Configure GSLB for disaster recovery

Disaster recovery capability is critical, because downtime is costly. A NetScaler appliance configured for GSLB forwards traffic to the least-loaded or the best-performing data center. This configuration, referred to as an active-active setup, not only improves performance, but also provides immediate disaster recovery by routing traffic to other data centers if a data center that is part of the setup goes down. Alternatively, you can configure an active-standby GSLB setup for disaster recovery only.

Configure GSLB for disaster recovery in an active-standby data center setup

A conventional disaster recovery setup includes an active data center and a standby data center. The standby data center is a remote site. When a failover occurs as a result of a disaster event that causes the primary active data center to be inactive, the standby data center becomes operational.

Configuring disaster recovery in an active-standby data-center setup consists of the following tasks.

  • Create the active data center.
    • Add a local GSLB site.
    • Add a GSLB vserver, which represents the active data center.
    • Bind the domain to the GSLB virtual server.
    • Add gslb services and bind the services to active GSLB virtual server.
  • Create the standby data center.
    • Add a remote gslb site.
    • Add a gslb vserver, which represents standby data center.
    • Add gslb services which represents standby data center and bind the services to the standby gslb vserver.
    • Designate the standby data center by configuring the standby GSLB virtual server as the backup virtual server for the active GSLB virtual server.

Once you have configured the primary data center, replicate the configuration for the backup data center and designate it as the standby GSLB site by designating a GSLB virtual server at that site as the backup virtual server.

For details on how to configure a basic GSLB setup, see Configuring GSLB Entities Individually.

To designate the standby GSLB site by using the command line interface

At both the active site and the remote site, at the command prompt, type:

set gslb vserver <name> -backupVserver <string>

Example:

set gslb vserver vserver-GSLB-1 -backupVServer vserver-GSLB-2

To configure the standby site by using the configuration utility

  1. Navigate to Traffic Management > GSLB > Virtual Servers and double-click the GSLB virtual server for the primary site.
  2. Click the Backup Virtual Server section and select a backup virtual server.

By default, once the primary virtual server becomes active, it starts receiving traffic. However, if you want the traffic to be directed to the backup virtual server even after the primary virtual server becomes active, use the ‘disable primary on down’ option.

Configure for disaster recovery in an active-active data center setup

An active-active GSLB deployment, in which both GSLB sites are active, removes any risk that may arise in having a standby data center. With such a setup, web or application content can be mirrored in geographically separate locations. This ensures that data is consistently available at each distributed data center.

To configure GSLB for disaster recovery in an active-active data center set up, you must first configure the basic GSLB setup on the first data center and then configure all other data centers.

First create at least two GSLB sites. Then, for the local site, create a GSLB virtual server and GSLB services and bind the services to the virtual server. Then create ADNS services and bind the domain for which you are configuring GSLB to the GSLB virtual server in the local site. Finally, at the local site, create a load balancing virtual server with the same virtual server IP address as the GSLB service.

Once you have configured the first data center, replicate the configuration for other data centers part of the setup.

For details on how to configure a basic GSLB setup, see Configuring GSLB Entities Individually.

Configuring for Disaster Recovery with Weighted Round Robin

When you configure GSLB to use the weighted round robin method, weights are added to the GSLB services and the configured percentage of incoming traffic is sent to each GSLB site. For example, you can configure your GSLB setup to forward 80 percent of the traffic to one site and 20 percent of the traffic to another. After you do this, the NetScaler appliance will send four requests to the first site for each request that it sends to the second.

To set up the weighted round robin method, first create two GSLB sites, local and remote. Next, for the local site create a GSLB virtual server and GSLB services, and bind the services to the virtual server. Configure the GSLB method as round robin. Next, create ADNS services and bind the domain for which you are configuring GSLB to the GSLB virtual server. Finally, create a load balancing virtual server with the same virtual server IP address as the GSLB service.

Each service that represents a physical server in the network has weights associated with it. Therefore the GSLB service is assigned a dynamic weight that is the sum of weights of all services bound to it. Traffic is then split between the GSLB services based on the ratio of the dynamic weight of the particular service to the total weight. You can also configure individual weights for each GSLB service instead of the dynamic weight.

