Create services

Last published : Sep 30, 2026
The services tab comprises services and service profiles. Services represent the back-end servers in your VPC. These servers are the origin servers of your application. Service profiles contain the load balancing and SSL settings for these services. Two service profiles are available by default: Default HTTP Profile and Default HTTPS Profile. While creating a service, if you select one of these default profiles then creating a service profile is not mandatory. However, if you create a service profile, you must create a load balancer.
  1. Click Create Service.
  2. Specify values for the following parameters:
    • Name
    • Protocol: Select HTTPS to create an SSL service.
    • Port
    • Service Profile: Select an existing profile or create a service profile. For information about creating a service profile, see Create service profiles.
    • App Server Type: For an AWS environment, select from Auto Scaling Group, EC2 Instances, or Servers. Select an auto-scaling group, EC2 instance, or server type respectively. Only app servers from the selected environment are displayed. For an Azure environment, select from Azure VM Scale Set, Azure VMs, or IPAddress/FQDN. Select a scale set, aVM, or IP address/FQDN respectively.
      • Server Type/Address Type includes two options: IP address and FQDN. Select an option.
      • Specify an IP address or FQDN, and click Add Server. The details appear in the List of Servers table. You can view the status of the server and also enable or disable the server from the application dashboard. For more information, see View and change the back-end app server status.
  3. Click Create.
  4. Click Next.
Related information
Deliver a NetScaler® managed app

Self-heal slow application servers

Note: This feature is not currently supported in an Azure environment.
A poorly performing server can affect the performance of your applications. If an Autoscale™ group, or EC2 instances with IP address in AWS is configured, the self-healing capability of the service can detect a slow server and take remedial action accordingly.
For example, if the response time of one of the servers is consistently poor, the service identifies this server, and gracefully replaces it with a healthy server.
The service considers the following conditions while replacing a faulty server:
  • All persistent connections are honored.
  • Existing connections are completed.
  • New connections aren't accepted for the faulty server.
Related actions taken by the NetScaler Console service are logged and admins can view them in the Action History. For more information, see Monitor self-healing.

Configure detecting and auto-replacing a slow server

  1. Navigate to NetScaler® Managed Apps > App Delivery > New Application.
  2. Type a name for the application, and select an environment.
  3. Click Next.
  4. Click Create Service.
  5. In the Create Service page, type values for the following parameters:
    • Name
    • Protocol
    • Port
    • Service Profile
    • App Server Type - Select either Auto Scaling Group, or Servers
      You can auto-replace a slow server, if the server is added using an IP address and hosted as an EC2 instance in AWS.
      If you select Servers, select the IPAdress in the Address Type list, and enter the required IP address in the IP Address field. FQDN is not supported.
    • Select a group from the list.
  6. Slide the following toggles to enable the functionality:
    • Detect slow server
    • Auto-replace slow server

View and change the back-end app server status

The status of the back-end app servers that host the application is determined using health checks. The health checks send probes to the servers and based on the responses received, the status of the server is determined.
The NetScaler Console service allows you to view the status of the back-end app servers and their associated health checks. The back-end app server status information is helpful in troubleshooting. For example, consider a scenario where you are unable to access an application. One of the reasons for this scenario can be that the back-end app server is down or disabled. As part of troubleshooting, you can view the status of the back-end app server.
To view the status of the back-end app server, do the following:
  1. Navigate to the NetScaler Managed Apps > App Delivery and click the Analyzer icon in the Actions column.
  2. Click the required service tile. The list of back-end app servers associated with the service is listed in the lower pane.
  3. Click Get Status in the Actions column corresponding to the required server.
The status of the back-end app server might change over time. To view the latest status, click Refresh.
You can enable or disable the configured state of the back-end server without removing it from the application. For example, consider a planned network maintenance activity that needs to disable a back-end app server. In such a scenario, you can disable the required back-end app server without redeploying the application. You can enable back the server after the maintenance activity is completed.
To disable the back-end app server, click Disable. Optionally, select the Graceful checkbox and enter the Delay in seconds to disable the server gracefully.
Note:
After disabling a back-end server gracefully, the status shows Graceful: Yes only until the server is shutting down gracefully. Once the server is shut down, the status shows Graceful: No.
To enable the server, click Enable.

Monitor self-healing

The NetScaler managed application offer self-healing capability for the application server. The self-healing capability provides deep application analytics with improved application experience and security.
With the self-healing capability:
  • You can automatically detect, remediate, and replace the defective server with a healthy server whenever the performance of an application server degrades or starts to malfunction.
  • You can automatically detect the security attacks and take remediation actions.
The following use cases are supported as part of the self-healing capability:
  • Automatically replace Slow Server: In a stack of applications running as a back-end server in an Autoscale group, if one of the instances goes faulty and if that instance starts responding slowly based on the response time, then the self-healing feature enables detecting the faulty server and replacing the faulty server with the healthy server automatically.
  • Block past offenders: Enables you to check and send alerts if the client IPs are causing excessive AppSec violations in the last 20 minutes. For more information, see SQL injection and Cross-site scripting.