Running an on-demand backup of the management database

Deploy the Backup CR to initiate a backup of the management database.

About this task

If you are taking a management database backup, you should also take a portal database backup for all registered portal services. Management and portal databases must be kept in sync.

Note: For OpenShift users: The steps that are detailed in this topic use the Kubernetes kubectl command. On OpenShift, use the equivalent oc command in its place.

Procedure

Verify the health of your ManagementCluster:
```
kubectl -n <management namespace> get ManagementCluster
```
The status should show Running, and all pods should be up:
```
NAME   READY   STATUS    VERSION      RECONCILED VERSION   AGE
mgmt   n/n     Running   10.0.8.0     10.0.8.0             19h
```
Note: On Cloud Pak for Integration a keycloak cluster is also shown in this output. Check the status on the ManagementCluster only.

Identify the management database cluster name:

kubectl -n <management namespace> get cluster

Example output that shows the database cluster name is m1-dc1-db:

NAME        AGE   INSTANCES   READY   STATUS                     PRIMARY
m1-dc1-db   8d    3           3       Cluster in healthy state   m1-dc1-db-1

Create a file called mgmtbackup_cr.yaml and paste in the following text:

apiVersion: postgresql.k8s.enterprisedb.io/v1
kind: Backup
metadata:
  generateName: mgmt-backup- # prefix for the name of the backup CR that is generated.
spec:
  cluster:
    name: <database cluster name>

where <database cluster name> is the cluster name that you identified in step 2.

Create the Backup CR from the mgmtbackup_cr.yaml file:

kubectl -n <management namespace> create -f mgmtbackup_cr.yaml

Verify that the backup is running with:

kubectl -n <management namespace> get backup -o custom-columns="name:.metadata.name,backupId:.status.backupId,endpoint:.status.endpointURL,path:.status.destinationPath,servername:.status.serverName,status:.status.phase"

The expected output when the backup is completed is:

name                 backupId          endpoint                                     path                       servername                                        status
test-backup-1.17.1   20230831T110022   https://apic-mgmt-s3proxy.default.svc:8765   s3://backups/backup01ibm   apic-mgmt-d54a2194-db-2023-08-31T11:49:10+01:00   completed

Do not delete the Backup CR when the backup is complete. The Backup CR is required to restore the backup.

What to do next

Take a database backup of any portal subsystems that are registered with this management subsystem: Running an on-demand backup of the portal database.