Tài liệu ITIL V2 : Service Support doc
Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (250.8 KB, 31 trang )
ITIL V2 : Service Support
Summary
Author: Pieter Dubois
mail address:
Total Pages: 30
Document Version: 1.2
June, 2003
IGS non-Confidential
Document Control
Distribution
Pieter Dubois
ITIL interest groups
Revision History
Date of this revision: 16/06/2003
Date of next revision: TBD
Revision Number Revision Date
Summary of Changes
Changes marked
(1) 20/11/2002 Version 1.0
1.1 December 2002 slight adaptations
1.2 June 2003 prepare for publication
References:
Service Support (CCTA) - ISBN 0 11 330015 8
IBM Belgium
Page 2 of 31
IGS non-Confidential
!"#$$%
!& %
!'( %
!!)
!*+ )
!&# )
!",
*& -
*-
*. -
*!/ -
**+
*&
*"$
*"$
*0+1
*%"2"$
*%&
*%&
'( *
*
. *
!/ *
*+
$
'($
'(
&'(
!3'(
%"$
)'($
)
' )
)
' )
. )
!/ )
*$&21,
+,
"$,
%"2"$ ,
%,
%4#$$5-
%!$$-
%+
%$
%
%$
%!'
%.
%+&
%6&
%!++
%!/
IBM Belgium
Page 3 of 31
IGS non-Confidential
%*+
%+$!
%"$!
%%" !
%%+!
%%+$$!
)&#*
)7$ *
) *
)$ *
)!' *
).
)!/
)*&2+
)+2$
)"$
)%"2"$ %
)%&#%
)%&#./%
,")
,7$2' )
, )
,$ )
,!' )
,. ,
,!/ ,
,*+,
,+2$!-
,& !-
,%"$ !-
,%0+1!-
,%"$!
,)"2$!
IBM Belgium
Page 4 of 31
IGS non-Confidential
1. Overview
1.1 Management Summary
This document provides a summary of the most relevant information of the ITIL Service
Support V2 book to help with preparation of the Exin certificate in ITIL Service Management.
The following topics for all processes got particular attention:
• What is ITIL
• Mission statement different ITIL processes
• Activities of each ITIL process
• Benefits/difficulties/costs implementing ITIL process
• Links dependencies with other ITIL processes
• Information flow with other ITIL processes
• ITIL process roles and responsibilities
• Order of implementation within ITIL process
• Key Performance Indicators of and reports about the ITIL Process
IBM Belgium
Page 5 of 31
IGS non-Confidential
2 Introduction to ITIL
2.1 ITIL is
• A Public Domain Framework: the information is publicly available and can be used
by everyone for implementation
• Best practice Framework: the information documents industry best practice
guidance. While not detailing every action that should be done, which is organization
dependant, it provides the goals, general activities, inputs and outputs of the various
processes. It provides a framework in which to place existing methods and activities.
• De facto standard: ITIL by its widespread use became a de facto standard. The
advantage is that there is a common language for people to speak. This makes
education very important to provide all people in the process this common language
• Quality Approach: ITIL focuses on providing high quality services with a focus on
customer relationships. There is a strong relationship between the IT organization
and the Customers.
• Process Based: ITIL is a process based method. It sees IT Service Management
also as a Business Process and tries to organize it that way.
2.2 General Service Management benefits
• Improved quality of service – more reliable business support
• Formalizes the use of procedures so that they are more reliable to follow
• Provides better view of current IT capabilities.
• Provides better information on current IT services provided.
• Better understanding of IT support by the business.
• Better motivated staff through better management of expectations and
responsibilities.
• Enhanced Customer satisfaction as it is clear what service providers know and
deliver.
• More flexibility and adaptability through acknowledgement and formalizing working in
a changing environment.
• Faster cycle times and greater success rates on implementing changes.
IBM Belgium
Page 6 of 31
IGS non-Confidential
3 Relationship between processes
3.1 Configuration Management
Configuration management is an integral part of all other service management processes.
