Kubernetes rake <arbitrary tasks> job runner

kuberake

#!/bin/bash

function join_by { local d=$1; shift; echo -n "$1"; shift; printf "%s" "${@/#/$d}"; }

template=`cat ./jobs/rake.yaml`
joined=$(join_by \",\" $@)
args=[\"$joined\"]
job="${template//"\$\$RAKECMD"/$args}"
echo "$job" | kubectl create -f -
kubectl get pods -a

zjobs_rake.yaml

apiVersion: batch/v1
kind: Job
metadata:
  name: rake
spec:
  template:
    metadata:
      name: rake
    spec:
      restartPolicy: Never
      containers:
      - name: rails
        image: us.gcr.io/counting-company-production/rails
        envFrom:
          - secretRef:
              name: rails-environment
        volumeMounts:
          - name: cloudstorage-instance-credentials
            mountPath: /secrets
            readOnly: true
        command: ["rake"]
        args: $$RAKECMD
      volumes:
        - name: cloudstorage-instance-credentials
          secret:
            secretName: cloudstorage-instance-credentials

Usage:

./kuberake db:migrate db:seed

output

NAME                                READY     STATUS      RESTARTS   AGE
cloudsqlproxy-1733271362-kzrrc      1/1       Running     0          1h
ml-deployment-1367760623-9j1bl      1/1       Running     0          1h
rails-deployment-4017203545-fc324   1/1       Running     0          1h
rake-bbxvc                          0/1       Completed   0          38s

Used fancy bash join found here: https://stackoverflow.com/questions/1527049/join-elements-of-an-array