Skip to main content

Deploy Airbyte on Kubernetes using Helm


Airbyte allows scaling sync workloads horizontally using Kubernetes. The core components (api server, scheduler, etc) run as deployments while the scheduler launches connector-related pods on different nodes.


If you don't want to configure your own Kubernetes cluster and Airbyte instance, you can use the free, open-source project Plural to bring up a Kubernetes cluster and Airbyte for you. Use this guide to get started.

Alternatively, you can deploy Airbyte on Restack to provision your Kubernetes cluster on AWS. Follow this guide to get started.

Getting Started

Cluster Setup

For local testing we recommend following one of the following setup guides:

For testing on GKE you can create a cluster with the command line or the Cloud Console UI.

For testing on EKS you can install eksctl and run eksctl create cluster to create an EKS cluster/VPC/subnets/etc. This process should take 10-15 minutes.

For production, Airbyte should function on most clusters v1.19 and above. We have tested support on GKE and EKS. If you run into a problem starting Airbyte, please reach out on the #troubleshooting channel on our Slack or create an issue on GitHub.

Install kubectl

If you do not already have the CLI tool kubectl installed, please follow these instructions to install.

Configure kubectl

Configure kubectl to connect to your cluster by using kubectl use-context my-cluster-name.

For GKE:

  1. Configure gcloud with gcloud auth login.

  2. On the Google Cloud Console, the cluster page will have a Connect button, which will give a command to run locally that looks like

    gcloud container clusters get-credentials $CLUSTER_NAME --zone $ZONE_NAME --project $PROJECT_NAME.

  3. Use kubectl config get-contexts to show the contexts available.

  4. Run kubectl config use-context $GKE_CONTEXT to access the cluster from kubectl.

For EKS:

  1. Configure your AWS CLI to connect to your project.
  2. Install eksctl
  3. Run eksctl utils write-kubeconfig --cluster=<CLUSTER NAME> to make the context available to kubectl
  4. Use kubectl config get-contexts to show the contexts available.
  5. Run kubectl config use-context <eks context> to access the cluster with kubectl.

Install helm

To install helm simply run:

For MacOS:

brew install helm

For Linux:

  1. Download installer script curl -fsSL -o
  2. Assign required premissions chmod 700
  3. Run script ./

Add Helm Repository

From now charts are stored in helm-repo thus there're no need to clone the repo each time you need to deploy the chart.

To add remote helm repo simply run: helm repo add airbyte

Where airbyte is the name of the repository that will be indexed locally.

After adding the repo, perform the repo indexing process by running helm repo update.

After this you can browse all charts uploaded to repository by running helm search repo airbyte

It'll produce the output below:

NAME                            CHART VERSION   APP VERSION     DESCRIPTION
airbyte-oss/airbyte 0.30.23 0.39.37-alpha Helm chart to deploy airbyte
airbyte-oss/airbyte-bootloader 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-bootloader
airbyte-oss/pod-sweeper 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-pod-sweeper
airbyte-oss/server 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-server
airbyte-oss/temporal 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-temporal
airbyte-oss/webapp 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-webapp
airbyte-oss/worker 0.30.23 0.39.37-alpha Helm chart to deploy airbyte-worker

Deploy Airbyte

Default deployment

If you don't intend to customise your deployment, you can deploy airbyte as is with default values.

In order to do so, run the command:

helm install %release_name% airbyte/airbyte

Custom deployment

In order to customize your deployment, you need to create values.yaml file in the local folder and populate it with default configuration override values.

values.yaml example can be located in charts/airbyte folder of the Airbyte repository.

After specifying your own configuration, run the following command:

helm install --values path/to/values.yaml %release_name% airbyte/airbyte

(Early Access) Airbyte Enterprise deployment

Airbyte Enterprise is in an early access stage, so this section will likely evolve. That said, if you have an Airbyte Enterprise license key and wish to install Airbyte Enterprise via helm, follow these steps:

  1. Checkout the latest revision of the airbyte-platform repository

  2. Add your Airbyte Enterprise license key and auth configuration details to a file called airbyte.yml in the configs directory of airbyte-platform. You can copy airbyte.sample.yml to use as a template:

cp configs/airbyte.sample.yml configs/airbyte.yml

Then, open up airbyte.yml in your text editor to fill in the indicated fields.


For now, auth configurations aren't easy to modify once initially installed, so please double check them to make sure they're accurate before proceeding! This will be improved in the near future.

  1. Make sure your helm repository is up to date:
helm repo update
  1. Install Airbyte Enterprise on helm using the following command:

The default release name is airbyte-pro. You can change this via the RELEASE_NAME environment variable.

Migrate from old charts to new ones

Starting from 0.39.37-alpha we've revisited helm charts structure and separated all components of airbyte into their own independent charts, thus by allowing our developers to test single component without deploying airbyte as a whole and by upgrading single component at a time.

In most cases upgrade from older monolith chart to a new one should go without any issue, but if you've configured custom logging or specified custom configuration of DB or Logging then follow the instructions listed bellow

Minio migration

Since the latest release of bitnami/minio chart, they've changed the way of setting up the credentials for accessing the minio. (written mid-2022)

Going forward in new version you need to specify the following values in values yaml for user/password instead old one


rootUser: airbyte-user
rootPassword: airbyte-password-123


rootUser: minio
rootPassword: minio123

Before upgrading the chart update values.yaml as stated above and then run:

  • Get the old rootPassword by running export ROOT_PASSWORD=$(kubectl get secret --namespace "default" %release_name%-minio -o jsonpath="{.data.root-password}" | base64 -d)
  • Perform upgrade of chart by running helm upgrade %release_name% airbyte/airbyte --set auth.rootPassword=$ROOT_PASSWORD
    • If you get an error about setting the auth.rootPassword, then you forgot to update the values.yaml file

Custom logging and jobs configuration

Starting from 0.39.37-alpha if you've configured logging yourself using logging or jobs section of values.yaml file, you need to update your configuration so you can continue to use your custom logging and jobs configuration.

Simply declare global value in values.yaml file and move everything related to logging and jobs under that section like in the example bellow:


After updating values.yaml simply upgrade your chart by running command:

helm upgrade -f path/to/values.yaml %release_name% airbyte/airbyte

Database external secrets

If you're using external DB secrets, then provide them in values.yaml under global.database section in the following format:

secretName: "myOctaviaSecret"
secretValue: "postgresql-password"
host: ""
port: "5432"

And upgrade the chart by running:

helm upgrade -f path/to/values.yaml %release_name% airbyte/airbyte