High availability

KubeBlocks integrates the open-source Patroni solution to realize high availability and adopts Noop as the switch policy.

Before you start

• Install kbcli if you want to manage the PostreSQL cluster with kbcli.

• Install KubeBlocks.

• Create a PostgreSQL Replication Cluster.

• Check whether the following role probe parameters exist to verify the role probe is enabled.

  kubectl get cd postgresql -o yaml
  >
  probes:
    roleProbe:
      failureThreshold: 2
      periodSeconds: 1
      timeoutSeconds: 1

Steps

kubectl

1. View the initial status of the PostgreSQL cluster and pods.

   kubectl get cluster mycluster -n demo
   
   kubectl -n demo get pod -L kubeblocks.io/role

   

   Currently, mycluster-postgresql-0 is the primary pod and mycluster-postgresql-1 is the secondary pod.

2. Simulate a primary pod exception.

   # Enter the primary pod
   kubectl exec -it mycluster-postgresql-0 -n demo -- bash
   
   # Delete the data directory of PostgreSQL to simulate an exception
   root@mycluster-postgresql-0:/home/postgres# rm -fr /home/postgres/pgdata/pgroot/data

3. View logs to observe how the roles of pods switch when an exception occurs.

   # View the primary pod logs
   kubectl logs mycluster-postgresql-0 -n demo

   In the logs, the leader lock is released from the primary pod and an HA switch occurs.

   2024-05-17 02:41:23,523 INFO: Lock owner: mycluster-postgresql-0; I am mycluster-postgresql-0
   2024-05-17 02:41:23,702 INFO: Leader key released
   2024-05-17 02:41:23,904 INFO: released leader key voluntarily as data dir empty and currently leader
   2024-05-17 02:41:23,905 INFO: Lock owner: mycluster-postgresql-1; I am mycluster-postgresql-0
   2024-05-17 02:41:23,906 INFO: trying to bootstrap from leader 'mycluster-postgresql-1'

   # View secondary pod logs
   kubectl logs mycluster-postgresql-1 -n demo

   In the logs, the original secondary pod has obtained the lock and become the leader.

   2024-05-17 02:41:35,806 INFO: no action. I am (mycluster-postgresql-1), the leader with the lock
   2024-05-17 02:41:45,804 INFO: no action. I am (mycluster-postgresql-1), the leader with the lock

4. Connect to the PostgreSQL cluster to view the replication information.

   kubectl get secrets -n demo mycluster-conn-credential -o jsonpath='{.data.username}' | base64 -d
   >
   postgres
   
   kubectl get secrets -n demo mycluster-conn-credential -o jsonpath='{.data.password}' | base64 -d
   >
   shgkz4z9
   
   kubectl exec -ti -n demo mycluster-postgresql-1 -- bash
   
   root@mycluster-postgresql-0:/home/postgres# psql -U postgres -W
   Password: shgkz4z9

   postgres=# select * from pg_stat_replication;

   

   From the output, mycluster-postgresql-0 has been assigned as the secondary pod.

5. View the status of the PostgreSQL cluster and pods again.

   kubectl get cluster mycluster -n demo
   
   kubectl -n demo get pod -L kubeblocks.io/role

   

   After the failover, mycluster-postgresql-0 becomes the secondary pod and mycluster-postgresql-1 becomes the primary pod.

kbcli

1. View the initial status of the PostgreSQL cluster.

   kbcli cluster describe mycluster -n demo

   

   Currently, mycluster-postgresql-1 is the primary pod and mycluster-postgresql-0 is the secondary pod.

2. Simulate a primary pod exception.

   # Enter the primary pod
   kubectl exec -it mycluster-postgresql-1  -n demo -- bash
   
   # Delete the data directory of PostgreSQL to simulate an exception
   root@mycluster-postgresql-0:/home/postgres# rm -fr /home/postgres/pgdata/pgroot/data

3. View logs to observe how the roles of pods switch when an exception occurs.

   # View the primary pod logs
   kubectl logs mycluster-postgresql-1 -n demo

   In the logs, the leader lock is released from the primary pod and an HA switch occurs.

   2024-09-28 09:54:15,199 INFO: Lock owner: mycluster-postgresql-1; I am mycluster-postgresql-1
   2024-09-28 09:54:15,419 INFO: Leader key released
   2024-09-28 09:54:15,632 INFO: released leader key voluntarily as data dir empty and currently leader
   2024-09-28 09:54:15,634 INFO: Lock owner: mycluster-postgresql-0; I am mycluster-postgresql-1
   2024-09-28 09:54:15,635 INFO: trying to bootstrap from leader 'mycluster-postgresql-0'

   # View secondary pod logs
   kubectl logs mycluster-postgresql-0 -n demo

   In the logs, the original secondary pod has obtained the lock and become the leader.

   2024-09-28 09:54:17,117 INFO: no action. I am (mycluster-postgresql-0), the leader with the lock
   2024-09-28 09:54:17,516 INFO: no action. I am (mycluster-postgresql-0), the leader with the lock
   

4. Connect to the PostgreSQL cluster to view the replication information.

   kbcli cluster connect mycluster -n demo

   postgres=# select * from pg_stat_replication;

   

   From the output, mycluster-postgresql-1 has been assigned as the secondary pod.

5. Describe the cluster and check the instance role.

   kbcli cluster describe mycluster -n demo

   

   After the failover, mycluster-postgresql-1 becomes the secondary pod and mycluster-postgresql-0 becomes the primary pod.