Troubleshooting issues in Proactive problem management
Issues with the job state
Issue symptom
The job stays in pending state and never starts.
Issue scope
One or more problem coordinators face this issue
Resolution
- Navigate to
> Proactive Problem Management > Settings. - Check the Max # of jobs per user count.
- Verify if jobs are already running for the tenant and if the maximum count of jobs has been reached.
New jobs are run only after the queued jobs are complete.
Issues with the job results
Issue symptom
A job is completed but see the clusters are not displayed on the dashboard.
Issue scope
One or more problem coordinators face this issue.
Resolution
This issue might occur due to the following reasons:
- The incident data is of low quality.
- A large number of clusters of clusters have been created
- The data used for the job is huge.
Use the following steps to resolve this issue:
- Data should be in the proper UTF8-encoded format without any garbage characters.
- Run the job again by providing a specific number of clusters.
- Reduce the job data size by selecting a shorter date range or by applying filters, and run the job again.
Job failure
Issue symptom
A job fails and the following message is displayed on the Jobs table:
Issue scope
One or more problem coordinators face this issue.
Resolution
Verify if the value specified in the Helix-Portal-URL parameter in the Centralized configuration is correct.
Issues with job cluster quality
Issue symptoms
The job quality of one or more jobs is poor.
Issue scope
One or more problem coordinators face this issue.
Resolution
This issue can occur due to the following reasons:
- The data set used by the job is very discrete.
- The problem coordinator has not selected a proper Group By field or has selected a field that is not appropriate.
- The problem coordinator has specified a low cluster count in the job configuration.
Use the following steps to resolve this issue:
- Add proper stop words to clean up data in preprocessing.
- Increase the number of clusters in the job configuration.
- Select an appropriate Group by field.
Issues with permissions
Issue symptoms
On selecting an incident from the dashboard, the following error is shown:
Issue scope
One or more problem coordinators face this issue.
Resolution
Verify if the administrator has provided the following permissions:
- Incident User or Incident Submitter
- Problem User or Problem Submitter or Problem viewer
If the issue persists, contact the BMC SaaS Operations Team.
Cluster IDs being displayed on the Proactive Problem Management heatmap and list view
Issue symptoms
For the jobs that have already been run successfully, cluster IDs are displayed on the Proactive Problem Management heatmap and list view instead of cluster labels.
Issue scope
All users face this issue after the latest patch is applied.
Resolution
To fix this issue, rerun existing jobs, or delete old jobs and create new ones. Once job execution is complete, you can see cluster labels again.