Skip to main content

Configuring an HA pair

Bringing up and configuring a standard or mirrored HA pair for the first time can require enabling HA mode capability and failover, setting options, configuring network connections, and testing the configuration.

These tasks apply to all HA pairs regardless of disk shelf type.

  1. Verifying and setting the HA state on the controller modules and chassis
    For systems that use the HA state value, the value must be consistent in all components in the HA pair. You can use the Maintenance mode ha-config command to verify and, if necessary, set the HA state.
  2. Setting the HA mode and enabling storage failover
    You need to set the HA mode and enable storage failover functionality to get the benefits of an HA pair.
  3. Enabling cluster HA and switchless-cluster in a two-node cluster
    A cluster consisting of only two nodes requires special configuration settings. Cluster high availability (HA) differs from the HA provided by storage failover, and is required in a cluster if it contains only two nodes. If you have a switchless configuration, the switchless-cluster option must be enabled. Detection of a switchless cluster is automatic.
  4. Configuring hardware-assisted takeover
    You can configure hardware-assisted takeover to speed up takeover times. Hardware-assisted takeover uses the remote management device to quickly communicate local status changes to the partner node.
  5. Configuring automatic takeover
    Automatic takeover is enabled by default. You can control when automatic takeovers occur by using specific commands.
  6. Configuring automatic giveback
    You can configure automatic giveback so that when a node that has been taken over boots up to the Waiting for Giveback state, giveback automatically occurs.
  7. Testing takeover and giveback
    After you configure all aspects of your HA pair, you need to verify that it is operating as expected in maintaining uninterrupted access to both nodes' storage during takeover and giveback operations. Throughout the takeover process, the local (or takeover) node should continue serving the data normally provided by the partner node. During giveback, control and delivery of the partner's storage should return to the partner node.