Run Metis on top of StackGres

This runbook will show you how to install Metis on Kubernetes, with a production-grade database provided by StackGres. MetisData guardrail enables a proactive approach to database code, empowering devs to better understand, control, troubleshoot, fix, and own the entire data layer.improving their experiences and preventing code from breaking production.

Scenario

In this runbook we’ll assume that you already have a Kubernetes cluster with the StackGres operator installed. We will create an SGCluster with a configuration that fits MetisData’s requirements. You can find the example resources in the apps-on-stackgres GitHub repository. Please clone the repository and change to the examples/metisdata directory, where all the referenced files here are present.

Creating an SGCluster

To properly group all related resources together, let’s first create a namespace:

kind: Namespace
apiVersion: v1
metadata:
  name: metisdata

To apply run from within the examples/metisdata folder of the apps-on-stackgres GitHub repository:

kubectl apply -f 01-namespace.yaml

In this step we are creating the database script for our cluster:

apiVersion: stackgres.io/v1
kind: SGScript
metadata:
  name: create-db-script
  namespace: metisdata
spec:
  continueOnError: false
  managedVersions: true
  scripts:
  - id: 0
    name: create-metisdata-database
    script: |
            create database metisdata;
  - id: 1
    database: metisdata
    name: create-extension-pgstoreplans
    script: |
            create extension if not exists pg_store_plans;

kubectl apply -f 02-db-script.yaml

We are now ready to create the Postgres cluster:

apiVersion: stackgres.io/v1
kind: SGCluster
metadata:
  namespace: metisdata
  name: metisdata
spec:
  postgres:
    version: '15'
    extensions:
    - name: pg_store_plans
  instances: 1
  pods:
    persistentVolume:
      size: '5Gi'
  managedSql:
    scripts:
      - sgScript: create-db-script

Please execute:

kubectl apply -f 03-sgcluster.yaml

After some seconds to a few minutes, the cluster should be brought up:

kubectl -n metisdata get pods
NAME                           READY   STATUS    RESTARTS   AGE
postgres-0                     6/6     Running   0          16m

And the database metisdata should exist and being owned by the user with the same name:

kubectl -n metisdata exec -it metisdata-0 -c postgres-util -- psql -l 
                                             List of databases
   Name    |  Owner   | Encoding | Collate |  Ctype  | ICU Locale | Locale Provider |   Access privileges   
-----------+----------+----------+---------+---------+------------+-----------------+-----------------------
 metisdata | postgres | UTF8     | C.UTF-8 | C.UTF-8 |            | libc            | 
 postgres  | postgres | UTF8     | C.UTF-8 | C.UTF-8 |            | libc            | 
 template0 | postgres | UTF8     | C.UTF-8 | C.UTF-8 |            | libc            | =c/postgres          +
           |          |          |         |         |            |                 | postgres=CTc/postgres
 template1 | postgres | UTF8     | C.UTF-8 | C.UTF-8 |            | libc            | =c/postgres          +
           |          |          |         |         |            |                 | postgres=CTc/postgres

Deploy

Add the Metis Helm repository to your local Helm installation and update the Helm repository to ensure that you have the latest version:

helm repo metis-data https://metis-data.github.io/helm-charts/
helm repo update

Create helm chart with specific api-key and pg connection on your relevant namespace:

helm install metis-mmc metis-data/metis-md-collector \
  --set apiKey="*****1" \
  --set dbConnectionStrings=postgresql://postgres:*****2@postgres.metisdata.svc:5432/metisdata;

Where the: *****1 - Represents the API token from Metis, that you could find in the web console at: Metisdata-API

*****2 - Represents the StackGres password, that could be obtained by the command:

kubectl get secrets -n metisdata metisdata -o jsonpath='{.data.superuser-password}' | base64 -d

The output should be similar with:

NAME: metis-mmc
LAST DEPLOYED: Thu Jun  8 09:49:09 2023
NAMESPACE: default
STATUS: deployed
REVISION: 1
TEST SUITE: None

Cleanup

Deleting the namespace should clean all used resources:

kubectl delete namespace metisdata