SAP Web Dispatcher Top Consumers Test

During peak business events such as month-end processing, user traffic through the SAP Web Dispatcher can increase steadily. As backend systems become overloaded, response times may degrade significantly. This slowdown can in turn impact the SAP Web Dispatcher, causing it to refuse new requests from hosts connecting to it. The resulting chain of events may lead to visible errors, failed user logins, and periods of service unavailability. Administrators must therefore closely monitor rising load and system stress to proactively mitigate issues and minimize user impact so that business operations remain unaffected. The SAP Web Dispatcher Top Consumers test helps administrators in this regard!

For each host connecting to the target SAP Web Dispatcher, this test reports the number of active connections to the target SAP Web Dispatcher. Using this test, administrators can figure out the rate at which requests were triggered with warning conditions as well as the rate at which the requests were rejected for each host. Using these metrics, administrators can identify the host from which maximum requests were rejected by the SAP Web Dispatcher. This also helps administrators proactively detect and prevent potential outages before business operations are disrupted.

Target of the test : A SAP Web Dispatcher

Agent deploying the test : An internal/remote agent

Outputs of the test : One set of results for each host connecting the target SAP Web Dispatcher being monitored.

Configurable parameters for the test
Parameter	Description
Test period	How often should the test be executed
Host	Specify the host for which the test is to be configured.
Port	The port number at which the specified host listens. The default is 2375.
Web Disp URL	By default, the eG agent connects to the SAP Web Dispatcher Admin Login URL to collect the required metrics by scraping the URL. Therefore, specify the full path of the SAP Web Dispatcher Admin Login URL against this text box. For example, your specification can be: https://<SAP Web Dispatcher Host>:<SAP Web Dispatcher Port>/sap/wdisp/admin/public/default.htm.
Separate Login Page	A separate login page may appear in your environment if the target SAP Web Dispatcher cannot authenticate you via HTTP authentication or if HTML FORM-based authentication is enabled or session-based authentication is enforced. By default, this flag is set to Yes indicating that a separate login page appears to access the target SAP Web Dispatcher. If your environment invokes a pop up window for login in order to access the target SAP Web Dispatcher, then, set this flag to No.
Web Disp Username, Web Disp Password and Confirm Password	To enable the test to access the SAP Web Dispatcher Admin Login URL and collect the required metrics, you need to configure the Web Disp Username and Web Disp Password parameters of the test with the credentials (i.e., name and password) of a user with monitor privilege. To know more on how to create a new user, refer to Creating a New User with Monitor Privilege. Confirm the Web Disp Password by retyping it in the Confirm Password text box.

Measurements made by the test
Measurement	Description	Measurement Unit	Interpretation
Active connections	Indicates the number of open TCP / HTTP connections that are currently handled by this host.	Number	This measure is a good indicator of connection load on the target SAP Web Dispatcher. A sudden spike in the value of this measure may be due to traffic bursts or slow backend which may require the intervention of administrators.
Warning rate	Indicates the rate at which requests triggered ICM warning conditions and not errors through this host.	Warnings/min	A high value for this measure is a cause of concern as this may be an early indicator of performance degradation.
Reject rate	Indicates the rate at which requests from this host were rejected by the SAP Web Dispatcher before reaching backend.	Rejects/min	Ideally, the value of this measure should be zero. Requests may often be rejected when: maximum connection limit is reached queue is full there are no free worker threads backend is not reachable timeout is exceeded A sudden/gradual increase in the value of this measure is an indication for the administrators to initiate troubleshooting at the earliest.