Managing the cluster

This section contains information about managing the cluster servers.

Viewing the table of servers of the cluster

To view the table of cluster servers:

1. In a browser on any computer on which access to the Central Node server has been allowed, enter the IP address of any server in the Central Node cluster or the fully qualified domain name (FQDN) of the cluster into the the address bar of your browser.

   To ensure uninterrupted access to the program web interface, you can configure the Round Robin function on the DNS server. To do this, log in to the web interface of the first healthy server of the Central Node cluster.

   An input window for account credentials of the Kaspersky Anti Targeted Attack Platform user opens.

2. Enter the administrator user name "admin" and the password that was set during installation of the application.
3. Select the Local administrator check box.
4. Click Log in.

   This opens a web interface window in which you can manage application sizing.

5. Go to the Cluster section.

A window with a table will open.

The table contains the following information:

• Server type—server type depending on its role in the cluster.

  The following values can be displayed:

  • Storage.
  • Processing.
• Status—server status.

  The following values can be displayed:

  • Connected.
  • Not connected.
• Host name—server name.
• IP— IP address of the server.
• RAM—RAM load level of the server.
• CPU—CPU load level of the server.
• Action—Actions that you can perform with the server.

  The following action is available: Delete.

Adding a server to a cluster

To add a server to the cluster, you need to start the installation of Kaspersky Anti Targeted Attack Platform on this server and follow the steps to install the components. The added server is displayed in the cluster server list.

Increasing the disk space on the storage server

You can increase the disk space on an operational storage server by installing an additional disk.

To increase the disk space of the storage server by means of an additional disk, you need to contact Technical Support.

The server is configured in Technical Support Mode.

Decommissioning servers

To decommission an operational server, you need to contact Technical Support.

If a server fails, you can decommission it on your own.

To decommission an inoperable processing server:

1. Add a new processing server to the cluster.
2. Remove the server from the cluster.
3. Configure the sizing of the application for the new configuration.

The processing server will be decommissioned.

To decommission an inoperable storage server:

1. Add a new storage server to the cluster.
2. Contact Technical Support to remove the inoperable server from the cluster.

The storage server will be decommissioned.

Removing a server from a cluster

If you are using the

distributed solution

Two-level hierarchy of servers with Central Node components installed. This hierarchy allocates a primary control server (Primary Central Node (PCN)) and secondary servers (Secondary Central Nodes (SCN)).

multitenancy

Operation mode in which Kaspersky Anti Targeted Attack Platform is used to protect the infrastructure of multiple organizations or branch offices of the same organization simultaneously.

A removed server cannot be restored. Make sure that the selected server is not operational.

To remove a server from the cluster:

1. Log in to the web interface for sizing management.
2. Go to the Cluster section.
3. In the Action column, click the Delete link opposite the server that you want to remove.
4. Click Proceed.

The removal process will start. Removal may take about a day. Information about the removed server will not be displayed in the table of servers.

After removing the server, you can reconfigure the servers in the cluster or add a server with the same role to maintain the same level of application performance.

Starting up and shutting down the cluster

To shut down or start the cluster, we recommend contacting Technical Support. Do not shut down or start the cluster if you encounter problems with application health.

If you want to power off the healthy servers in the cluster, you must first shut down the cluster to avoid data loss.

To shut down a cluster:

1. In a browser on any computer on which access to the Central Node server has been allowed, enter the IP address of the server with the Central Node component into the browser's address bar.

   If you are using the high availability version of the application, you can enter the IP address of any server of the Central Node cluster or the fully qualified domain name (FQDN) of the cluster.

   To ensure uninterrupted access to the program web interface, you can configure the Round Robin function on the DNS server. To do this, log in to the web interface of the first healthy server of the Central Node cluster.

   An input window for account credentials of the Kaspersky Anti Targeted Attack Platform user opens.

2. Enter the administrator user name "admin" and the password that was set during installation of the application.
3. Select the Local administrator check box.
4. Click Log in.

   This opens a web interface window in which you can manage application sizing.

5. Go to the Cluster section.
6. Click the Shut down button.

The main components of the application are stopped. You can now power off the cluster servers.

To start up the cluster servers:

1. Disconnect power to the servers if it has not been previously disconnected.
2. Power on the storage server.
3. Power up the remaining servers.

The cluster servers will start up.

The scaling management web interface becomes available when more than half of the cluster servers are started. For example, if there are 7 servers in the cluster, the web interface will be available when 4 servers of the cluster are powered on.

Powering off servers in a cluster

If necessary, you can power off a server in the cluster in one of the following ways:

• End the session in the application menu.
• Power off the server over SSH or through the terminal.

To power off the server over SSH or through the terminal:

1. Sign in to the management console of the server that you want to power off over SSH or through a terminal.
2. Run the shutdown -h now command.

The server is powered off.

High availability of the application is not guaranteed when a server in the cluster is powered off.

The recommended interval between powering on a server and powering off another server is 6 hours.