• Change Management: current, accurate information about all components in the IT
infrastructure makes Change management is more effective. It can even be
integrated with Configuration Management. As a minimum logging of all changes
should be done under control of Configuration Management. Also impact
assessment should be done based on data of the CMDB. All change requests
should be entered and updated in the CMDB.
• Incident Management/Problem Management: Information from the CMDB help the
service support group to solve Incidents and Problems faster by understanding the
probably cause of failure on the Configuration Item. The CMDB should link Incident
records and Problem records to the failing CI and the User.
• Release Management: Requires the CMDB to know what CI’s have to be changed
with a new release.
• Service Level Management: uses info from the CMDB to identify CI’s making up a
particular service for which SLA’s are in place. With this information, underpinning
agreements can be set up.
• Financial Management for IT: needs to know which components are used by whom
in order to be able to charge the right departments
• IT Service Continuity and Availability management: needs to know the CI’s to
perform risk analysis and impact analysis on the service and business when a
component fails.
• Service Desk: uses CMDB information on CI’s to help resolve incidents or provide
feedback on the status of a CI to the User.
3.2 Service Desk
Service desk is the SPOC between service provider and Users on a day-to-day basis.
• Incident Management: SD is the focal point in the incident management process. It
its activities are: registering Incidents, informing Users about incidents, and
management of the incident end-to-end.
• Change Management: As changes can cause incidents, and SD is the SPOC for
incidents, it should know about all changes. SD can act as a focal point for Change
requests of Users, and inform Users on progress of changes. SD can be given
delegation to implement Changes to circumvent Incidents within a scope.
• Problem Management: informs the Service Desk on the progress of managing
problems.
• Service Level Management: Any impact on SLA will arrive at the SD. So SD needs
to know what the relevant SLA’s are such that incidents related to services with SLA’s
get appropriate priority.
• IT Service Continuity: If business continuity plans are invoked, SD plays an
important coordination role.
• Configuration Management: good info on CI’s helps the SD managing requests
and incidents.
IBM Belgium
Page 7 of 31
IGS non-Confidential
3.3 Incident Management
• Problem Management: an incident that has a large impact or several incidents
having the same cause can lead to investigation of a problem. Therefore all
Incidents should be recorded and available for investigation by Problem
Management.
• Service Desk: Is the prime owner of the incident management process. It
coordinates all actions required to manage an incident.
• Change Management: Changes can lead to new incidents; therefore a way of
linking incidents to changes is required.
• Configuration Management: the CMDB can help incident management with
identification of the CI in question. Also because of interrogation & reporting reasons
it would be good to link or insert incident records in the CMDB.
• Service Level Management: priorities given to incidents and escalation procedures
should be agreed upon as part of the SLM process and documented in the SLA’s.
3.4 Problem Management
• Incident management: PM has to have access to accurate and comprehensive
recording of Incidents in order to examine the cause of incidents and trends that are
happening.
• Configuration Management: PM uses the CMDB to record/update its problem
records, and to identify CI’s exhibiting Problems
• Availability Management: is strongly connected with PM in order to identify trends
and to instigate remedial action.
• Change Management: Known Errors give rise to Request for Changes which fall
under the responsibility of Change Management
• Service Desk: PM keeps SD up-to-date of ongoing Problem Resolution.
3.5 Change Management
• Configuration Management: is required for change management as it depends on
the accuracy of the CMDB to assess the full impact of a change.
• Release Management: All releases have to go through change management when
implemented in production
• Service Level Management: is provided details of the Change process in the SLA’s
such that the change management procedures are known to the Users. Also the
impact of changes to the SLA’s are important to take into account
• Service Desk: provided details of the changes that happen such that when incidents
occur on a CI due to the changes it can inform the User about the cause.
• Service Level Management: Before changes happen, their impact on SLA’s have to
be assessed.
• Availability: Changes to CI’s should be evaluated against the impact on availability
of the services it provides.
