English
Menu

What is Problem management?

The process whose goal is to identify the root cause (the problem) behind one or more incidents and prevent recurrence with a permanent fix. It includes reactive work, investigating causes from incidents that have occurred, and proactive work, analyzing incident trends to prevent failures that have not happened yet. A problem whose root cause or workaround is known but whose fix is not yet complete is recorded as a known error and shared with the service desk and others. When a fix changes a system, it goes through change management.

Japanese name: 問題管理

Problem management appears in 4 practice questions across 2 topics, including IT Service Management and Systems Audit, Software Development and Project Management.

Field: IT Service Management and Systems Audit

Problems and known errors

In problem management, the not-yet-understood root cause behind incidents is called a problem. A problem is registered and investigated when incidents with the same symptoms keep recurring or when an incident with a large impact occurs. A problem whose root cause has been identified, or for which a workaround has been found, but whose permanent fix is not yet complete is called a known error and is recorded together with the workaround. The service desk uses these records to respond quickly to the same incident.

How causes are investigated

To find the root cause, incident records, logs and configuration management information (what changed just before) are compared, and events are put in time order. Methods include repeatedly asking "why" to trace the cause, listing candidate causes with a cause-and-effect diagram, and narrowing the targets with a Pareto chart that orders items by count. Once the cause is known, a permanent fix is considered, and if the system must be changed, a change request is submitted to change management.

Reactive and proactive work

Investigating causes in response to incidents that have already happened is reactive problem management. Proactive problem management, by contrast, analyzes trends in incident records and monitoring data to find signs before they become major failures and acts first. For example, noticing that disk replacements for one model are increasing and replacing them in a planned way is proactive work.

Try it: questions about Problem management

Which of the following describes a "known error" in problem management appropriately?

  1. AA problem whose root cause has been identified or for which a workaround is known, but whose permanent fix has not yet been completed
  2. BAn incident caused by a user's operating mistake that was resolved by explaining the correct operation to the user
  3. CA defect found during development that was fixed and tested before release and does not remain in the production environment
Show answer and explanation

Answer: A. A problem whose root cause has been identified or for which a workaround is known, but whose permanent fix has not yet been completed

In problem management, the not-yet-understood root cause behind one or more incidents is treated as a "problem". A problem whose root cause has been identified through investigation, or for which a workaround has been found, but whose permanent fix has not yet been completed is called a "known error", and it is recorded together with the workaround and shared with the service desk and others. This allows the workaround to be used immediately when the same incident occurs.

A file server stops responding with the same error several times a week, and each time an operations staff member restarts it to restore service. From the standpoint of problem management, which of the following is an appropriate action?

  1. ASince a restart restores service, do not treat it as a problem and keep restarting the server each time
  2. BWithout knowing the cause, keep changing the settings of the production server one after another at the staff member's discretion to see what happens
  3. CRegister the series of incidents as a problem, investigate the root cause, and carry out the permanent fix through change management
Show answer and explanation

Answer: C. Register the series of incidents as a problem, investigate the root cause, and carry out the permanent fix through change management

When the same incident keeps recurring, a root cause probably remains, so it is registered as a problem and investigated. Until the root cause is found, the restart is recorded as a workaround and shared as a known error. When a permanent fix changes the server's settings or programs, its impact and risk are assessed in change management and it is approved before being carried out.

FAQ

If a workaround keeps the business running, is problem management unnecessary?

No. A workaround is a temporary measure; as long as the root cause remains, the same incident keeps recurring, each time costing effort and affecting the business. The workaround is recorded as a known error, and the cost and benefit of a permanent fix are compared to decide whether to carry it out.

Can a fix found in problem management be applied to production right away?

Changes to the production environment go through change management, where their impact and risk are assessed and they are approved before being carried out. Even a fix for a root cause can cause new failures if it is applied without assessment.

Problem management in the explanations

Questions that mention Problem management (4)

  1. Which of the following describes a "known error" in problem management appropriately?
  2. A file server stops responding with the same error several times a week, and each time an operations staff member restarts it to restore service. From the standpoint of problem management, which of the following is an appropriate action?
  3. In IT service management, which process mainly aims to investigate the root cause of recurring failures and prevent them from happening again through permanent fixes?
  4. The service desk received a call from a user saying, "In the expense reimbursement system, the submit button does not respond when I attach an image of a receipt." The same symptom is registered as a known error, with the recorded workaround "the image can be submitted if it is first registered from a separate registration screen." Which of the following is the most appropriate action in incident management?

Books on IT Service Management and Systems Audit Ad

As an Amazon Associate I earn from qualifying purchases.

Related terms

Change management Service desk Incident management Audit trail BCP (business continuity plan) CAAT (computer-assisted audit techniques) Configuration management Escalation Internal control IT governance

Back to the glossary