Creating PostgreSQL instances

Edit online

A PostgreSQL cluster custom resource is a YAML file that defines the PostgreSQL cluster size, configuration, storage, and behavior. Create a cluster custom resource (CR) to create a PostgreSQL instance.

Before you begin

System owner responsibilities

Configure a suitable StorageClass for the tenant environment. For more information, see Configuring storage.
Link the storage class to tenant. For more information, see Handing over configuration files.
Publish cluster catalog on the Platform capabilities page. For more information, see Configuring cluster.
Enable the Postgres Operator Catalog setup in Platform capabilities page. For more information, see Setting up Postgres catalog.

Tenant user responsibilities

An instance of a cluster service is deployed in the workspace. For more information, see Provisioning clusters.
Access to the created cluster using Red Hat® OpenShift® Console or oc CLI. For more information, see Accessing clusters.
Deploy the Postgres Operator on tenant cluster. For more information, see Deploying Postgres operator on tenant cluster.

Note: In the following commands, my-postgres is used as an example PostgreSQL cluster name. Replace it with the name of your actual PostgreSQL cluster. Additionally, postgres-namespace is an example namespace and must be replaced with the namespace in which your PostgreSQL cluster is deployed.

Procedure

Verify which ClusterImageCatalogs are available in your cluster. To view the list, run the command:
```
oc get clusterimagecatalogs.pg.ibm.com
```
Verify that a namespace exists for PostgreSQL resources
```
oc get namespace <postgres-namespace>
```
If the namespace does not exist, create one
```
 oc create namespace <postgres-namespace>
```
Get block storage class name.
- View available storage classes
```
oc get storageclass
```
- Select a block storage class from the list.
- If no block storage class exist, contact the system owner or editor to setup storage class for your tenant.

Use the IBM sample PostgreSQL cluster CR YAML template and update configuration parameters with values from step 3.

Basic CR example

apiVersion: pg.ibm.com/v1                 
kind: Cluster 
metadata: 
 name: my-postgres 
 namespace: postgres-namespace           # From step 2 
spec: 
  imageCatalogRef: 
    apiGroup: pg.ibm.com
   kind: ClusterImageCatalog 
   name: postgresql-image-catalog-latest  # Replace with value from step 1 
   major: 16                              # PostgreSQL major version 
 
 instances: 3                             # Number of replicas 
   storage: 
   size: 10Gi                             # Storage size 
    storageClass: <storageclass-name>     # Storage class name  
   
 # Resource configuration 
 resources: 
   requests: 
     memory: "2Gi" 
      cpu: "1" 
   limits: 
     memory: "4Gi" 
      cpu: "2" 
 
 # Enable superuser access 
  enableSuperuserAccess: true

Note: For more information on configuring the PostgreSQL cluster, see Configuring TLS/SSL for secure connections and Configuring access from other clusters.

Validate the YAML file in dry-run mode.
- Validate on the server
```
oc apply -f <cluster yaml file name> --dry-run=server
```
- Validate on the client
```
oc apply -f <cluster yaml file name> --dry-run=client 
```
Resolve any errors before proceeding.
Deploy PostgreSQL cluster by applying the PostgreSQL cluster CR.
Apply using oc apply or by creating the resource directly through the OpenShift Console YAML editor:
```
oc apply -f <cluster yaml file name> 
```

What to do next

After applying the CR, verify the deployment:

Check cluster resource status

oc get clusters.pg.ibm.com -n postgres-namespace

Verify that PostgreSQL pods services and PVCs are created

oc describe clusters.pg.ibm.com my-postgres -n postgres-namespace

Verify that the cluster status is Running

oc get pods -n postgres-namespace -l pg.ibm.com/cluster=my-postgres