Groups(authenticationGroup).Resources("tokenreviews").RuleOrDie(), // Needed for all shared informers rbac.NewRule("list", "watch").Groups("*").Resources("*").RuleOrDie(), }, }, { // a role to use for the kube-scheduler ObjectMeta: metav1.ObjectMeta{Name: "system:kube-scheduler"}, Rules: []rbac.PolicyRule{ eventsRule(), // this is for leaderlease access // TODO: scope this to the kube-system namespace rbac.NewRule("create").Groups(legacyGroup).Resources("endpoints").RuleOrDie(), rbac.NewRule("get", "update", "patch", "delete").Groups(legacyGroup).Resources("endpoints").Names("kube-scheduler").RuleOrDie(), // fundamental resources rbac.NewRule(Read...).Groups(legacyGroup).Resources("nodes", "pods").RuleOrDie(), rbac.NewRule("create").Groups(legacyGroup).Resources("pods/binding", "bindings").RuleOrDie(), rbac.NewRule("update").Groups(legacyGroup).Resources("pods/status").RuleOrDie(), // things that select pods rbac.NewRule(Read...).Groups(legacyGroup).Resources("services", "replicationcontrollers").RuleOrDie(), rbac.NewRule(Read...).Groups(extensionsGroup).Resources("replicasets").RuleOrDie(), rbac.NewRule(Read...).Groups(appsGroup).Resources("statefulsets").RuleOrDie(), // things that pods use rbac.NewRule(Read...).Groups(legacyGroup).Resources("persistentvolumeclaims", "persistentvolumes").RuleOrDie(), }, }, { // a role to use for the kube-dns pod ObjectMeta: metav1.ObjectMeta{Name: "system:kube-dns"}, Rules: []rbac.PolicyRule{ rbac.NewRule("list", "watch").Groups(legacyGroup).Resources("endpoints", "services").RuleOrDie(), }, }, { // a role for an external/out-of-tree persistent volume provisioner ObjectMeta: metav1.ObjectMeta{Name: "system:persistent-volume-provisioner"}, Rules: []rbac.PolicyRule{ rbac.NewRule("get", "list", "watch", "create", "delete").Groups(legacyGroup).Resources("persistentvolumes").RuleOrDie(), // update is needed in addition to read access for setting lock annotations on PVCs rbac.NewRule("get", "list", "watch", "update").Groups(legacyGroup).Resources("persistentvolumeclaims").RuleOrDie(), rbac.NewRule(Read...).Groups(storageGroup).Resources("storageclasses").RuleOrDie(), // Needed for watching provisioning success and failure events rbac.NewRule("watch").Groups(legacyGroup).Resources("events").RuleOrDie(), eventsRule(), }, }, } addClusterRoleLabel(roles) return roles } // ClusterRoleBindingFilter can modify and return or omit (by returning nil) a role binding type ClusterRoleBindingFilter func(*rbac.ClusterRoleBinding) *rbac.ClusterRoleBinding // AddClusterRoleBindingFilter adds the given filter to the list that is invoked when determing bootstrap roles to reconcile. func AddClusterRoleBindingFilter(filter ClusterRoleBindingFilter) { clusterRoleBindingFilters = append(clusterRoleBindingFilters, filter) } // ClearClusterRoleBindingFilters removes any filters added using AddClusterRoleBindingFilter func ClearClusterRoleBindingFilters() { clusterRoleBindingFilters = nil } const systemNodeRoleName = "system:node" var clusterRoleBindingFilters []ClusterRoleBindingFilter // OmitNodesGroupBinding is a filter that omits the deprecated binding for the system:nodes group to the system:node role. var OmitNodesGroupBinding = ClusterRoleBindingFilter(func(binding *rbac.ClusterRoleBinding) *rbac.ClusterRoleBinding { if binding.RoleRef.Name == systemNodeRoleName { subjects := []rbac.Subject{} for _, subject := range binding.Subjects { if subject.Kind == rbac.GroupKind && subject.Name == user.NodesGroup { continue } subjects = append(subjects, subject) } binding.Subjects = subjects } return binding }) // ClusterRoleBindings return default rolebindings to the default roles func ClusterRoleBindings() []rbac.ClusterRoleBinding { rolebindings := []rbac.ClusterRoleBinding{ rbac.NewClusterBinding("cluster-admin").Groups(user.SystemPrivilegedGroup).BindingOrDie(), rbac.NewClusterBinding("system:discovery").Groups(user.AllAuthenticated, user.AllUnauthenticated).BindingOrDie(), rbac.NewClusterBinding("system:basic-user").Groups(user.AllAuthenticated, user.AllUnauthenticated).BindingOrDie(), rbac.NewClusterBinding("system:node-proxier").Users(user.KubeProxy).BindingOrDie(), rbac.NewClusterBinding("system:kube-controller-manager").Users(user.KubeControllerManager).BindingOrDie(), rbac.NewClusterBinding("system:kube-dns").SAs("kube-system", "kube-dns").BindingOrDie(), rbac.NewClusterBinding("system:kube-scheduler").Users(user.KubeScheduler).BindingOrDie(), // This default system:nodes binding is deprecated in 1.7 with the availability of the Node authorizer. // If an admin wants to grant the system:node role (which cannot partition Node API access), they will need to create their own clusterrolebinding. // TODO: Remove the subjects from this binding in 1.8 (leave the empty binding for tightening reconciliation), and remove AddClusterRoleBindingFilter() rbac.NewClusterBinding(systemNodeRoleName).Groups(user.NodesGroup).BindingOrDie(), } addClusterRoleBindingLabel(rolebindings) retval := []rbac.ClusterRoleBinding{} for i := range rolebindings { binding := &rolebindings[i] for _, filter := range clusterRoleBindingFilters { binding = filter(binding) if binding == nil { break } } if binding != nil { retval = append(retval, *binding) } } return retval }