Deployment

The recommended means of deployment for a production web server of Beacon-python is via a container image (e.g. Docker image). In this section we illustrate several means of building and running a beacon-python application via a Docker container image.

Dockerfile

Using vanilla docker in order to build the image - the tag can be customised:

$ git clone https://github.com/CSCfi/beacon-python
$ docker build -t cscfi/beacon-python .
$ docker run -p 5050:5050 cscfi/beacon-python

Source to Image

Using OpenShift’s s2i means of building the Docker image requires source2image installed.

$ git clone https://github.com/CSCfi/beacon-python
$ cd beacon-python
$ s2i build . centos/python-36-centos7 cscfi/beacon-python

After the image has been built, one can use it with the simple Docker run (requires connection to a DB in the same Docker network) or as part of a docker-compose file or as illustrated below with Kubernetes.

$ docker run -p 5050:5050 cscfi/beacon-python

Docker Compose

$ git clone https://github.com/CSCfi/beacon-python
$ cd beacon-python/deploy
$ docker build -t cscfi/beacon-python .
$ docker-compose up -d

Kubernetes Integration

For use with Kubernetes we provide YAML configuration.

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    role: beacon
  name: beaconpy
  namespace: beacon
spec:
  selector:
    matchLabels:
      app: beaconpy
  template:
    metadata:
      labels:
        app: beaconpy
        role: beacon
    spec:
      containers:
        - image: cscfi/beacon-python
          imagePullPolicy: Always
          name: beacon
          ports:
            - containerPort: 5050
              name: beaconpy
              protocol: TCP
          volumeMounts:
            - mountPath: /files
              name: data
      volumes:
        - name: data
        # change below with preferred volume class
          hostPath:
            path: /local/disk/path
---
apiVersion: v1
kind: Service
metadata:
  name: beacon
  labels:
    app: beaconpy
spec:
  type: NodePort
  ports:
    - port: 5050
      targetPort: 5050
      protocol: TCP
      name: web
  selector:
    app: beaconpy

1000 Genome Loader

Note

We use data from: 1000 Genome FTP.

For use with loading the whole 1000 genome dataset we provide a docker image cscfi/beacon-dataloader that downloads the whole 1000 genome vcf.gz files (>18GB disk space) and a YAML configuration for Kubernetes illustrated below.

The container uses the same Environment Variables specified at: Environment Setup and adds two more:

ENV	Default	Description
FTP_URL	ftp://ftp.1000genomes.ebi.ac.uk	The URL for the FTP server.
FTP_DIR	/vol1/ftp/release/20130502/	Name of the directory.

apiVersion: batch/v1
kind: Job
metadata:
  name: dataloader
spec:
  template:
    metadata:
      name: dataloader
    spec:
      containers:
      - name: dataloader
        image: cscfi/beacon-dataloader
        env:
        - name: TABLES_SCHEMA
          value: /app/init.sql
        - name: DATABASE_URL
          valueFrom:
            secretKeyRef:
              key: uri
              name:
        - name: DATABASE_NAME
          valueFrom:
             secretKeyRef:
                key: database_name
                name:
        - name: DATABASE_USER
          valueFrom:
            secretKeyRef:
              key: username
              name:
        - name: DATABASE_PASSWORD
          valueFrom:
            secretKeyRef:
              key: password
              name:
        volumeMounts:
        - name: data
          mountPath: /app/data
      restartPolicy: Never
      imagePullPolicy: Always
      volumes:
      - name: data
        persistentVolumeClaim:
          claimName: 1000genome