Argo workflow clusters use a serverless model. They run workflows on Alibaba Cloud Container Compute Service (ACS) and optimize Kubernetes cluster parameters for efficient, elastic scheduling of large-scale workflows. This approach also uses the BestEffort Quality of Service (QoS) to reduce costs. This topic describes how to create an Argo workflow cluster.
Prerequisites
Distributed Cloud Container Platform for Kubernetes (ACK One) is activated.
ACS is activated.
The Resource Access Management (RAM) user must have the AliyunAdcpFullAccess permission to create a cluster. For more information, see Grant permissions to a RAM user.
Create a workflow cluster
Log on to the Argo Workflow Clusters console.
Click Create Workflow Cluster. In the panel that appears, configure the parameters and click Next. After dependency checks pass, click Create Cluster.
Parameter
Description
Cluster Name
Enter a name for the cluster.
Must be 1 to 63 characters long. It can contain letters, digits, underscores (_), or hyphens (-). It must start with a letter.
Region
Select the region where you want to deploy the cluster.
VPC
Select an existing VPC from the drop-down list.
vSwitch
Select an existing vSwitch from the drop-down list.
Resource Group
Assign the cluster to a resource group to simplify permission management and cost allocation.
A resource can belong to only one resource group.
Tags
Attach key-value pair tags to identify the cluster as a cloud resource.
API Server Access
No configuration is required. When you create an Argo workflow cluster, a pay-as-you-go private network Classic Load Balancer (CLB) instance is automatically created as the internal endpoint for the API Server.
ImportantDo not delete this instance. Doing so will prevent you from accessing the cluster API Server.
Create and Attach EIP
If enabled, an elastic IP address (EIP) is attached to the CLB instance used by the cluster API Server. This enables public network access to the cluster API Server.
Enable Components and Audit Log
Enables the API Server audit feature for the cluster to collect requests to the Kubernetes API and their results. To enable this feature later, see Collect container logs from ACK clusters and Use the API Server audit feature.
Delete a workflow cluster
Before deleting an Argo workflow cluster, delete all workflows in the cluster to release ACS resources.
Log on to the Argo Workflow Clusters console.
In the Actions column of the cluster that you want to delete, click
> Delete Cluster. In the confirmation dialog box, click OK.
References
Before you use the cluster, enable Argo Server and access the workflow console.