• Capacity Management: before implementing a change, the impact on the capacity
of the CI should be assessed. Also capacity management activities will give raise to
RFC’s to ensure proper capacity is available
IBM Belgium
Page 8 of 31
IGS non-Confidential
• Security Management: before implementing a change, the impact on the security of
the environment should be assessed.
• IT Service Continuity: before implementing a change on a CI, it has to be assessed
what the impact is on the IT Service continuity plans.
• Release Management: the implementation of new releases happens under the
control of change management.
3.6 Release Management
• Change Management: Changes may require new releases of HW & SW. The
preparation of this before it goes in production is part of Release Management, but
bringing it in production happens under control of Change Management
• Configuration Management: Creating a new Release should be in close integration
with Configuration Management, where the state of all CI’s, whether they are in
development/test/validation/production are updated in the CMDB.
• Incident/Problem Management: should be provided with the documentation of the
new Release and procedures of incident remedial should be provided.
• Service Desk: after the roll-out of a new Release, SD should be able to support this
new release in terms of requests from Users, and managing of incidents relevant to
this new release.
• Service Level Management: Implementation of a new release can have impact on
the SLA’s. This has to be assessed and remediate in advance.
• Capacity Management: The impact of a new Release on the capacity demands of
the systems on which it is implemented should be assessed.
• IT Service Continuity: the impact to the business by a Release when a failure
occurs has to be validated by the IT Service Continuity process. Both when during
the rollout as well as during normal operation, should a failure occur, impacting the
usage of this Release.
IBM Belgium
Page 9 of 31
IGS non-Confidential
4 Configuration Management
4.1 Mission Statement
Configuration Management provides IT infrastructure control through the identification,
registration, monitoring and management of:
• All the Configuration Items of the IT infrastructure in scope.
• All configurations, versions and their documentation
• All changes, errors, service level agreements and history of the components in
general.
• Relationships between the different components.
• Exceptions between configuration records and the real infrastructure.
Configuration Management provides a sound basis for other processes like Incident,
Problem, Change and Release management by providing accurate information on all CI’s
and to the SD function.
4.2 Activities
1. Planning: What is the purpose, scope, objectives, policies, procedures,
organizational & technical context for configuration management
2. Identification & registration: selection, identification and labeling of CI’s and the
relationships between them. CI’s can be hardware, software, documentation. Also
Incident, Problem, Change records can be associated with a CI. Important is the
level of detail required. Each CI should have a unique identifier.
3. Control: Only authorized and identifiable CI’s are recorded during their whole
lifecycle time. It ensures updates to CI’s are associated with Change Management
documents.
4. Status accounting: Reporting of current and historical data associated with each CI.
It tracks the state of a CI from one state to another: test, broken, production,
withdrawn.
5. Verification and audit: validate data in the CMDB with the situation in the real
environment.
6. Reporting: Information about the IT infrastructure should be provided to the other
processes.
4.3 Benefits
• Provides accurate information on CI’s: supports all other Service Management
processes
• Controls valuable CI’s: who is responsible for what CI. If a CI is stolen/broke, what
should be the configuration of the replacement.
• Helps with adherence to licenses: As the CMDB contains an Inventory of all
software installed in the environment, it helps with the verification of legality of
installed software during audits
• Helps financial planning: is helped by interrogating the CMDB to know expected
maintenance costs, life expiry dates, replacement costs.
IBM Belgium
Page 10 of 31
IGS non-Confidential
• Changes don’t go undetected: It provides information on all changes in the
environment which allows for investigation on other required changes like licensing
eg.
• Helps contingency planning: The CMDB identifies the CI’s making up a service
which helps to know what is required to restore a certain service.
• Supports Release Management: When rollouts of new hardware or software
happen, information in the CMDB helps to find out what versions should be in the
new release, as well as which CI’s are affected.
