DocsInstall with K8S
Community Maintained Helm Chart
We recommend you use the community maintained helm chart to install Baserow on K8S.
Raw K8S starting point
See below for a starting point for a K8S configuration file which deploys a production ready Baserow.
It assumes you want the most performant version of Baserow possible and so deploys separate wsgi and asgi backend services, separate async task workers etc.
You will need to also provide a redis and postgres instance configured using the environment variables below. See Configuring Baserow for more details on these variables.
You should also set up and configure Baserow to use an S3 compatible storage service for uploading and serving user uploaded files.
Example baserow.yml
# A secret containing all the required Baserow settings.
apiVersion: v1
kind: Secret
metadata:
name: YOUR_ENV_SECRET_REF
type: Opaque
stringData:
SECRET_KEY: "TODO"
PRIVATE_BACKEND_URL: "http://backend-wsgi"
PUBLIC_BACKEND_URL: "TODO"
PUBLIC_WEB_FRONTEND_URL: "TODO"
DATABASE_HOST: "TODO"
DATABASE_USER: "TODO"
DATABASE_PASSWORD: "TODO"
DATABASE_PORT: "TODO"
DATABASE_NAME: "TODO"
REDIS_HOST: "TODO"
REDIS_PORT: "TODO"
REDIS_USER: "TODO"
REDIS_PASSWORD: "TODO"
REDIS_PROTOCOL: "TODO rediss or redis"
BASEROW_AMOUNT_OF_GUNICORN_WORKERS: "5"
# S3 Compatible storage is recommended with K8S to get the exports and file storage working
# See the docs for more info https://baserow.io/docs/installation%2Fconfiguration#user-file-upload-configuration
AWS_ACCESS_KEY_ID: "TODO"
AWS_SECRET_ACCESS_KEY: "TODO"
AWS_STORAGE_BUCKET_NAME: "TODO"
# An example ingress controller routing to the correct services
---
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: balancer
annotations:
nginx.ingress.kubernetes.io/affinity: "cookie"
nginx.ingress.kubernetes.io/session-cookie-name: "route"
nginx.ingress.kubernetes.io/session-cookie-expires: "172800"
nginx.ingress.kubernetes.io/session-cookie-max-age: "172800"
nginx.ingress.kubernetes.io/proxy-body-size: 20m
kubernetes.io/ingress.class: nginx
# removed ssl settings, add in your own desired ones
spec:
# TODO a tsl block
rules:
- host: REPLACE_WITH_YOUR_BACKEND_HOST
http:
paths:
- pathType: Prefix
path: "/ws/"
backend:
service:
name: backend-asgi
port:
number: 80
- pathType: Prefix
path: "/"
backend:
service:
name: backend-wsgi
port:
number: 80
- host: REPLACE_WITH_YOUR_WEB_FRONTEND_HOST
http:
paths:
- pathType: Prefix
path: "/"
backend:
service:
name: web-frontend
port:
number: 80
---
apiVersion: v1
kind: Service
metadata:
name: backend-asgi
spec:
type: ClusterIP
ports:
- port: 80
targetPort: 8000
selector:
app: backend-asgi
---
apiVersion: v1
kind: Service
metadata:
name: backend-wsgi
spec:
type: ClusterIP
ports:
- port: 80
targetPort: 8000
selector:
app: backend-wsgi
---
apiVersion: v1
kind: Service
metadata:
name: web-frontend
spec:
type: ClusterIP
ports:
- port: 80
targetPort: 3000
selector:
app: web-frontend
# The backend ASGI worker handling websockets
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: backend-asgi
labels:
app: backend-asgi
spec:
replicas: 2
selector:
matchLabels:
app: backend-asgi
template:
metadata:
labels:
app: backend-asgi
spec:
affinity:
podAntiAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 1
podAffinityTerm:
labelSelector:
matchExpressions:
- key: app
operator: In
values:
- backend-asgi
topologyKey: "kubernetes.io/hostname"
containers:
- name: backend-asgi
image: baserow/backend:1.24.2
workingDir: /baserow
args:
- "gunicorn"
ports:
- containerPort: 8000
name: backend-asgi
imagePullPolicy: Always
