Service Health

When monitoring the performance of their mission-critical business services, service managers may want to know what percentage of service components are currently in an abnormal state. This knowledge will not only enable them to quickly identify the services that are experiencing performance degradations, but will also help them swiftly assess the extent of the damage! The optional service health page of the eG monitoring console imparts this knowledge to service managers. By default, this page is not available in the eG monitoring console. To enable this page, do the following:

  • Edit the eg_ui.ini file (in the <eg_install_dir>\manager\config directory).
  • Set the enable_service_health parameter in the file to true. By default, this is set to false.
  • Finally, save the file.

Once this is done, then, you would be able to access the service health page by clicking the icon in the eG monitoring console.

This uses a wide variety of chart types, namely - gauge chart, pie chart, HLinear gauge chart, and column chart - to indicate the state of components engaged in the delivery of each managed business service.

By default, the chart that corresponds to a service denotes the following:

  • the percentage of service components that are currently in a Critical state, indicated by the color Red;
  • the collective percentage of components that are in the non-critical states (i.e., Normal, Major,Minor, and Unknown), indicated by the color Green;

This default setting is governed by the show_states parameter in the [SERVICE_HEALTH] section of the

eg_ui.ini file. By default, the show_states parameter is set as follows:


Since the show_states parameter is set to HIGH and GOOD by default, each service chart indicates the percentage of service components in the CRITICAL state and GOOD states (by default) only. While HIGH denotes Critical, the GOOD here is the collective term used to denote all non-critical states, such as Normal, Unknown, Major, and Minor.

You can override this default setting so that the service charts indicate more or less number of 'distinct' states. For instance, to make sure that each of the service charts 'distinctly denote' the percentage of components in the Critical, Major, Minor, Normal, and Unknown states, your show_states parameter setting should be changed as follows:


Also under each service chart, the list of service components in the abnormal state (if any) is displayed. While the chart itself can alert you to any slowdown that is currently experienced by the service, a quick look below the chart will help you instantly identify those service components that could be contributing to the service slowdown. By default, this service component list will display the top-5 abnormal components involved in delivering that service. To override this default setting, do the following:

  • Edit the eg_ui.ini file.
  • Against the show_top parameter in the [SERVICE_HEALTH] section, specify any number of your choice. By default, this is set to 5.
  • Finally, save the file.

Moreover, as mentioned already, the SERVICE HEALTH page uses any or all of the following types of charts to indicate the current state of configured services:

  • Angular Gauge chart
  • HLinear Gauge chart
  • Pie chart
  • Column chart

If you choose to use all the above-mentioned chart types in the SERVICE HEALTH page, then set the show_charts parameter in the [SERVICE_HEALTH] section of the eg_ui.ini file to all, as indicated below:


If the show_charts parameter is set to all, you then need to indicate the chart types to be used and the sequence in which they are to be used. For this, provide a comma-separated list of chart types against the chart_list parameter in the [SERVICE_HEALTH] section of the eg_ui.ini file:


At runtime, the charts will be displayed in the same sequence in which they appear in the comma-separated chart_list above.

Each entry against the chart_list parameter should be of the following format:


For instance, if an Angular Gauge chart is one of the charts you wish to use in the SERVICE HEALTH page, and you want the legend of the chart to appear below the chart, then your chart_list specification will be:


Sometimes, you may not want to use a variety of charts in the SERVICE HEALTH page, and instead use a single chart type across the entire interface. In this case, you will need to alter the show_charts parameter as indicated below:


For instance, if you want the health of all services to be depicted using only pie charts, and you want the legend to be displayed to the right of every pie chart, your show_charts specification will be:


Clicking on the service chart of a non-web-site service in the SERVICE HEALTH page here will lead you to the Topology tab page of the Service Dashboard, where you can view the state of all components engaged in the delivery of that service and the inter-dependencies they share. Clicking on the service chart of a web site service on the other hand, will list the transactions configured for that web site, the current state of each transaction, and the metrics that are collected for every transaction.