• Improves security by controlling versions: Information on versions of CI’s are in
the CMDB so inadvertent changes to other versions becomes more difficult to go
unnoticed.
• Helps with impact analysis of changes: reduces the risks of implementing
Changes in the life environment.
• Provides data on trends to Problem Management: Incidents affecting particular
CI’s help to find weaknesses in the IT infrastructure.
4.4 Problems
• CI’s with too much detail/too little detail: causing too much work or not sufficient
control
• Implementation without design/analysis
• Overambitious schedules: making CM a bottleneck because not enough time to do
CM activities is provided
• No commitment from management: resulting in lack of control
• Perception is too bureaucratic process: resulting in trying to bypass process.
• Bypass process: for reasons of speed or malice.
• Inefficient process: because of too many manual activities
• Unrealistic expectations: CM is not a total solution
• Inflexible: Inability to process new requirements
• Isolation: implemented with no link to other processes. Less effective, reducing
benefits.
• Lack of configuration control: users can order HW/SW and install it without
passing CM.
4.5 Costs
Costs are outweighed by benefits, because CM allows handling large volume of changes
keeping quality. CM control reduces risks of viruses, wrong software, fraud, theft …
• Staff costs developing & running procedures
• HW/SW configuration identification
• HW & SW for the CMDB & DSL
• Users with access to CM system
• Customization of CM tool
• Integration with other Service Management tools
• Training and education
4.6 Reporting
IBM Belgium
Page 11 of 31
IGS non-Confidential
4.6.1 Management Reporting
Supports Service Management activities.
• Configuration audit reports
• Number of registered CI’s, versions, categories reports
• Repository capacity growth
• Change rate of CI’s in CMDB/DSL
• Backlog of CM jobs or delays caused by CM
• CM staffing issues & overtime
• Value of CI’s
• Reports on CI’s per location, business unit, service
4.6.2 Key Performance Indicators
Provides reports on how effective the process is.
• Number of times the configuration is not as authorized
• Incidents/Problems tracked to wrong changes
• RFC’s that failed due to bad data in CMDB, wrong impact assessment
• Cycle time to approve/implement Changes.
• Unused licenses
• Exception reports from audits
4.7 Roles & Responsibilities
4.7.1 Configuration Manager
• Develops & Implements CM policy & standards
• Evaluates existing systems and design/implementation new systems
• Proposes/agrees scope CM
• Organizes communication plan & awareness campaign.
• Creates/manages the CI registration procedures
• Ensure roles & responsibilities are defined in the CM procedures
• Develops CI naming conventions and manages staff to comply with standards
• Manages interfaces with other ITIL processes and other business units
• Plans & executes population, maintenance of CMDB
• Provides reports on management and status issues
• Ensures impact analysis is possible by making sure the links between CI’s are
implemented.
• Ensures Changes update the CMDB
• Performs configuration audits and initiates corrective actions
• Obtain budget for having required infra and staffing taking growth of the CM process
in scope
4.7.2 Configuration librarian
• Controls receipt, identification, storage and removal of all CI’s
• Provide information on status of CI
• Assist CM manager
• Create identification scheme for CM libraries and DSL
• Create libraries to hold CI’s
IBM Belgium
Page 12 of 31
IGS non-Confidential
• Assist in identification of CI’s
• Maintain current status on CI’s
• Archive superceded CI’s
• Hold master copies
• Issues copies for change/review when authorized
• Maintain record on all copies issued
• Produce configuration status reports
• Assist CM audits
IBM Belgium
Page 13 of 31
IGS non-Confidential
5 Service Desk
Function, not process
5.1 Mission Statement
The Service Desk intends:
• To be the focal point for all service requests from the business Users towards the IT
organization (including 3
rd
party providers), and to manage them within Service Level
Agreements
• To record and manage the life cycle of incidents, with an emphasis in rapid
restoration of service with minimal business impact.
• To keep the User up-to-date with the status of his service request.
