Skip to content
This repository was archived by the owner on Dec 31, 2025. It is now read-only.

Getting started with cctl

Jump to bottom
Arun Sriraman edited this page Apr 16, 2019 · 1 revision

Table of Contents

Introduction to the CLI

A single command is used for all cluster operations. All logging is donein the UI, so there are no logs on disk – other than Kubernetes-specific logs that could be found onany Kubernetes cluster.

To invoke the CLI, run: ​ 

  /opt/bin/cctl

Example 

 node1 ~ # /opt/bin/cctl
 Platform9 tool for Kubernetes cluster management. This tool lets you create, scale, backup and restoreyour on-premise Kubernetes cluster.
 Usage:  
   cctl [command]
 Available Commands:  
   backup      Create an archive with the current cctl state and an etcdsnapshot from the cluster.
   bundle      Used to create cctl bundle
   create      Used to create resources
   delete      Used to delete resources
   deploy      Used to deploy app to the cluster
   get         Display one or more resources
   help        Help about any command
   migrate     Migrate the state file to the current version
   recover     Used to recover the cluster
   restore     Restore the cctl state and etcd snapshot from an archive.
   snapshot    Used to get a snapshot
   status      Used to get status of the cluster
   upgrade     Used to upgrade the cluster
   version     Print version information
 Flags:
   -h, --help               help for cctl
   -l, --log-level string   set log level for output, permitted values debug,info, warn, error, fatal and panic (default "info")      
   --state string       state file (default "/etc/cctl-state.yaml")
 Use "cctl [command] --help" for more information about a command.

The CLI can be executed from any node in the cluster. The only requirement for this is that the cluster configuration file, located in /etc/cctl-state.yaml, must be synchronized between the nodes.

Cluster Configuration

Choose a cluster node to use for initial cluster configuration. We'll call this the "management node". From this node, execute the following steps to bring up a Multi-Master Kubernetes cluster:

Step 1 : Add Credential

Use ​cctl create credential​ to configure the username and private key to use for accessing all nodes in the cluster: 

 node1 ~ $ sudo /opt/bin/cctl create credential --user core --private-key/home/core/.ssh/id_rsa
 2018/07/16 12:50:36 Credentials created with user:root and private-key file:/home/core/.ssh/id_rsa

Step 2 : Create Cluster

Use ​cctl create cluster to create a cluster. This step includes the CIDR configuration for both the services and pod networks, as well as the VIP (optional) and Router ID (optional) for the VRRP-based Virtual IP address in front of the master node(s): 

 node1 ~ $ sudo /opt/bin/cctl create cluster --pod-network 192.168.0.0/16 --service-network 192.169.0.0/24 --vip 10.105.16.38 --router-id 201
 2018/07/16 12:51:08 Cluster created successfully.

Step 3: Creating a Cluster with Configurable Parameters

The cctl create cluster​ ​command also supports passing a YAML config file used to supply values for configurable parameters as follows: 

  node1 ~ $ sudo /opt/bin/cctl create cluster --pod-network 192.168.0.0/16 --service-network 192.169.0.0/24 --vip 10.105.16.38 --router-id 201 --cluster-config ./clusterconfig.yaml

Sample YAML config file: Following shows the default values used for current configurable parameters. 

 kubeAPIServer:
   service-node-port-range: "80-32767"
   allow-privileged: "true"
   secure-port: 6443
 kubeControllerManager:  
   pod-eviction-timeout: "20s"
 kubeScheduler:
 kubeProxy:  
   mode: iptables
 kubelet:  
   kubeAPIQPS: 20
   kubeAPIBurst: 40
   maxPods: 500
   failSwapOn: "false"

References:

Step 4 : Add Master Node(s)

To add one or more master nodes, run cctl create machine with --role master 

 node1 ~ $ sudo /opt/bin/cctl create machine --ip 10.0.0.51 --role master

To create additional master nodes, run the same command against the additional nodes – changing only the IP address.

Step 5 : Add Worker Node(s)

To add one or more worker nodes, run cctl create machine with --role node 

 node1 ~ $ sudo /opt/bin/cctl create machine --ip 10.0.0.54 --role node

Step 6 : Configure Kubectl

As soon as the master and worker nodes have been setup, the cluster is ready and can be managed with kubectl. Use “sudo” with non-root user as follows or export KUBECONFIG if user is root: 

 node1 ~ $ sudo KUBECONFIG=/etc/kubernetes/admin.conf /opt/bin/kubectl get nodes
 NAME  	STATUS	ROLES 	AGE   	VERSION
 node1 	Ready 	master	13m   	v1.11.9
 node2 	Ready 	master	3m    	v1.11.9
 node3 	Ready 	master	2m    	v1.11.9
 node4 	Ready 	<none>	1m    	v1.11.9

 node1 ~ # export KUBECONFIG=/etc/kubernetes/admin.conf
 node1 ~ # source <(kubectl completion bash)

 node1 ~ # kubectl get nodes
 NAME  	STATUS	ROLES 	AGE   	VERSION
 node1 	Ready 	master	13m   	v1.11.9
 node2 	Ready 	master	3m    	v1.11.9
 node3 	Ready 	master	2m    	v1.11.9
 node4 	Ready 	<none>	1m    	v1.11.9

Clone this wiki locally