Run Discovery Manually for a Service Model

Run Discovery Manually for a Service Model

By default, service discovery jobs run automatically once each day. If necessary, you can use the following procedure to run a discovery job manually.

1. In the Service Modeling dashboard, click an existing service model.
2. Click .

3. Check and edit the entry point before running the discovery process:

   Click the Entry Point value to enter the Edit Entry Point mode, where you can add or edit the URL that is the starting point of the service.

   Once you have provided the entry point URL, you can click Check URL Availability to validate the URL.

   • If the URL is available, URL Online is displayed with a success indicator
   • If the URL is not available, URL Offline is displayed with an error indicator .

   You can click URL Online or URL Offline again to check the current status of the entry point URL.

   To exit the Edit Entry Point mode, simply click .

4. Click .

   Indicators for the discovery progress of each CI are displayed on the topology map.

   While the discovery process is running, the status Running Discovery is displayed, and the number of discovery issues found is updating. You can click to deactivate the currently running process, or click to restart the process.

   After the discovery process completes, you can see a message indicating the discovery process has finished successfully or finished with errors.

   In case of errors, you can click Discovery Issues to open the Discovery Issues pane, which lists errors grouped by the types of discovery issues . For details, see Discovery Issues.

   When you click a discovery issues type a pane opens and lists all the errors within the selected category.

   To see in the Discovery Issues pane only the relevant discovery issues for a specific CI and for its trigger CIs, simply click the CI's Discovery indicator.

   

5. After you have provided the required information, click to rerun discovery.

Discovery Issues

The Discovery Issues pane lists errors grouped by the types of discovery issues . The following types of discovery issues may occur during discovery:

• Internal Errors
• Connection Errors

• Credential-Related Errors

  For credential-related errors, you can provide new credentials for the selected protocol, and you can save multiple protocols. When specifying credentials, you can specify the network scope for those credentials:

  • IPv4 (either as IP range or as Classless Inter-Domain Routing (CIDR)
  • IPv6

  Note The newly provided credentials takes effect only after the discovery process has been restarted.

• Timeout Errors
• Unexpected/Invalid Behaviors
• Information Retrieval Errors
• Resources-Related Errors
• Parsing Errors
• Encoding Errors
• SQL Related Errors
• HTTP Related Errors
• Specific Application Errors

• Probe Errors

  For probe errors, you can see a pause indicator and meaningful error messages in the following situations:

  Clicking Probe Errors in the Discovery Issues pane opens the Probe Errors pane, where you can click each of the probe errors to display detailed meaningful error messages.

  Detailed error messages are available in the following situations:

  • When the discovery job cannot run because the probe is suspended or disconnected, or there is no setup for the probe.

  • When the discovery job cannot run because the probe or the activity is in a blackout period.

  • When the discovery job cannot run because the IP is not in the discovery IP ranges.

    To fix this probe error, you can create a new IP range to contain the problematic IP address. To do so, click the probe error entry and then click Add IP Range . For details, see Add an IP Range Manually.

• Software Related Errors

  For Software Related Errors, you can see the ignored TCP connections of a running software and you can add the missing ports in the portNumberToPortName.xml file. See Add a missing port manually.

• Others

  For all other types of errors, you can click the link to send a message to an administrator for assistance.

  Note In some cases, if the message is too long, there might be an HTML limitation or an Email client limitation, and the message may not be displayed entirely in the email.

Add a missing port manually

1. In the Discovery Issues pane, click Software Related Errors, and then select an error entry with the following Problem Source message: "Some outgoing ports are filtered".

2. Click the Application Signature button.
3. Enter an Application Name.
4. Select all the ports you want to add in the portNumberToPortName.xml file.

5. Click Save Mapping.

   Note  

   • You can see the newly added ports in the UCMDB UI > Data Flow Management > Adapter Management > portNumberToPortName.xml file.
   • You can also click the link to send a message to an administrator for assistance.
6. After adding the missing ports, click to rerun discovery.