Adding a New Node to a Sisense Cluster

[vsolodkyi]

Adding a New Node to a Sisense Cluster

The process of updating the nodes varies depending on your environment. Below, we outline the steps for both AWS EKS deployments with autoscaling enabled and self-hosted clusters.

For AWS EKS Deployments with Autoscaling:

If you deployed your Sisense cluster on AWS EKS and configured autoscaling in the Sisense installation configuration file (autoscaling: true), adding a new node is streamlined:

1. Create a new node group in EKS with the desired configuration.
2. Ensure the new node group is in the same availability zone as the node you intend to replace.
3. Once the new node is up and running, it will automatically join your Sisense cluster.
4. Safely drain and shut down the old node to complete the process.

For Self-Hosted Clusters:

If you are using a self-hosted cluster, follow these steps to add a new node:

Adding the New Node:

1. Open the cluster_config.yaml file.
2. Set "new_node: yes" and "update: true".
3. Specify the details of the new node in the k8s_nodes section, adjusting the information as necessary. Here is an example:
   

k8s_nodes:

  - { node: 1, internal_ip: 0.0.0.0, external_ip: 0.0.0.0, disk_volume_device: /dev/sdb, roles: "application, query" }

  - { node: 2, internal_ip: 0.0.0.0, external_ip: 0.0.0.0, disk_volume_device: /dev/sdb, roles: "application, query" }

  - { node: 3, internal_ip: 0.0.0.0, external_ip: 0.0.0.0, disk_volume_device: /dev/sdb, roles: "build" }

  - { node: new-node, internal_ip: 0.0.0.0, external_ip: 0.0.0.0, disk_volume_device: /dev/sdb, roles: "application, query" }

1. Run the installation script with the updated configuration file:

./sisense.sh cluster_config.yaml

Removing the Old Node:

1. In the cluster_config.yaml file, set "remove_node: true".
2. Specify the node to remove using "node_to_remove: node-name".
3. Run the installation script again to complete the removal process:

./sisense.sh cluster_config.yaml

By following these steps, you can seamlessly expand your Sisense cluster, ensuring high availability and enhanced performance for your analytics workloads.

If you have any questions or encounter any issues during this process, feel free to reach out to our support team for assistance.