From backup to test result.
You decide which backup to test and when. RestoreTrust runs the test in your cluster, records the result, and can notify you when it is ready.
How a backup test works
Download the backup
From the source you have chosen.
Restore
In a temporary test database.
Run your checks
Using the criteria you have defined.
Record the result
Status, logs, and results in the Console.
Send a notification
By email or webhook, if configured.
Set up access to your backups
For your first test, you need a RestoreTrust worker in your Kubernetes cluster. Assign it to a project and add your backup source to that project. The worker needs to be able to reach the storage location.
Store the credentials as a Kubernetes Secret in your cluster. In RestoreTrust, enter the Secret’s name and link it to the source. You then choose which file to download when you set up the test schedule.
Create your test schedule
Select the worker and decide what it should check:
- Choose a backup. Select the source and path. You can use the latest file or narrow the selection with a filename pattern. If needed, set a maximum age and file size limits.
- Choose the test type. For a database test, select the database, version, and backup format that match your backup. To check a file without restoring a database, use a file check.
- Define your checks. Set the results you expect, such as a query returning a particular value or a test step finishing within a time limit.
- Set the timing. Run tests daily or weekly. Hourly tests are available depending on your plan.
Once you have saved the schedule, you can start the first test manually or wait for the next scheduled run. See Sources for supported storage locations and backup formats.
What happens during a test
The worker downloads the selected backup. For a database test, it restores the backup into a temporary test database and runs your checks there. File checks verify the download and the file requirements you have set.
In the Console, you can follow the test and see which checks passed. If something fails, you can identify the affected step, such as the download or restore. After fixing the issue, you can start a new test. See Execution for more about the test environment.
Set up notifications
Select one or more active notification channels from your project in the test schedule. Receive results by email or through an HTTPS webhook that passes them to your own systems.
You can choose to be notified after every test or only when a test fails. You can also turn notifications off and review the results in the Console.
Use the results with your team and in audits
Depending on your plan, you can combine results from several tests in a PDF report. This records which backups you tested, when you tested them, and the outcome.
Each completed test also has a SHA-256 hash of its recorded test data. Once confirmed, an optional OpenTimestamps proof establishes that those data existed in that form no later than the confirmed time. See Reports for how to create reports and verify proofs independently.