If the services do not have weights associated with them, you can configure the GSLB virtual server to use the number of services bound to it to calculate the weight dynamically.

For details on how to configure a basic GSLB setup, see Configuring GSLB Entities Individually.

Once you configure a basic GSLB setup, you must configure the weighted round robin method such that the traffic is split between the configured GSLB sites according to the weights configured for the individual services.

To configure a virtual server to assign weights to services by using the command line interface

At the command prompt, type one of the following commands, depending upon whether you want to create a new load balancing virtual server or configure an existing one:

add lb vserver <name>@ -weight <WeightValue> <ServiceName>  
set lb vserver <name>@ -weight <WeightValue> <ServiceName>  

Example:

add lb vserver Vserver-LB-1 -weight 4 Service-HTTP-1
set lb vserver Vserver-LB-1 -weight 4 Service-HTTP-1

To set dynamic weight by using the command line interface

At the command prompt, type:

set gslb vserver <name> -dynamicWeight DynamicWeightType

Example:

set gslb vserver Vserver-GSLB-1 -dynamicWeight ServiceWeight

To add weights to the GSLB services by using the command line interface

At the command prompt, type:

set gslb vserver <name> -serviceName GSLBServiceName -weight WeightValue

Example:

set gslb vserver Vserver-GSLB-1 -serviceName Service-GSLB-1 -weight 1

To configure a virtual server to assign weights to services by using the configuration utility

  1. Navigate to Traffic Management > Load Balancing > Virtual Servers and double-click the virtual server (for example, Vserver-LB-1).
  2. Click the Services section and set the weight of a service.

To add weights to the GSLB services by using the configuration utility

  1. Navigate to Traffic Management > GSLB > Virtual Servers and double-click the virtual server (for example, vserver-GSLB-1)
  2. Click the Services section and set the weight of the service in the Weight field.

To set dynamic weight by using the configuration utility

  1. Navigate to Traffic Management > GSLB > Virtual Servers and double-click the virtual server (for example, vserver-GSLB-1).
  2. Click the Method section and, from the Dynamic Weight drop-down list select SERVICEWEIGHT.

Configuring for Disaster Recovery with Data Center Persistence

Data center persistence is required for web applications that require maintaining a connection with the same server instead of having the requests load balanced. For example, in an e-commerce portal, maintaining a connection between the client and the same server is critical. For such applications, HTTP redirect persistence can be configured in an active-active setup.

To configure GSLB for disaster recovery with data center persistence, you must first configure the basic GSLB set up and then configure HTTP redirect persistence.

First create two GSLB sites, local and remote. Next, for the local site, create a GSLB virtual server and GSLB services and bind the services to the virtual server. Next, create ADNS services and bind the domain for which you are configuring GSLB to the GSLB virtual server at the local site. Next, create a load balancing virtual server with the same virtual server IP address as the GSLB service. Finally, duplicate the previous steps for the remote configuration, or configure the NetScaler appliance to autosynchronize your GSLB configuration.

For details on how to configure a basic GSLB setup, see Configuring GSLB Entities Individually.

Once you have configured a basic GSLB setup, configure HTTP redirect precedence to enable data center persistence.

To configure HTTP redirect by using the command line interface

At the command prompt, type the following commands to configure HTTP redirect and verify the configuration:

set gslb service <serviceName> -sitePersistence <sitePersistence> -sitePrefix <string>
show gslb service <serviceName>

Example:

set gslb service Service-GSLB-1 -sitePersistence HTTPRedirect -sitePrefix vserver-GSLB-1
show gslb service Service-GSLB-1

To configure HTTP redirect by using the configuration utility

  1. Navigate to Traffic Management > GSLB > Services and double-click the GSLB service to be configured.
  2. Click the Site Persistence section, select the HTTPRedirect option, and in the Site Prefix text box, enter the site prefix (for example, vserver-GSLB-1).

Note

When site persistence is not configured and if a load balancing virtual server that is configured as a local GSLB service is DOWN, the HTTP requests are redirected to other healthy GSLB sites using a 302 redirect.