Disaster recovery: Active/Passive configurations

Install secondary cluster for disaster recovery in Active/Passive or Active/Active Automation Suite deployments.

Note:

Some Automation Suite products are not supported in Disaster Recovery - Active/Passive. You can install these products while installing the primary cluster only. For details, refer to Disaster recovery - Active/Passive.

The disaster recovery configuration requires that you install the two Automation Suite clusters separately. To install both the primary and secondary clusters in an Active/Passive deployment, you must configure the following input.json parameters:

Parameter	Description
fqdn	It represents the FQDN that, at the time of installation, points to the load balancer of the primary cluster. For details, refer to DNS routing logic.
cluster_fqdn	It represents the cluster-specific FQDN (DNS) that points to the load balancer of the cluster you set up using the input.json file. For details, refer to DNS routing logic.
multisite.enabled	It indicates that Automation Suite must be configured to work multi-site. It must be set to true.
multisite.primary	It indicates that this cluster is a primary cluster and must be set to true. It defaults to false to denote the secondary cluster.
multisite.other_kube_config	It indicates the base64-encoded kubeconfig file of another cluster. While installing the primary Automation Suite cluster, this value is unavailable and can be left as is. However, you must provide the value when rebuilding the primary automation suite later during recovery.
multisite.type	It specifies the deployment type. You must set it to active-passive.

Optional: Proxy configuration

If you choose to configure a proxy, in addition to the standard proxy configuration described in Configuring the proxy, make sure the no_proxy variable includes the following values:

• <traffic-manager-fqdn> - same value as fqdn in the disaster recovery configuration
• <primary-cluster-fqdn> - same value as cluster_fqdn set in the primary cluster
• <secondary-cluster-fqdn> - same value as cluster_fqdn set in the secondary cluster

The following example shows a valid proxy configuration:

"proxy": {
  "enabled": true,
  "http_proxy": "http://20.110.210.6:3128",
  "https_proxy": "http://20.110.210.6:3128",
  "no_proxy": "<secondary-cluster-fqdn>,<primary-cluster-fqdn>,<traffic-manager-fqdn>"
}

Multi-site configuration

This page describes how to set up a multi-site configuration with a primary and secondary cluster. The primary cluster is active and the secondary cluster is passive.

1. In the configuration for the primary cluster option, the enabled option must be set to true.

     "multisite": {
       "enabled": true,
       "primary": true,
       "type": "active-passive"
     }
   

2. In the configuration for the secondary cluster, the primary option must be set to false:

     "multisite": {
       "enabled": true,
       "primary": false,
       "other_kube_config": "[base64 encoded kubeconfig]",
       "type": "active-passive"
     }
   

   You must supply the primary kubeconfig in base64 encoded string.

3. Services that are not compatible with being in a passive state must be disabled. For details on services that do not support Active/Passive mode, refer to the Disaster recovery - Active/Passive page.

4. Ensure the certificates are consistent across the primary and secondary cluster, as this is not automatically checked or enforced.