Adding cluster nodes | Tdg

Adding cluster nodes

Добавление нового узла (экземпляра) в кластер TDG – это фактически та же операция развертывания.

In this example, you will take the already deployed TDG cluster, deploy a new TDG instance and configure it via web interface by creating a new replica set with the storage cluster role.

For deploying a new instance, you will use the same tar.gz package and the Ansible tool with the predefined inventory and playbook configuration as for the initial cluster deployment.

  1. Navigate to the deploy directory extracted earlier from the tar.gz package and edit the Ansible inventory file hosts.yml. You need to add the description of a new instance and its parameters. In the code block below, parameters that should be added are marked with the comments # <-- Add ....

        ### Instances ###
          storage_3:  # <-- Добавьте новый экземпляр и параметры конфигурации для него
              advertise_uri: ""
              http_port: 8085
              memtx_memory: 1073741824  # 1024 Mb
          ### Machines ###
              storage_3:  # <-- Добавьте этот экземпляр к hosts на vm1
              ansible_host: ""


    While editing hosts.yml, double-check the following parameters:

    • cartridge_package_path—use the same package of the same application version you used for initial cluster deployment.

    • cartridge_cluster_cookie—should be the same as it was during the initial deployment. Otherwise, the new instance won’t be included in the cluster.

  2. Deploy a new instance by using the deploy_without_topology.yml playbook:

    $ ansible-playbook -i hosts.yml --limit storage_3 playbooks/deploy_without_topology.yml

    The --limit option is used to apply the playbook steps only to the specified instance and avoid any changes for the existing ones.

  3. Откройте или обновите страницу с интерфейсом TDG. В нашем примере она находится по адресу Новый экземпляр появится на вкладке Cluster в разделе Unconfigured servers.

    Добавлен несконфигурированный экземпляр
  4. The last step is to create a new replica set with the “storage” role. For the “storage_3” instance, click Configure. In the Configure server dialog, specify the following parameters and click Create replica set:

    • Replica set name: storage_3

    • Roles: storage

    • Replica set weight: 1

    Настройка нового набора реплик

The Replica set weight parameter should be set to the same value as for other replica sets with the “storage” role. It is necessary to run automatic data rebalancing between the storages upon creating a new storage.

You can verify if rebalancing was done correctly by checking out the Buckets parameter: the value should be the same for storage instances on the same server ( in this example). Rebalancing process takes some time, so you may need to wait a bit and refresh the page to see the result in web interface.

Создан новый набор реплик
Found what you were looking for?