Product Documentation

Configuring Reverse Monitoring for a Service

Oct 19, 2016

A reverse monitor marks a service as DOWN if the probe criteria are satisfied and UP if they are not satisfied. For example, if you want a backup service to receive traffic only when the primary service is DOWN, you can bind a reverse monitor to the secondary service but configure it to probe the primary service.

The NetScaler appliance supports the following reverse monitors:

  • HTTP
  • ICMP
  • TCP (from release 11.1 build 49.x)

Configuring HTTP Reverse Monitoring for a Service

The following table describes the conditions of HTTP direct and reverse monitoring for a service:

Condition

Direct

Reverse

Connection not established.

Fail

Fail

HTTP response code matches the probe's specifications.

Success

Fail

HTTP response code does not match the probe's specifications.

Fail

Success

Probe timed out.

Fail

Fail

To configure HTTP reverse monitoring for a service by using the NetScaler command line

At the command prompt, type:

>add lb monitor <Monitor_Name> HTTP -respCode 200 -httpRequest "HEAD /" -destIP <Primary_Service_IP_Address> -destPort 80 -reverse YES

>bind service <Secondary_Service_Name> -monitorname <Monitor_Name>

Configuring ICMP Reverse Monitoring for a Service

The following table describes the conditions of ICMP direct and reverse monitoring for a service:

Condition

Direct

Reverse

ICMP echo reply is received.

Success

Fail

Probe timed out.

Fail

Success

To configure ICMP reverse monitoring for a service by using the NetScaler command line

At the command prompt, type:

>add lb monitor <Monitor_Name> PING -destIP <Primary_Service_IP_Address> -reverse YES

>bind service <Secondary_Service_Name> -monitorname <Monitor_Name>

Configuring TCP Reverse Monitoring for a Service

If a direct TCP monitor receives a RESET in response to a monitor probe, the service is marked DOWN. However, if a reverse TCP monitor receives a RESET response, the probe is considered successful, and the service is marked UP.

The following table describes the conditions of TCP reverse monitoring for a service:

Condition

Direct

Reverse

TCP connection is established.

Success

Fail

Probe timed out.

Fail

Fail

Response to probe is RESET.

Fail

Success

To configure TCP reverse monitoring for a service by using the NetScaler command line

At the command prompt, type:

add lb monitor <Monitor_Name> TCP –destip <Primary_Service_IP_Address> -destport <primary_service_port> –reverse YES

>bind service <Secondary_Service_Name> -monitorname <Monitor_Name>

 

To configure reverse monitoring by using the NetScaler GUI

  1. Navigate to Traffic Management > Load Balancing > Monitors.
  2. Create an HTTP, ICMP, or TCP monitor and select Reverse.