Configure > Networking > Health Check URLs

Force Health Checks to Report Proxy Down
Enable/Disable	When enabled, all health check URLs sent to this proxy report WSDOWN. The URL response will be similar to: HTTP/1.0 503 Service Unavailable Server: Content Gateway Manager 7.7.0 Date: Thu, 26 Jul 2012 20:26:14 GMT Cache-Control: no-store Pragma: no-cache Content-type: text/plain Content-length: 6 WSDOWN
Health Check URLs	The load balancer should consider the service down if the URL request fails for the following reasons: No TCP connection -- proxy down Response too slow -- proxy deadlocked or not responsive Invalid response
http://[Content Gateway IP address]: 8083/health.basic	Checks connectivity with Content Gateway and responds with WSUP or WSDOWN.
http://[Content Gateway IP address]: 8083/health.app.filtering	Checks the health of Filtering Service responses to Content Gateway requests and reports WSUP or WSDOWN.
http://[Content Gateway IP address]: 8083/health.load	If the health.basic URL reports WSDOWN, this URL also reports WSDOWN. Otherwise, health.load returns: CPU usage (operating system load average) Connection usage (number of open connections) Bandwidth usage How these values are calculated and how they can be customized is described below. The default response will look similar to: HTTP/1.0 200 OK Server: Content Gateway Manager 7.7.0 Date: Thu, 26 Jul 2012 20:26:14 GMT Cache-Control: no-store Pragma: no-cache Content-type: text/plain Content-length: xx Load=2253 Conns=5150 Mbps=6.42

A format file, /opt/WCG/config/health.load.template, allows for customization of the response format.

Format specifiers are:

%L = Load (integer)

%C = Connections integer)

%B = Bandwidth in Mbps (double)

%% = %

The default health.load.template file is:

Load=%L Conns=%C Mbps=%B

Here is health.load.template modified to respond with an xml-like format:

<load>

</load>