Incident vs. Problem vs. Change: The Twelve ITIL Processes, ISO/IEC 20000-1, and How CCSP Tells Them Apart

Chris Rees
25+ years in IT · Pluralsight author, 4.6/5 across 2,000+ ratings

The bug that took a grocery chain's checkout offline three Fridays in a row was fixed three times. Each incident was closed inside forty minutes, so the incident metrics looked excellent, and nobody investigated why it kept happening because no step in the process required anyone to. That story is CCSP objective 5.3 in one paragraph. The outline lists twelve ITIL-derived management processes, and the exam does not ask you to run a service desk. It asks you to tell the processes apart, because in ordinary speech they overlap completely and in definition they do not overlap at all.
Twelve processes, one job each
Each process has one goal, and the goal is the test. When a stem describes an activity, ask which goal it serves and the process names itself.
| Process | Its one job | The tested distinction |
|---|---|---|
| Incident management | restore normal service as fast as possible | a workaround is a legitimate resolution |
| Problem management | find and remove the underlying cause | recurrence is the signal |
| Change management | authorize, assess and control modifications | it authorizes; it builds and moves nothing |
| Release management | plan and package what goes out | what goes out, when, grouped how |
| Deployment management | move the release into the environment | deployed is not the same as enabled |
| Configuration management | record configuration items and their relationships | the CMDB; change assesses impact against it |
| Service level management | agree, monitor and report service targets | the agreement, not the engineering |
| Availability management | engineer and measure the agreed availability | the engineering, not the agreement |
| Capacity management | meet current and future demand at justifiable cost | forward-looking |
| Continuity management | restore service after major disruption | disaster, not daily availability |
| Information security management | ensure confidentiality, integrity and availability across all of it | sets the requirements the other eleven satisfy |
| Continual service improvement | measure and improve services and processes | the loop that keeps the rest honest |
Hold it as four groups rather than twelve items. Five agree and design the service; four move changes into it safely; two keep it running when something breaks; one improves all of the others.
Incident restores, problem eliminates
This is the most reliably tested distinction in the objective. Incident management has one goal: restore normal service as quickly as possible, by any legitimate means. A workaround, a restart, a failover, a rollback, all count. The incident closes when the service is back, whether or not anyone understands why it failed.
Problem management has a different goal: identify and remove the underlying cause so the incident class stops occurring. It records the known error and its workaround, owns the investigation, and does not close until the cause is gone or a decision has been made to live with it. It can be reactive, triggered by incidents, or proactive, hunting for weaknesses before they produce one.
The exam signal is repetition. A stem that describes the same symptom returning, or a team that keeps restoring service without anyone asking why, is a problem management question. The grocery chain's fix was structural: an incident that recurs raises a problem, and a problem has an owner. The incident process had worked perfectly all along; the missing piece was the handoff.
Change authorizes, release packages, deployment moves
The second pair is really a trio, and the verbs separate them. Change management authorizes: it weighs risk against benefit and approves or refuses a modification. It builds nothing and moves nothing. Standard changes are pre-approved and low risk, normal changes go to a review board, and emergency changes are expedited and reviewed afterward.
Release management decides what goes out, when, and how it is grouped. Deployment management performs the movement into the target environment. The distinction that trips people is that a release can be deployed but not yet enabled: with a feature flag the code is in production, customers cannot see it, and turning it on is the release decision.
Configuration management sits underneath all three. It records configuration items and the relationships between them in a CMDB, and change management uses it to assess impact: what depends on the thing we are about to modify? A stem about a change that broke a system nobody knew was connected is a configuration management failure, not a change management one, because the approval was made against an incomplete map.
Agreeing, engineering, and surviving availability
Three processes all sound like "keep it up", and the exam separates them by what they produce. Service level management produces the agreement and the report: it negotiates targets and monitors against them, distinguishing the SLA with the customer from the operational level agreement between internal teams and the underpinning contract with a supplier. A customer-facing SLA cannot promise more than the underpinning contracts deliver. Availability management produces the engineering: redundancy, failover, monitoring, and measurement of whether the agreed number is met. Continuity management produces the recovery after a major disruption, within agreed timescales, which is where the RTO and RPO live. A failed disk is availability; a lost region is continuity; the number the customer was promised is service level.
Capacity management is forward-looking: meet current and future demand at justifiable cost. In cloud nobody waits for procurement, but quotas, service limits, reservations and spend all run out, and a stem about an autoscaler that hit an account limit at peak is a capacity question.
ISO/IEC 20000-1, ITIL, and the frameworks beside them
The objective opens by naming frameworks, and two pairings decide most of those questions. ITIL is a body of guidance describing service management practices; an organization adopts and adapts it, and nobody is certified against it. ISO/IEC 20000-1 is the certifiable standard for a service management system, aligned with ITIL practice. Remember the pairing: 27001 certifies an information security management system, 20000-1 certifies a service management system, and information security management is the process where 27001 attaches to the twelve.
| Framework | What it is | The stem that points to it |
|---|---|---|
| NIST SP 800-53 | a comprehensive security and privacy control catalog | control selection; US federal and contractor compliance |
| ISO/IEC 27001 | certifiable ISMS standard, with 27002 for control guidance | demonstrating a managed security program to customers |
| ISO/IEC 20000-1 | certifiable service management system standard | certifying that service management is systematically run |
| COBIT | IT governance and management framework | board-level oversight; separating governance from management |
| COSO | internal control and enterprise risk framework | internal control over financial reporting; Sarbanes-Oxley |
| CIS Controls | prioritized defensive actions in implementation groups | "what should we do first" |
| HIPAA | US health data legislation with security, privacy and breach rules | a legal obligation flowing to providers as business associates |
COBIT and COSO are the pair candidates swap. COBIT's signature contribution is separating governance (evaluate, direct, monitor, the board's job) from management (plan, build, run, monitor). COSO is not an IT framework at all; it is the internal control framework auditors reference for financial reporting. Board oversight of IT points to COBIT; controls over the numbers in the annual report point to COSO.
