You can add a worker node to a managed edge Kubernetes cluster if the node can communicate with the Kubernetes API server. You can add Elastic Compute Service (ECS) instances, Edge Node Service (ENS) instances, and on-premises servers to a managed edge Kubernetes cluster.

Prerequisites

  • A managed edge Kubernetes cluster is created. For more information, see Create a managed edge cluster.
  • To enable the cluster to automatically add ENS instances, ENS must be activated.
Notice Limits on a managed edge Kubernetes cluster:
  • The managed edge Kubernetes cluster feature is in the public preview phase. Each cluster can contain up to 40 nodes. To add more nodes, you must Submit a ticket.
  • You can add nodes that run on CentOS 7.4, CentOS 7.6, or Ubuntu 18.04.
  • An ENS instance to be automatically added must have a 2-core CPU and 4 GB memory or a higher specification and must be running on CentOS 7.4 or 7.6.
  • If the version of a managed edge Kubernetes cluster is 1.14.8-aliyunedge.1 or later, Advanced RISC Machine (ARM) or ARM64 nodes can be added to this cluster. The cluster allows ARM nodes to run on CentOS 7.4 and ARM64 nodes to run on Ubuntu 18.04.

Add a node

  1. Log on to the Container Service console.
  2. Add an existing node in either of the following ways:
    • Method 1:
      1. In the left-side navigation pane under Container Service - Kubernetes, choose Cluster > Cluster. The Clusters page appears.
      2. Find the target cluster and choose More > Add Existing Node in the Actions column.
        Add an edge node
    • Method 2:
      1. In the left-side navigation pane under Container Service - Kubernetes, choose Cluster > Nodes. The Nodes page appears.
      2. Select the target cluster and click Add Existing Node in the upper-right corner.
        Add a node
  3. On the Add Node page, you can select the Manual mode to add existing instances.
    If you select the Manual mode, you can add only one instance at a time.
    Note In Manual mode, you can add ECS instances, ENS instances, and on-premises servers.
    1. Click Next.Manual mode
    2. On the page that appears, enter the node configuration. For more information about the parameters, see Parameters.
      Note If the version of a managed edge Kubernetes cluster is 1.14.8-aliyunedge.1 or later, Advanced RISC Machine (ARM) or ARM64 nodes can be added to this cluster.
    3. After the configuration is complete, click Next.Add instances in Manual mode
      Note
    4. On the Complete page, click Copy. Then, log on to the edge node to run the following command.Copy code
      wget http://aliacs-k8s-cn-beijing.oss-cn-beijing.aliyuncs.com/public/pkg/run/attach/1.14.8-aliyunedge.1/amd64/edgeadm -O edgeadm; chmod u+x edgeadm; ./edgeadm join --openapi-token=XXXXX --node-spec="{\"enableIptables\":true,\"quiet\":true,\"manageRuntime\":true,\"nodeNameOverride\":\"*.ack.edge\",\"allowedClusterAddons\":[\"kube-proxy\",\"flannel\",\"coredns\"]}" --region=cn-beijing
    If information shown in the following figure appears, the edge node is added.Node added

Parameters

Parameter Description Default value
flannelIface The name of the network interface card (NIC) that is used by the Flannel plug-in. The name of the NIC that is recorded in the default route entry of the node.
enableIptables Specifies whether to enable iptables. false
quiet Specifies whether to send yes as the answer to all questions during the node adding process. false
manageRuntime Specifies whether to use edgeadm to install and manage the runtime. false
nodeNameOverride The name of the node.
  • By default, this parameter is not specified, and the host name is used as the node name.
  • The value can be a random string that contains six characters.
  • The value can be a random string that is followed by a suffix. The random string contains six characters.
allowedClusterAddons The list of addons to be installed. By default, this parameter is not specified, which indicates that no addons are to be installed. For a standard edge node, set this parameter to ["kube-proxy","flannel","coredns"]. []
gpuVersion Specifies whether the node to be added is a GPU node. By default, this parameter is not specified. Supported GPU versions are Nvidia_Tesla_T4, Nvidia_Tesla_P4, and Nvidia_Tesla_P100. By default, this parameter is not specified, which indicates that the node to be added is not a GPU node.