Monitoring a process

You can define a data source that monitors a process or several processes which run on a server. The processes must run on the same host as the agent. For every process, the data source adds a row to the Availability data set.

Procedure

  1. On the Agent Initial Data Source page or the Data Source Location page, click A process in the Monitoring Data Categories area.
  2. In the Data Sources area, click A process.
  3. Click Next.
  4. On the Process Monitor page, in the Process information area, provide the display name and process name. You can type the process name manually or obtain it by clicking Browse. Clicking Browse shows a list of processes that are currently running on the local system or on a remote system.

    You can further discriminate processes by selecting the Use argument match and Match full command line options. For example, if multiple instances of the same processes are running on the system, one instance can be distinguished from another by using these options.

    Table 1. Fields on the Process Monitor page. A table listing the fields in the Process Monitor page and their descriptions
    Field name Description Acceptable values
    Display name Descriptive name for the component of the application that is implemented by the process as it is shown in the Tivoli® Enterprise Portal or in the IBM® Cloud Application Performance Management console Descriptive string
    Process name Name of the process that is being monitored Valid executable file name
    Use argument match Select if you want to match on the process arguments. On or Off
    Argument Argument string on which to match. Argument matching looks for the provided string as a substring of the arguments. Matching is successful if you provide any part of the arguments as the input string. String
    Match full command line Specify the entire name of the executable file that might include the path On or Off
    Command line Matches the provided string against the fully qualified command name that is used to start the process. Command arguments are not included. Fully qualified means the path to the command must be included. String
    Operating systems Select the operating systems on which this process runs Any selection
  5. If you click Browse, the Process Browser window opens. This window initially contains detailed information about each process on the Agent Builder system. The information includes the ID, the process name, and the full command line for the process. Select one or more processes or work with the list in the Process Browser window by using one or more of the following actions:
    1. To sort the list of processes, click the column heading.
    2. To refresh the information in the window, click the Refresh (lightening bolt) icon.
    3. To search for specific processes, click the Search (binoculars) icon.
      You can enter a search phrase and select options section to search by process identifier, name, and command line.
    4. To view processes on a different system, select a previously defined system from the Connection Name list. Or click Add to enter the system information for a new system.
      For more information, see Defining connections for process browsing. You can load processes from more than one system at a time, and switch between connections while processes are loading for one or more connections.
      Note: When you browse remote systems, the command-line details are available only when you browse through a Tivoli Enterprise Portal Server.
    In the following example, after you select svchost.exe, it is shown in the Process name field on the Process Monitor page (Figure 1).
    Figure 1. Process Monitor page example
    Process Monitor page with svchost entered as the display name and svchost.exe entered as the process name
  6. Complete the Process Monitor page by using the information in (Table 1).
    Note: If the process you described in this monitor is applicable to only some of the operating systems that your application runs on, you might want to create one or more process monitors with the same display name to cover the other operating systems. Add the process monitors one at a time. Ensure that the display name is the same for each monitor, but that the process name can be found on the operating systems that are selected.
  7. Do one of the following steps:
    • If you are using the Agent wizard, click Next.
    • Click Finish to save the data source and open the Agent Editor.

What to do next

If you want to use the data from this data source in the summary dashboard for IBM Cloud Application Performance Management, you must create a filtered data set (attribute group) based on the Availability data set and configure it as providing a single row. Use the NAME field to select the row for your process.

You can use the Status field for status; DOWN means that the process is not running, while UP means it is running. In the new filtered attribute group, select the Status field and specify the severity values for it.

If several copies of the process are running, several rows with this process name are present in the Availability data set, and all of then include the UP status. Your filtered data set must be configured to return one row, so any of these rows might be returned, but the Status value is valid in any case.