• To provide management information on relevant topics regarding the service, like:
o Service deficiencies
o Customer training needs
o Resource usage
o Service performance & targets
• To be the window on the level of service offered by the IT organization towards the
business.
5.2 Activities
• Call handling.
• Recording & tracking service requests, including incidents, until closure and
verification with the User.
• Request/Incident classification and initial support, including forwarding to 2
nd
level, within Service Level agreements.
• Keeping customer informed on request status and priority.
• Initiate escalation procedures relative to the SLA.
• Coordinating incident handling until resolution with 2
nd
and 3
rd
line support
• Helping to identify problems by recording all incidents
• Providing management information and recommendations for service
improvement
• Communication of planned changes to the Users.
5.3 Benefits
• Improvement in Customer and business service by- and perception of IT organization
• Faster response times to standard requests
• Customers better informed of progress of requests/incidents/changes
• Reduced Incident resolution times
• Improved Business & management reporting
• 2
nd
& 3
rd
level teams can work more organized, less interrupt driven
• More efficient use of resources and technology
IBM Belgium
Page 14 of 31
IGS non-Confidential
• More proactive and focused approach to service provision when using automated
infrastructure monitoring.
5.4 Problems
• Underestimation of scale of work: implementing a Service Desk is a major Change
• No real management commitment to SD. Abandoning at first sign of difficulty
(Silver Bullet lifecycle).
• Working ‘around’ the SD (or the ‘Helpless Desk).
• Not enough training for SD staff, or wrong skill set (no communication skills)
• No information provided to SD when implementing Changes or Releases,
resulting in less good service to the Business.
• No alignment with business needs
5.5 Implementation considerations
• Business needs & objectives must be clear, as well as Customer requirements.
• Management commitment is obtained and budget & resources is made available.
• Decide on metrics of effectiveness of the Service Desk.
• Investment is made for training Customers, SD staff & support teams.
• Decide on the right Service Desk structure (insourced, outsourced, local, remote,
virtual, customer built, off-the-shelf).
• Get professional support from outside
• Consult Customers and End Users
• Adopt a phased approach, generating quick wins. Communicate them.
• Measure progress.
• Acknowledge it is hard work.
• Advertise and sell the service
5.6 Service Desk type considerations
5.6.1 Local Service Desk
• Service Requests are logged locally
• Difficult to provide support services over multiple locations.
• Duplication of skills & resources can make it expensive
• Requires common operational standards over the different SD’s.
• Local skills should be known to other SD’s.
• Ensure compatibility of infrastructure
• Use common management reporting metrics
• Make exchange of Incidents/Requests between SD possible
• Try to establish a logically shared database
5.6.2 Central Service Desk
• Service Requests are logged centrally
• Reduced operational costs
• Consolidated management overview
• Improved usage of available resources
IBM Belgium
Page 15 of 31
IGS non-Confidential
5.6.3 Virtual Service Desk
• Physical location and associated services of SD are largely immaterial due to
advances in telecommunications.
• Can be accessed from anywhere in the world
• Reduced operational costs
• Consolidated management overview
• Improved usage of available resources
• Restriction is a need of physical local presence of a specialist
• Common processes & procedures should be used
• Network performance should be right sized
• User/Role based authorization views on Requests should be possible
5.7 Reporting
• Incident reports by application/support group/customer
• Service Desk functioning reports
o Telephony pickup times
o Telephony talk times
o % calls answered within x seconds
o time worked on resolving incidents
• Workload reports
• Daily reports on:
o Escalations by support group
o Outstanding incidents
o Service breaches
• Weekly reports on:
o Service availability
o Incidents that occur the most, require the most work to resolve, take the
longest to resolve.
o Related incidents requiring creation of problem records
o Known Errors and RFC’s
o Service breaches
o Customer satisfaction
o Trends
o Workload
• Proactive reports on:
o Planned changes for the following week
o Major incidens/problems/changes of the last week with workarounds/fixes
o Customer incidents which were resolved unsatisfied
o Recent poorly performing infrastructure.
