Cortx ha Replace Node

Node Name

faulty node healthy node new node

Replace Node

1. Csm will do poweroff of faulty node
2. Provision can call remove node from healthy node cortx-ha setup-ees.yaml

  remove_node:
    script: /opt/seagate/cortx/ha/conf/script/cluster_update
    args: remove_node

3. Auth new node for HA: when new node up then configure auth for pacemaker ha

4. Then provision or any step for other component needed that can be performed

5. On new node for cortx-ha setup.yaml Run post-install, config, init

ha:
    post_install:
        script: /opt/seagate/cortx/ha/conf/script/ha_setup
        args: post_install

    init:
        script: /opt/seagate/cortx/ha/conf/script/ha_setup
        args: init

    config:
        script: /opt/seagate/cortx/ha/conf/script/ha_setup
        args: config

6. From healthy node run refresh context for cortx-ha

    refresh_config:
        script: /opt/seagate/cortx/ha/conf/script/ha_setup
        args: replace_node

7. At last all configuration done, Execute add node from healthy node from cortx-ha setup-ees.yaml

  add_node:
    script: /opt/seagate/cortx/ha/conf/script/cluster_update
    args: add_node

8. all strep should be done check $ pcs status it should take time to get node to stable.