Quick Installation

This guide will use the easiest and most convenient way to install Kato. Help you quickly evaluate Kato.

Start Kato Console

You need to prepare a machine that can run Docker containers, which can be Linux, Windows, or Mac. Of course we recommend using a Linux server. If you have not installed the Docker service, you can install the Docker service first through the following command.

curl https://storage.googleapis.com/kato-files/install_docker | bash

Please note that this command only supports Linux x86 operating systems.

All-In-One Mode to Start the Console

docker run -d -p 7070:7070 -v ~/.ssh:/root/.ssh -v ~/katodata:/app/data \
      --name=kato-allinone --restart=always \

By default sqlite3 is used as the database, and the data is stored in the katodata directory under the user’s home directory. Kato 5.3 supports console data migration, which is convenient for subsequent data migration to the production environment. Please feel free to try it.

After the container is successfully created and started, access the machine’s port 7070 to access the Kato console.

Initialize Kato Cluster Resources

After the first registration is completed, you will be directed to the page for adding clusters. Kato needs to connect to computing resources to create and manage cloud native applications. Version 5.3 began to support direct connection to cloud resources (Alibaba Cloud ACK cluster), installed Kubernetes cluster resources, and bare metal resources provided by users.

  1. Docking cloud resources is suitable for users who use Alibaba Cloud resources (AWS, Huawei Cloud, Tencent Cloud or other cloud users can purchase Kubernetes services by themselves and use the Connect to Kubernetes cluster method for docking) Using cloud resources will automatically purchase RDS, ACK, NAS, SLB and other resources can quickly build a highly available production environment for you.

  2. Access to a Kubernetes cluster is suitable for users who have already built a Kubernetes cluster. You should have certain Kubernetes management capabilities. This method requires the cluster version to be 1.16 or higher.

  3. If you only have a Linux machine, please use the installation entry from the host, and Kato will automatically install the Kubernetes cluster for you.

Please choose an appropriate method to install and initialize the cluster based on the product guide. Reference document Add cluster

Please note that the cluster installation and initialization can be performed in parallel, and you can install and initialize multiple clusters at the same time.

After the task is completed, it is recommended to migrate the control to Kato for management, see reference document

Offline Installation

Please note that the v5.3.1 version does not currently support offline environment installation. For offline environment, please refer to 5.2 Offline Installation

Common Problems

Error reporting failed to connect to following etcd hosts when installing the cluster

The problem is that the console cannot connect to the node that reported the error. First, make sure that when configuring and planning cluster nodes, node initialization is correctly performed on all nodes, and password-free login settings are completed. When checking the method, execute ssh docker@nodeIP in the console container to log in directly without password.

Initialization of the Kato cluster is blocked for a long time in the Local processing of non-componentized images required by the system

The problem may occur in the connection to a self-built Kubernetes cluster. Please judge the problem by the output of the following command. A common situation is that the gateway component cannot be started due to port conflicts.

kubectl get katocluster katocluster -n rbd-system -o yaml
kubectl get pod -n rbd-system

If there is no kubectl command based on the Kubernetes cluster installed on the host, please refer to: install kubectl command