What every process turns into in the cloud
The twelve were written for data centers and change advisory boards that meet weekly. The exam expects you to know how each survives an estate that deploys hundreds of times a day.
Configuration management becomes continuous and API-driven, because a manually maintained CMDB is stale on arrival; infrastructure as code becomes the definitive record and drift is treated as an incident. Incident management converges with security incident handling, because a compromise and an outage look identical for the first hour: one intake, one severity scale, and an explicit escalation into the security process. Problem management becomes the blameless post-incident review, and in cloud the cause is frequently a missing guardrail rather than a defect, so the fix is a policy. Service level management runs in both directions: you manage the provider's SLA as well as offering your own, and provider credits never compensate business loss. Continuity management becomes multi-region and multi-provider planning that depends on exit capability rather than a second data center, and the forensic readiness objective 5.4 demands is decided here, before the incident.
Four stems, decoded
- The same database deadlock takes an ordering service offline every month, and each time the on-call engineer restarts it and closes the ticket. What is missing? Problem management. Incidents are being restored correctly; nobody owns removing the cause.
- A team ships code to production on Tuesday but exposes the feature to customers on Friday. Which process governs the Friday decision? Release management. Deployment moved the code; release decides when it goes live, and a feature flag is the mechanism.
- An approved change to a storage service broke a reporting job nobody knew depended on it. Which process failed? Configuration management. The CMDB did not hold the dependency, so the impact assessment was made against an incomplete map.
- An organization wants a certification proving its IT service management is systematically operated. Which standard? ISO/IEC 20000-1. ITIL is guidance and cannot be certified against; 27001 certifies the security management system.
Key takeaways
- Incident restores; problem eliminates. A workaround closes an incident and leaves the problem open; repetition in a stem is the problem management signal.
- Change authorizes, release packages, deployment moves, configuration records. Deployed is not enabled, and the CMDB is what change assesses impact against.
- Service level agrees, availability engineers, continuity recovers, capacity anticipates. A failed disk, a lost region and a promised number are three different processes.
- ITIL is guidance; ISO/IEC 20000-1 is the certifiable standard. 27001 for security, 20000-1 for service management; COBIT for IT governance, COSO for financial reporting control.
- In cloud the processes move rather than disappear. The pipeline becomes the change control, infrastructure as code becomes the CMDB, and security and service incidents share one intake.
Operational controls and standards are objective 5.3 of CCSP Domain 5, Cloud Security Operations, 17% of the exam, the management layer between the infrastructure controls of 5.1 and 5.2 and the forensic and communication duties of 5.4 and 5.5. Work all 42 Domain 5 topics with our CCSP Domain 5 study guide.
#CCSP #ISC2 #CloudSecurity #ITIL #IncidentManagement #ProblemManagement #ChangeManagement #ISO20000 #InfoSec #CyberSecurity
Run it, and investigate it
Digital forensics and communication are two of six objectives in CCSP Domain 5, 17% of the exam. Work all 42 topics, from building and operating cloud infrastructure through operational controls, forensics, communication with relevant parties and managing security operations, with 60 practice questions.
- Find your gapsA free readiness quiz scores you domain by domain.
- Study what's on the testWork through every objective, with practice questions.
- Book it with confidenceSchedule the exam when you're scoring well in every domain.
Read a free excerpt from CCSP Domain 5: Cloud Security Operations
Keep reading
WAF, API Gateway, DAM, CASB: What Each Cloud Application Component Can See, and the IAM Stack CCSP Tests
CCSP Domain 4 placement questions all hide the same test: what can this component actually inspect, and what does this identity control actually prove? Objectives 4.6 and 4.7 as a request path and an identity stack: WAF, API gateway, XML firewall, load balancer and database activity monitoring by what they see; cryptography and sandboxing inside the application; federation versus SSO, phishing-resistant MFA, the three CASB deployment modes, and why the best secret is the one that does not exist.
Read CCSP deep divesWhose Law Applies? Jurisdiction, Blocking Statutes, Transfer Mechanisms, and eDiscovery in the Cloud
Cloud data has at least three addresses: where it is stored, where the provider is incorporated, and where the person it describes lives. CCSP Domain 6 on what happens when their laws disagree: conflicting jurisdiction and lawful access demands, blocking statutes, the mechanisms that make cross-border transfer legal, controller and processor duties and the privacy impact assessment, and eDiscovery and legal hold when the evidence is spread across a provider's regions.
Read CCSP deep divesSnapshot, Hash, Isolate: Digital Forensics When the Server Is Someone Else's
Every forensic rule you learned assumes you can pull the plug. In the cloud you cannot, and the disk belongs to someone else. CCSP Domain 5 on investigating on infrastructure you do not own: forensic readiness before the incident, isolate versus terminate, acquisition by snapshot and hash, chain of custody across a provider, what the provider will and will not give you, and the communication clock that starts the moment personal data is involved.
ReadFollow Skillthropic
Keep learning between study sessions
The same exam-focused teaching, in smaller pieces, wherever you scroll.
Enjoyed this? Get the AI security news that matters.
Join The AI Security Brief for the top AI security news, plus what's important to the C-suite. Free, straight to your inbox.
No spam. Unsubscribe anytime.