Files
patroni/kubernetes/patroni_k8s.yaml
T
Alexander KukushkinandGitHub 23dcfaab49 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.
2020-08-14 12:39:47 +02:00

278 lines
6.0 KiB
YAML

# headless service to avoid deletion of patronidemo-config endpoint
apiVersion: v1
kind: Service
metadata:
name: patronidemo-config
labels:
application: patroni
cluster-name: patronidemo
spec:
clusterIP: None
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: &cluster_name patronidemo
labels:
application: patroni
cluster-name: *cluster_name
spec:
replicas: 3
serviceName: *cluster_name
template:
metadata:
labels:
application: patroni
cluster-name: *cluster_name
spec:
serviceAccountName: patronidemo
containers:
- name: *cluster_name
image: patroni # docker build -t patroni .
imagePullPolicy: IfNotPresent
readinessProbe:
httpGet:
scheme: HTTP
path: /readiness
port: 8008
initialDelaySeconds: 3
periodSeconds: 10
timeoutSeconds: 5
successThreshold: 1
failureThreshold: 3
ports:
- containerPort: 8008
protocol: TCP
- containerPort: 5432
protocol: TCP
volumeMounts:
- mountPath: /home/postgres/pgdata
name: pgdata
env:
- name: PATRONI_KUBERNETES_POD_IP
valueFrom:
fieldRef:
fieldPath: status.podIP
- name: PATRONI_KUBERNETES_NAMESPACE
valueFrom:
fieldRef:
fieldPath: metadata.namespace
- name: PATRONI_KUBERNETES_BYPASS_API_SERVICE
value: 'true'
- name: PATRONI_KUBERNETES_USE_ENDPOINTS
value: 'true'
- name: PATRONI_KUBERNETES_LABELS
value: '{application: patroni, cluster-name: patronidemo}'
- name: PATRONI_SUPERUSER_USERNAME
value: postgres
- name: PATRONI_SUPERUSER_PASSWORD
valueFrom:
secretKeyRef:
name: *cluster_name
key: superuser-password
- name: PATRONI_REPLICATION_USERNAME
value: standby
- name: PATRONI_REPLICATION_PASSWORD
valueFrom:
secretKeyRef:
name: *cluster_name
key: replication-password
- name: PATRONI_SCOPE
value: *cluster_name
- name: PATRONI_NAME
valueFrom:
fieldRef:
fieldPath: metadata.name
- name: PATRONI_POSTGRESQL_DATA_DIR
value: /home/postgres/pgdata/pgroot/data
- name: PATRONI_POSTGRESQL_PGPASS
value: /tmp/pgpass
- name: PATRONI_POSTGRESQL_LISTEN
value: '0.0.0.0:5432'
- name: PATRONI_RESTAPI_LISTEN
value: '0.0.0.0:8008'
terminationGracePeriodSeconds: 0
volumes:
- name: pgdata
emptyDir: {}
# volumeClaimTemplates:
# - metadata:
# labels:
# application: spilo
# spilo-cluster: *cluster_name
# annotations:
# volume.alpha.kubernetes.io/storage-class: anything
# name: pgdata
# spec:
# accessModes:
# - ReadWriteOnce
# resources:
# requests:
# storage: 5Gi
---
apiVersion: v1
kind: Endpoints
metadata:
name: &cluster_name patronidemo
labels:
application: patroni
cluster-name: *cluster_name
subsets: []
---
apiVersion: v1
kind: Service
metadata:
name: &cluster_name patronidemo
labels:
application: patroni
cluster-name: *cluster_name
spec:
type: ClusterIP
ports:
- port: 5432
targetPort: 5432
---
apiVersion: v1
kind: Service
metadata:
name: patronidemo-repl
labels:
application: patroni
cluster-name: &cluster_name patronidemo
role: replica
spec:
type: ClusterIP
selector:
application: patroni
cluster-name: *cluster_name
role: replica
ports:
- port: 5432
targetPort: 5432
---
apiVersion: v1
kind: Secret
metadata:
name: &cluster_name patronidemo
labels:
application: patroni
cluster-name: *cluster_name
type: Opaque
data:
superuser-password: emFsYW5kbw==
replication-password: cmVwLXBhc3M=
---
apiVersion: v1
kind: ServiceAccount
metadata:
name: patronidemo
---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
name: patronidemo
rules:
- apiGroups:
- ""
resources:
- configmaps
verbs:
- create
- get
- list
- patch
- update
- watch
# delete and deletecollection are required only for 'patronictl remove'
- delete
- deletecollection
- apiGroups:
- ""
resources:
- endpoints
verbs:
- get
- patch
- update
# the following three privileges are necessary only when using endpoints
- create
- list
- watch
# delete and deletecollection are required only for for 'patronictl remove'
- delete
- deletecollection
- apiGroups:
- ""
resources:
- pods
verbs:
- get
- list
- patch
- update
- watch
# The following privilege is only necessary for creation of headless service
# for patronidemo-config endpoint, in order to prevent cleaning it up by the
# k8s master. You can avoid giving this privilege by explicitly creating the
# service like it is done in this manifest (lines 2..10)
- apiGroups:
- ""
resources:
- services
verbs:
- create
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
name: patronidemo
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: Role
name: patronidemo
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