Labeling Azure Kubernetes Service Nodes

Labeling is a means for identifying application processing and qualifying the application as a candidate to run on a specific node. For example, labeling an AKS node with the label kafka=yes specifies that a Kafka instance will run on that node.

Labels required for AKS nodes include the following:

Label Purpose
kafka=yes Run Kafka
zk=yes Run ZooKeeper
fusion=yes Run Fusion
th-processing=yes Process Transformation Hub data
th-platform=yes Run Transformation Hub
intelligence=yes Run pods that manage functions and services for the ArcSight Intelligence capability
intelligence-spark=yes Run Analytics services for the ArcSight Intelligence capability
intelligence-datanode=yes Run pods that manage HDFS services for the ArcSight Intelligence capability
intelligence-namenode=yes

Run HDFS NameNode services for the ArcSight Intelligence capability

Add this label to only one node in the cluster

To label your AKS nodes:

  1. On your jump host, get a list of AKS nodes by running the following command:
    # kubectl get nodes
NAME                                STATUS   ROLES   AGE    VERSION
aks-agentpool-36457641-vmss000000   Ready    agent   137m   v1.13.11
aks-agentpool-36457641-vmss000001   Ready    agent   137m   v1.13.11
aks-agentpool-36457641-vmss000002   Ready    agent   137m   v1.13.11
  1. Label the first AKS node by running the following command:
    # kubectl label node <node-name> zk=yes kafka=yes th-processing=yes th-platform=yes
For example: # kubectl label node aks-agentpool-36457641-vmss000000 zk=yes kafka=yes th-processing=yes th-platform=yes
  1. Repeat the command in step 2 for each additional node.