How to Snapshot and Rollback a PostgreSQL Helm Release on OpenShift

By June 7, 2019 August 15th, 2019 Blog, Kubernetes, storage

In this tutorial, we will create a snapshot of the PostgreSQL database that is deployed on OpenShift. We will then restore the database state using the roll back to a point-in-time snapshot feature.

Before you begin this tutorial, make sure you have installed ROBIN Storage on your existing OpenShift cluster, and your PostgreSQL database is deployed on OpenShift, has data loaded in it, and the Helm release is registered with ROBIN.

If you make a mistake, such as unintentionally deleting important data, you can simply undo it by restoring a snapshot. Snapshots allow you to restore the state of your application to a point-in-time.

ROBIN lets you snapshot not just the storage volumes (PVCs) but the entire database application including all its resources such as Pods, StatefulSets, PVCs, Services, ConfigMaps, etc. with a single command. To create a snapshot, run the following command.

robin snapshot create snap9movies movies --desc "contains 9 movies" --wait

Let us verify we have successfully created the snapshot.

robin snapshot list --app movies

You should see an output similar to this:

Snaptshot PostgreSQL on

We now have a snapshot of our entire database with information of all 9 movies.

Rolling back to a point-in-time using snapshot

We have 9 rows in our “movies” table. To test the snapshot and rollback functionality, let’s simulate a user error by deleting one movie, “June 9”, from the “movies” table.

PGPASSWORD="$POSTGRES_PASSWORD" psql -h $IP_ADDRESS -U postgres
-d testdb -c "DELETE from movies where title = 'June 9';"

Let us verify the movie titled “June 9” has been deleted.

PGPASSWORD="$POSTGRES_PASSWORD" psql -h $IP_ADDRESS -U postgres 
-d testdb -c "SELECT * from movies;"

You should see the row with the movie “June 9” does not exist in the table anymore.

Snapshot PostgreSQL on OpenShift 8 movies

Let’s run the following command to see the available snapshots:

robin app info movies

You should see an output similar to the following. Note the snapshot id, as we will use it in the next command.

Snapshot PostgreSQL on OpenShift - Snapshot ID

Now, let’s roll back to the point where we had 9 movies, including “June 9”, using the snapshot id displayed above.

robin app rollback movies Your_Snapshot_ID --wait
Snapshot PostgreSQL on OpenShift - Snapshot ID

To verify we have rolled back to 9 movies in the “movies” table, run the following command.

PGPASSWORD="$POSTGRES_PASSWORD" psql -h $IP_ADDRESS
-U postgres -d testdb -c "SELECT * from movies;"

You should see an output similar to the following:

Deploy PostgreSQL on OpenShift - Load data

We have successfully rolled back to our original state with 9 movies!

This concludes the Snapshot PostgreSQL on OpenShift tutorial. We can also clone the entire PostgreSQL database to improve collaboration across Dev/Test/Ops teams.

mm

Author Ripul Patel, Sr Software Engineer

More posts by Ripul Patel, Sr Software Engineer