Skip to main content

Creating destinations

A destination definition contains all the necessary information to execute a task on a cloud provider.

Destination Definition

  {      'id' : str,      'cloud_provider' : 'AWS' | 'Azure',       ... the additional elements are cloud provider-dependent, see sections below ...  }

Required:

  • id: unique identifier of the destination

  • cloud_provider: the cloud provider of this destination

AWS destinations#

  {      'id' : str,      'cloud_provider' : 'AWS',      'credentials' : {profile_name : str , aws_account_id : str}      'aws_region' : str      'aws_ecs_iam_role' : str      'aws_fargate_cluster' : str      'aws_cluster_vpc_subnet_1' : str      'aws_cluster_vpc_subnet_2' : str      'aws_lambda_iam_role' : str
  }

  • profile_name: name of the AWS credentials profile to use.

  • aws_account_id: id of AWS account to use

  • aws_region: AWS region to use

  • aws_ecs_iam_role: IAM role for ECS execution (containers)

  • aws_fargate_cluster: name of a Fargate cluster to use

  • aws_cluster_vpc_subnet_1: VPC subnet of the Fargate cluster

  • aws_cluster_vpc_subnet_2: VPC subnet of the Fargate cluster

  • aws_lambda_iam_role: IAM role for Lambda execution (functions)

Azure destinations#

  {      'id' : str,      'cloud_provider' : 'Azure',      'credentials' : {'azure_auth_file':str}      'azure_region' : str      'azure_resource_group' : str      'azure_container_registry' : str  }

  • azure_auth_file: path and filename of a file containing Azure authentication info.

  • azure_region: Azure region to use

  • azure_resource_group: name of the resource group to use

  • azure_container_registry: name of the container registry to use

Google destinations#

  {      'id' : str,      'cloud_provider' : 'Google',      'credentials' : {'service_account_file':str}      'google_location' : str      'google_project' : str      'google_storage_bucket' : str  }

  • service_account_file: path and filename of a file containing Google authentication info.

  • google_location: Google location (region) to use

  • google_project: name of the Google cloud project to use

  • google_storage_bucket: name of a bucket in Google cloud storage to store function code in, must exist under the given project