NOTE: To run this demo you will need a K8s environment with 2 worker nodes at least.
We ship some kubernetes resources files in order to allow installation of the StackGres operator for demonstration purpose. Assuming you have already installed the the kubectl CLI you can install the operator with the following command:
kubectl apply -f https://stackgres.io/downloads/stackgres-k8s/stackgres/1.1.0/stackgres-operator-demo.yml
The
stackgres-operator-demo.yml
will expose the UI as with a LoadBalancer. Note that enabling this feature will probably incur in some fee that depend on the host of the kubernetes cluster (for example this is true for EKS, GKE and AKS).
Use the command below to be sure when the operation is ready to use:
kubectl wait -n stackgres deployment -l group=stackgres.io --for=condition=Available
Once it’s ready you will see that the pods are Running
:
➜ kubectl get pods -n stackgres -l group=stackgres.io
NAME READY STATUS RESTARTS AGE
stackgres-operator-78d57d4f55-pm8r2 1/1 Running 0 3m34s
stackgres-restapi-6ffd694fd5-hcpgp 2/2 Running 0 3m30s
To create your first StackGres cluster you have to create a simple custom resource that reflect the cluster configuration. Assuming you have already installed the kubectl CLI you can proceed by installing a StackGres cluster using the following command:
cat << 'EOF' | kubectl create -f -
apiVersion: stackgres.io/v1
kind: SGCluster
metadata:
name: simple
spec:
instances: 2
postgres:
version: 'latest'
pods:
persistentVolume:
size: '5Gi'
EOF
This will create a cluster using latest available PostgreSQL version with 2 nodes each with a disk of 5Gi using the default storage class and a set of default configurations for PostgreSQL, connection pooling and resource profile.
A cluster called simple
will be deployed in the default namespace
that is configured in your environment (normally this is the namespace default
).
Follow the creation status:
kubectl get pods --watch
As final status you should see something like this:
NAME READY STATUS RESTARTS AGE
simple-0 6/6 Running 0 2m50s
simple-1 6/6 Running 0 1m56s
To open a psql console and manage the PostgreSQL cluster you may connect to the postgres-util
container of primary instance (with label role: master
):
kubectl exec -ti "$(kubectl get pod --selector app=StackGresCluster,cluster=true,role=master -o name)" -c postgres-util -- psql
IMPORTANT: Connecting directly trough the
postgres-util
sidecar will grant you access with the postgres user. It will work similar tosudo -i postgres -c psql
.
Please check about the postgres-util side car and how to connect to the postgres cluster for more details.
Now that the cluster is up and running, you can also open a shell in any instance to use patronictl and control the status of the cluster:
kubectl exec -ti "$(kubectl get pod --selector app=StackGresCluster,cluster=true,role=master -o name)" -c patroni -- patronictl list
You should see something similar to this:
+ Cluster: simple (6868989109118287945) ---------+----+-----------+
| Member | Host | Role | State | TL | Lag in MB |
+----------+------------------+--------+---------+----+-----------+
| simple-0 | 10.244.0.9:7433 | Leader | running | 1 | |
| simple-1 | 10.244.0.11:7433 | Replica| running | 1 | 0 |
+----------+------------------+--------+---------+----+-----------+
Now to test the automated failover, let’s simulate a disaster by killing the leader simple-0
:
kubectl delete pod simple-0
After deleted the leader simple-0
Patroni should perform the switchover, electing simple-1
as new leader and replace with a new container the simple-0
instance. After Patroni performs the failover operation, you can check the cluster status again:
kubectl exec -ti "$(kubectl get pod --selector app=StackGresCluster,cluster=true,role=master -o name)" -c patroni -- patronictl list
The final state of the failover will result with node simple-1
as the leader and simple-0
as the replica.
+ Cluster: simple (6868989109118287945) ---------+----+-----------+
| Member | Host | Role | State | TL | Lag in MB |
+----------+------------------+--------+---------+----+-----------+
| simple-0 | 10.244.0.9:7433 | Replica| running | 2 | 0 |
| simple-1 | 10.244.0.11:7433 | Leader | running | 2 | |
+----------+------------------+--------+---------+----+-----------+
Please check about the patroni-management for more details.
Now that you knows a little bit more about StackGres, you can easily manage all your clusters from the UI. It will ask for a username and a password. By default those are admin
and a randomly generated password. You can run the command below to get the user and password auto-generated:
kubectl get secret -n stackgres stackgres-restapi --template 'username = {{ printf "%s\n" (.data.k8sUsername | base64decode) }}password = {{ printf "%s\n" ( .data.clearPassword | base64decode) }}'
With the credentials in hand, let’s connect to the Web UI of the operator, for this you may forward port 443 of the operator pod:
POD_NAME=$(kubectl get pods --namespace stackgres -l "app=stackgres-restapi" -o jsonpath="{.items[0].metadata.name}")
kubectl port-forward "$POD_NAME" 8443:9443 --namespace stackgres
Then open the browser at following address localhost:8443/admin/
To uninstall all resources generated by this demo you can run:
kubectl delete --ignore-not-found -f https://stackgres.io/downloads/stackgres-k8s/stackgres/1.1.0/stackgres-operator-demo.yml
Check the uninstall section for more details.
Also, see installation via helm section in order to change those.