Configure a cluster

Requirements

How to create a cluster

1. Download senhasegura's application from the Virtual Appliances section in the PAM Solution Center.
2. Change the default application password.
3. Configure the application's Hostname.
4. Configure the NTP Server.
5. Configure the DNS Server.
6. Configure the Network.
7. Create a backup.
8. Back up your data and take a snapshot of each instance.

9. Go to Orbit Config Manager ➔ Replication ➔ Settings.
10. Change the Operating Mode to Cluster and enable replication.
11. Enable the File Synchronizer and set the Sync Timeout.
12. Add the IP address of each instance in the cluster.

13. Specify whether the Members are in different data centers.
14. Set a recovery screen display message in case of failures.
15. Click Save.
16. Click Yes to confirm that you want to change the settings of your database and reboot the service.

17. If this is the primary instance, click on Take over as Primary.

18. Check the instance's status to see if it is active after reboot and whether this first member of the cluster has been created successfully.
19. Repeat these steps on the next instance you want to add to the cluster.

Switch between instances automatically

You can enable and disable senhasegura instances remotely through HTTP requests. An instance is inactive and removed from load-balancing processes if it is under maintenance or unavailable.

To set up a list of IP addresses that are allowed to manage instances remotely:

1. Go to Orbit Config Manager ➔ Settings ➔ Recovery
2. Enable the parameter: Allow Remote System Wakeup.
3. Under Allowed source IPs for remote system activation, add the IP addresses with permission to send HTTP requests.
4. Click Save.

Now, these IP addresses can connect to your monitoring URL GET /flow/orbit/mntr .E.g., https://mysenhasegura/flow/orbit/mntr.

This URL displays the current state of the instance:

• HTTP 200: The application is active and available to users.
• HTTP 203: The application is active and available to users but is not the cluster's primary. 
• HTTP 403: The application has been deactivated and is unavailable to users. 
• HTTP 451: Expired activation license.
• HTTP 503: Application unavailable.

For example, if the administrator intentionally disables a particular member of the cluster, this member starts responding to load balancer requests as HTTP 403, and the load balancer stops redirecting user traffic to this instance. On the other hand, if an instance loses communication with the other members of the cluster, its database becomes unavailable, and this instance starts to send the HTTP 503 status to the load balancer, which, in turn, stops forwarding traffic to that particular instance.

Enable and disable instances automatically

Another useful feature of senhasegura clusters is the possibility to control which instances should be automatically activated and deactivated from an external system. Imagine a scenario where the load of an entire network must be redirected to a backup data center. In this case, the instance associated with the backup data center must be active and ready to receive a full load of requests, whereas the current instance must lose its role as the primary member of the cluster.

You can switch between instances using the activation/deactivation URLs below.

Activating an instance

This process makes an instance active and available to users as long as you have a valid activation license. If successful, this instance becomes the new primary member of the cluster, replacing the previous one. All the other instances are deactivated automatically.

Example: https://mysenhasegura/flow/orbit/mntr/activate.

Deactivating an instance

Makes an instance inactive. If this is the primary member of the cluster, please note that no other member of the cluster is automatically assigned as the new primary instance. Other inactive instances are not automatically re-activated either.

Example: https://mysenhasegura/flow/orbit/mntr/deactivate