Configure Intelligent Management properties for web servers

Configure Intelligent Management properties for web servers

After enabling Intelligent Management, we can specify new properties for our web server. The new properties display in the generated plugin-cfg.xml configuration file as <Property name="aaa" value="bbb"/> under the IntelligentManagement element. Use the new properties to configure or modify the Intelligent Management capabilities of the web server.
The following advanced custom properties are available by default. Do not modify these properties.

Name Value Description
DefaultRequestTraceSpec Any valid trace specification. For example: http.request:DEBUG
Default value: NOTICE
Trace individual requests, if none of the per-request conditional trace rules matches those requests. If this custom property is set, every request that does not match a trace rule condition is traced with this specification. If this custom property is not set, the requests are traced based on the overall control or HTTP specification.
MaxElapsedTimePerRequestInSeconds Seconds
Default value: There is no time limit
Amount of time in seconds Intelligent Management will try to find an end point for a request once it arrives. After this specified time, it will stop finding an end point and return.
MaxHttpContextPoolSize Maximum number of HTTP request contexts to be pooled or cached.
Default value: -1 (unlimited)
Maximum number of HTTP request contexts, which are pooled or cached. If the value is -1, there is no limit; otherwise, the pool size is limited. When the number of concurrent requests for a single Apache process exceeds this limit, contexts are disposed off rather than returned to the pool. Large amounts of pooled HTTP requests limits the amount of memory used, and decreases performance during a burst of concurrent requests.
MaxSecondsToWaitDuringAtomicRollout Seconds
Default value: 3 seconds
Maximum number of seconds that a single request can wait when it routs to an application during an atomic rollout. By allowing a request to wait, it returns successfully rather than returning with a service not available error response (for example, HTTP 503 error code).

Define an Intelligent Management property

In the administrative console, click...

Servers > Server types > Web servers > web_server > Intelligent Management > Intelligent Management plug-in properties > New

On the Intelligent Management properties configuration page, specify the plug-in properties. Enter an arbitrary name and value. The name is a property key. The value is a string value to be used for setting the internal system configuration properties. Click Apply, and the web server uses the new Intelligent Management property. Click OK to save your input.

Modify an Intelligent Management property

In the administrative console, click...

Servers > Server types > Web servers > web_server > Intelligent Management > Intelligent Management plug-in properties > property_name

On the Intelligent Management properties configuration page, modify the plug-in properties. Click Apply, and the web server uses the modified Intelligent Management property.

Configure Intelligent Management for web servers
Configure trace on the WebSphere plug-in

Name	Value	Description
DefaultRequestTraceSpec	Any valid trace specification. For example: http.request:DEBUG Default value: NOTICE	Trace individual requests, if none of the per-request conditional trace rules matches those requests. If this custom property is set, every request that does not match a trace rule condition is traced with this specification. If this custom property is not set, the requests are traced based on the overall control or HTTP specification.
MaxElapsedTimePerRequestInSeconds	Seconds Default value: There is no time limit	Amount of time in seconds Intelligent Management will try to find an end point for a request once it arrives. After this specified time, it will stop finding an end point and return.
MaxHttpContextPoolSize	Maximum number of HTTP request contexts to be pooled or cached. Default value: -1 (unlimited)	Maximum number of HTTP request contexts, which are pooled or cached. If the value is -1, there is no limit; otherwise, the pool size is limited. When the number of concurrent requests for a single Apache process exceeds this limit, contexts are disposed off rather than returned to the pool. Large amounts of pooled HTTP requests limits the amount of memory used, and decreases performance during a burst of concurrent requests.
MaxSecondsToWaitDuringAtomicRollout	Seconds Default value: 3 seconds	Maximum number of seconds that a single request can wait when it routs to an application during an atomic rollout. By allowing a request to wait, it returns successfully rather than returning with a service not available error response (for example, HTTP 503 error code).