readinessProbe:
exec:
command:
- curl
- --fail
- --silent
- http://localhost:8000/api/_health/
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 5
successThreshold: 1
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
imagePullSecrets:
- name: YOUR_PULL_SECRETS
# The backend WSGI worker handling normal http api requests
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: backend-wsgi
labels:
app: backend-wsgi
spec:
replicas: 2
selector:
matchLabels:
app: backend-wsgi
template:
metadata:
labels:
app: backend-wsgi
spec:
affinity:
podAntiAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 1
podAffinityTerm:
labelSelector:
matchExpressions:
- key: app
operator: In
values:
- backend-wsgi
topologyKey: "kubernetes.io/hostname"
containers:
- name: backend-wsgi
image: baserow/backend:1.24.2
workingDir: /baserow
args:
- "gunicorn-wsgi"
- "--timeout"
- "60"
ports:
- containerPort: 8000
name: backend-wsgi
imagePullPolicy: Always
readinessProbe:
exec:
command:
- curl
- --fail
- --silent
- http://localhost:8000/api/_health/
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 5
successThreshold: 1
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
imagePullSecrets:
- name: YOUR_PULL_SECRETS
# A set of celery workers handling realtime events, cleanup, async tasks etc.
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: backend-worker
labels:
app: backend-worker
spec:
replicas: 2
selector:
matchLabels:
app: backend-worker
template:
metadata:
labels:
app: backend-worker
spec:
# Set affinities to ensure the different replicas end up on different nodes.
affinity:
podAntiAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 1
podAffinityTerm:
labelSelector:
matchExpressions:
- key: app
operator: In
values:
- backend-worker
topologyKey: "kubernetes.io/hostname"
containers:
- name: backend-worker
image: baserow/backend:1.24.2
args:
- "celery-worker"
imagePullPolicy: Always
readinessProbe:
exec:
command:
- /bin/bash
- -c
- /baserow/backend/docker/docker-entrypoint.sh celery-worker-healthcheck
initialDelaySeconds: 10
timeoutSeconds: 10
periodSeconds: 10
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
- name: backend-export-worker
image: baserow/backend:1.24.2
args:
- "celery-exportworker"
imagePullPolicy: Always
readinessProbe:
exec:
command:
- /bin/bash
- -c
- /baserow/backend/docker/docker-entrypoint.sh celery-exportworker-healthcheck
initialDelaySeconds: 10
timeoutSeconds: 10
periodSeconds: 10
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
- name: backend-beat-worker
image: baserow/backend:1.24.2
args:
- "celery-beat"
imagePullPolicy: Always
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
imagePullSecrets:
- name: YOUR_PULL_SECRETS
# A web-frontend SSR server which renders the initial html/js when a client visits.
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: web-frontend
labels:
app: web-frontend
spec:
replicas: 3
selector:
matchLabels:
app: web-frontend
template:
metadata:
labels:
app: web-frontend
spec:
affinity:
podAntiAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 1
podAffinityTerm:
labelSelector:
matchExpressions:
- key: app
operator: In
values:
- web-frontend
topologyKey: "kubernetes.io/hostname"
containers:
- name: web-frontend
image: baserow/web-frontend:1.24.2
args:
- nuxt
ports:
- containerPort: 3000
name: web-frontend
imagePullPolicy: Always
readinessProbe:
httpGet:
path: /_health
port: 3000
initialDelaySeconds: 5
periodSeconds: 5
successThreshold: 1
envFrom:
- secretRef:
name: YOUR_ENV_SECRET_REF
imagePullSecrets:
- name: YOUR_PULL_SECRETS