Quick Start

This guide walks through creating a RestoreTest, triggering an immediate run, and checking the result. Total time: approximately 5 minutes, assuming Kymaros is already installed and Velero has at least one backup available.

Step 1 — Confirm Velero Has Backups

velero backup get

You need at least one backup in a Completed state. Note the backup name — you will reference it in the RestoreTest. If you want Kymaros to always pick the most recent backup automatically, use the special value latest.

Step 2 — Create a RestoreTest

Save the following manifest to a file named my-first-test.yaml. Replace my-app with the name of a namespace that exists in the backup.

apiVersion: restore.kymaros.io/v1alpha1
kind: RestoreTest
metadata:
  name: my-first-test
  namespace: kymaros-system
spec:
  backupSource:
    provider: velero
    backupName: latest
    namespaces:
      - name: my-app
  schedule:
    cron: "0 3 * * *"
  sandbox:
    namespacePrefix: rp-test
    ttl: 30m
    networkIsolation: strict
  sla:
    maxRTO: "15m"

Apply it:

kubectl apply -f my-first-test.yaml

Key fields explained:

Field	Value	Meaning
`backupSource.backupName`	`latest`	Always restore from the most recent completed backup
`schedule.cron`	`0 3 * * *`	Run at 03:00 UTC every night
`sandbox.namespacePrefix`	`rp-test`	Sandbox namespace will be named `rp-test-<run-id>`
`sandbox.ttl`	`30m`	Sandbox is deleted 30 minutes after the run completes
`sandbox.networkIsolation`	`strict`	Sandbox cannot initiate outbound connections to production namespaces
`sla.maxRTO`	`15m`	A restore taking longer than 15 minutes fails the RTO check

Step 3 — Trigger an Immediate Run

The RestoreTest is scheduled for 03:00 UTC, but you can force an immediate run now with an annotation:

kubectl annotate restoretest my-first-test \
  kymaros.io/trigger=now \
  -n kymaros-system

The controller detects the annotation and starts a run within a few seconds.

Step 4 — Watch Progress

kubectl get restoretest my-first-test -n kymaros-system -w

The STATUS column progresses through these phases:

Phase	Meaning
`Pending`	Run queued, waiting to start
`Restoring`	Velero restore in progress
`Validating`	Health checks running in the sandbox
`Scoring`	Confidence score being calculated
`Completed`	Run finished — check the report
`Failed`	Run encountered a fatal error before scoring

A typical run for a small application takes 3–8 minutes.

To watch the sandbox namespace appear and the pods start:

kubectl get pods -n rp-test-<run-id> -w

The exact sandbox namespace name is shown in the RestoreTest status:

kubectl get restoretest my-first-test -n kymaros-system -o jsonpath='{.status.currentSandboxNamespace}'

Step 5 — Read the Report

Once the RestoreTest reaches Completed, a corresponding RestoreReport is written:

kubectl get restorereports -n kymaros-system

NAME                         SCORE   RESULT    AGE
my-first-test-20240115-0300   94      Pass      2m

Get the full report:

kubectl get restorereport my-first-test-20240115-0300 \
  -n kymaros-system \
  -o yaml

See Your First Report for a complete explanation of every field in the report output.

Next Steps

Your First Report — understand the score breakdown and validation levels
Dashboard Access — view the same report in the web UI

Step 1 — Confirm Velero Has Backups​

Step 2 — Create a RestoreTest​

Step 3 — Trigger an Immediate Run​

Step 4 — Watch Progress​

Step 5 — Read the Report​

Next Steps​