AWS S3 Sink¶

The IntegrationSink supports the Amazon Web Services (AWS) S3 service, through its aws.s3 property.

Amazon credentials¶

There are two options for authenticating to AWS.

Access key and secret¶

To use an IAM User access key and secret, create a Kubernetes Secret in the namespace of the resource. The Secret can be created like:

kubectl -n <namespace> create secret generic my-secret --from-literal=aws.accessKey=<accessKey> --from-literal=aws.secretKey=<secretKey>

Then in the IntegrationSink .spec.aws.auth section reference the Secret like this:

      auth:
        secret:
          ref:
            name: "my-secret"

Pod Default Credentials¶

If you are using IRSA or Pod Identity, you can create a Kubernetes ServiceAccount and associate it with an AWS IAM role. Then in the IntegrationSink .spec.aws.auth section specify the name of the ServiceAccount. This will assign the ServiceAccount to the Deployment resource created for the IntegrationSink.

      auth:
        serviceAccountName: "my-service-account"

AWS S3 Sink Example¶

Below is an IntegrationSink to send data to an Amazon S3 Bucket:

apiVersion: sinks.knative.dev/v1alpha1
kind: IntegrationSink
metadata:
  name: integration-sink-aws-s3
  namespace: knative-samples
spec:
  aws:
    s3:
      arn: "arn:aws:s3:::my-bucket"
      region: "eu-north-1"
    auth:
      secret:
        ref:
        name: "my-secret"

Inside of the aws.s3 object we define the name of the bucket (or arn) and its region. The credentials for the AWS service are referenced from the my-secret Kubernetes Secret

More details about the Apache Camel Kamelet aws-s3-sink.