Make it possible to bypass kubernetes service (#1614)

When running on K8s Patroni is communicating with API via the `kubernetes` service, which is address is exposed via the
`KUBERNETES_SERVICE_HOST` environment variable. Like any other service, the `kubernetes` service is handled by `kube-proxy`, that depending on configuration is either relying on userspace program or `iptables` for traffic routing.

During K8s upgrade, when master nodes are replaced, it is possible that `kube-proxy` doesn't update the service configuration in time and as a result Patroni fails to update the leader lock and demotes postgres.

In order to improve the user experience and get more control on the problem we make it possible to bypass the `kubernetes` service and connect directly to API nodes.
The strategy is very simple:
1. Resolve list IPs of API nodes from the kubernetes endpoint on every iteration of HA loop.
2. Stick to one of these IPs for API requests
3. Switch to a different IP if connected to IP is not from the list
4. If the request fails, switch to another IP and retry

Such a strategy is already used for Etcd and proven to work quite well.

In order to enable the feature, you need either to set to `true` `kubernetes.bypass_api_service` in the Patroni configuration file or `PATRONI_KUBERNETES_BYPASS_API_SERVICE` environment variable.

If for some reason `GET /default/endpoints/kubernetes` isn't allowed Patroni will disable the feature.

kubernetes/openshift-example/templates/template_patroni_ephemeral.yml

@@ -118,6 +118,8 @@ objects:
               fieldRef:
                 apiVersion: v1
                 fieldPath: metadata.namespace
+          - name: PATRONI_KUBERNETES_BYPASS_API_SERVICE
+            value: 'true'
           - name: PATRONI_KUBERNETES_LABELS
             value: '{application: ${APPLICATION_NAME}, cluster-name: ${PATRONI_CLUSTER_NAME}}'
           - name: PATRONI_SUPERUSER_USERNAME
@@ -250,6 +252,34 @@ objects:
   subjects:
   - kind: ServiceAccount
     name: ${SERVICE_ACCOUNT}
+# Following privileges are only required if deployed not in the "default"
+# namespace and you want Patroni to bypass kubernetes service
+# (PATRONI_KUBERNETES_BYPASS_API_SERVICE=true)
+- apiVersion: rbac.authorization.k8s.io/v1
+  kind: ClusterRole
+  metadata:
+    name: patroni-k8s-ep-access
+  rules:
+  - apiGroups:
+    - ""
+    resources:
+    - endpoints
+    resourceNames:
+    - kubernetes
+    verbs:
+    - get
+- apiVersion: rbac.authorization.k8s.io/v1
+  kind: ClusterRoleBinding
+  metadata:
+    name: ${NAMESPACE}-${SERVICE_ACCOUNT}-k8s-ep-access
+  roleRef:
+    apiGroup: rbac.authorization.k8s.io
+    kind: ClusterRole
+    name: patroni-k8s-ep-access
+  subjects:
+  - kind: ServiceAccount
+    name: ${SERVICE_ACCOUNT}
+    namespace: ${NAMESPACE}
 parameters:
 - description: The name of the application for labelling all artifacts.
   displayName: Application Name

kubernetes/openshift-example/templates/template_patroni_persistent.yaml

@@ -130,6 +130,8 @@ objects:
               fieldRef:
                 apiVersion: v1
                 fieldPath: metadata.namespace
+          - name: PATRONI_KUBERNETES_BYPASS_API_SERVICE
+            value: 'true'
           - name: PATRONI_KUBERNETES_LABELS
             value: '{application: ${APPLICATION_NAME}, cluster-name: ${PATRONI_CLUSTER_NAME}}'
           - name: PATRONI_SUPERUSER_USERNAME
@@ -274,6 +276,34 @@ objects:
   subjects:
   - kind: ServiceAccount
     name: ${SERVICE_ACCOUNT}
+# Following privileges are only required if deployed not in the "default"
+# namespace and you want Patroni to bypass kubernetes service
+# (PATRONI_KUBERNETES_BYPASS_API_SERVICE=true)
+- apiVersion: rbac.authorization.k8s.io/v1
+  kind: ClusterRole
+  metadata:
+    name: patroni-k8s-ep-access
+  rules:
+  - apiGroups:
+    - ""
+    resources:
+    - endpoints
+    resourceNames:
+    - kubernetes
+    verbs:
+    - get
+- apiVersion: rbac.authorization.k8s.io/v1
+  kind: ClusterRoleBinding
+  metadata:
+    name: ${NAMESPACE}-${SERVICE_ACCOUNT}-k8s-ep-access
+  roleRef:
+    apiGroup: rbac.authorization.k8s.io
+    kind: ClusterRole
+    name: patroni-k8s-ep-access
+  subjects:
+  - kind: ServiceAccount
+    name: ${SERVICE_ACCOUNT}
+    namespace: ${NAMESPACE}
 parameters:
 - description: The name of the application for labelling all artifacts.
   displayName: Application Name

kubernetes/patroni_k8s.yaml

@@ -58,6 +58,8 @@ spec:
           valueFrom:
             fieldRef:
               fieldPath: metadata.namespace
+        - name: PATRONI_KUBERNETES_BYPASS_API_SERVICE
+          value: 'true'
         - name: PATRONI_KUBERNETES_USE_ENDPOINTS
           value: 'true'
         - name: PATRONI_KUBERNETES_LABELS
@@ -239,3 +241,37 @@ roleRef:
 subjects:
 - kind: ServiceAccount
   name: patronidemo
+
+# Following privileges are only required if deployed not in the "default"
+# namespace and you want Patroni to bypass kubernetes service
+# (PATRONI_KUBERNETES_BYPASS_API_SERVICE=true)
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: patroni-k8s-ep-access
+rules:
+- apiGroups:
+  - ""
+  resources:
+  - endpoints
+  resourceNames:
+  - kubernetes
+  verbs:
+  - get
+
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: patroni-k8s-ep-access
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: patroni-k8s-ep-access
+subjects:
+- kind: ServiceAccount
+  name: patronidemo
+# The namespace must be specified explicitly.
+# If deploying to the different namespace you have to change it.
+  namespace: default