Mittari

Observation statistics demo service.

Requirements:

• Local:
  • Docker or uv
• Deploying to OpenShift:
  • oc
  • gh

Note:

• Caching can be enabled/disabled in env var CACHING_ON. Visit /flush/{cache_secret} to flush the cache.

Set up locally

Copy .env.example to .env.local (local development) and .env.openshift (production), and fill in values.

Run locally with Docker Compose (recommended)

docker compose up --build

Then open: http://localhost:8080

Run locally with uv (optional)

From the repo root, install dependencies into .venv and start Gunicorn with variables from .env.local:

uv sync
uv run --env-file .env.local -- sh -c 'gunicorn --bind "0.0.0.0:${PORT:-8080}" --workers "${GUNICORN_WORKERS:-1}" --reload wsgi:app'

Then open: http://localhost:8080

Run in production on OpenShift

1. Push to main. GitHub Actions builds and pushes the image to GHCR.

2. Wait for the workflow to finish: https://github.com/luomus/mittari/actions

3. Log in to OpenShift (command from the Rahti web UI) and select the project:

oc project mittari

4. Deploy the newest image (this also syncs .env.openshift to the cluster by default):

./scripts/deploy-openshift.sh

5. Verify rollout and running image:

oc rollout status deployment/mittari
oc get pods
oc get deployment mittari -o jsonpath='{.spec.template.spec.containers[0].image}{"\n"}'

The app is served at: https://mittari.2.rahtiapp.fi

First-time setup

1. Select the OpenShift project:

oc project mittari

2. Create a GHCR pull secret (needed if the image is private):

oc create secret docker-registry ghcr-pull-secret \
  --docker-server=ghcr.io \
  --docker-username=<github-username> \
  --docker-password=<github-token-with-read-packages> \
  --docker-email=<email> \
  --dry-run=client -o yaml | oc apply -f -

Use a GitHub token with read:packages.

3. Create app resources from the template:

oc process -f openshift/mittari-app.yaml | oc apply -f -

4. Put production values in .env.openshift, then sync them to the cluster:

./scripts/sync-openshift-env.sh

5. After the first successful workflow run on main, point the deployment at a real image (the template starts from ghcr.io/luomus/mittari:latest):

./scripts/deploy-openshift.sh

6. Verify:

oc rollout status deployment/mittari
oc get pods
oc get route mittari

Update production env later

./scripts/deploy-openshift.sh