5.8 Service Desk profile
• Customer focused
• Articulate and methodical
• Trained in interpersonal skills
• Understand business objectives
• Teamplayer
• Emphasize with Users
• Act professional
IBM Belgium
Page 16 of 31
IGS non-Confidential
• Accept ownership
• Active listener
IBM Belgium
Page 17 of 31
IGS non-Confidential
6 Incident Management
6.1 Definition and mission statement
6.1.1 Mission statement Incident Management
To restore normal service operation, within Service Level Agreement limits, as quickly as
possible, after an incident has occurred to that service, and minimize the adverse impact on
business operations.
6.1.2 Definitions
1. Incident: is an event, which is not part of the standard operation of a service, and
which causes/may cause interruption/reduction in the quality of that service
2. Impact: measures business criticality of the incident. Extent to which an incident
leads to degradation of SLA. E.g. number of users that suffer
3. Urgency: measure for required speed of solving an incident
4. Priority: measure of the expected effort to solve the incident and the order in which
to solve the incidents. Priority=f(Impact, urgencym).
6.2 Activities
• Incident detection and recording: SD, record basic details (time, who, symptoms,
CI’s affected) and provide incident number
• Classification and initial support: SD, if service request go to SR procedure. Find
Service affected. Match against SLA, and assign priority. Also identify technical
details like application, version … If possible provide resolution advice. Match against
Known Errors, Problems and other Incidents.
• Investigation and diagnosis: SD/support group. If SD can’t help functional
escalation to 2
nd
line. If danger of SLA breach, hierarchical escalation. Try to
minimize impact by providing degraded service in the meantime. Iterative process.
• Resolution and recovery: SD/support group. Can be workaround, or
implementation of an RFC.
• Incident closure: SD, add details resolution, time spent by whom. Ask for
permission to close & close.
• Incident ownership, monitoring, tracking and communication: SD
6.3 Benefits
• Reduced business impact of incidents by timely resolution
• Proactive identification of possible enhancements
• Management information related to business focused SLA
• Improved monitoring
• Improved management information related to aspects of service
• Better staff utilization: no more interrupt based handling of incidents
• Elimination of lost Incidents & service requests
IBM Belgium
Page 18 of 31
IGS non-Confidential
• Better CMDB information
• Better User/Customer satisfaction
6.4 Implementation Considerations & Success factors
• IM is part of larger scope. When implementing, consider Service Desk, Problem
Management, Configuration Management, Change Management and Release
Management.
• Start with SD and IM. Together they represent quick wins in Service Management
implementation.
• Define IM processes and plan creation SD as early as possible (eg. When a new IT
service is implemented).
• Planning phase is 3-6 months, Implementation 3 mo-1 year (also ongoing activity)
• Up to date CMDB is a prerequisite for an efficient IM process.
• Resolving incidents is much easier when having a knowledge base of resolved
incidents/problems
• Paper based systems are not practical
• Know what the SLA’s are so that incidents resolution can be targeted against it.
6.5 Problems
• No management/staff commitment, hence no resources
• Lack of clarity of business needs
• Badly defined service goals
• Working practices not reviewed or changed
• No service levels defined with Customer
• Lack of knowledge on resolving incidents
• No integration with other processes
• Resistant to use the process
6.6 Reporting
• Total number of incidents
• Mean time to resolve incidents per impact code
• % of incidents resolved within response time SLA
• average cost per incident
• N° of incidents per PC
• % of remotely resolved incidents
6.7 Roles & Responsibilities
6.7.1 Incident Manager
• Drives the effectiveness of the IM process
• Produces Management information
• Manages the work of Incident support staff
• Makes recommendations for improvement
IBM Belgium
Page 19 of 31
IGS non-Confidential
• Develops and maintain the IM system
6.7.2 Incident-handling support staff
• Incident Registration
• Routing service requests to support groups when necessary
• Initial support and classification
• Ownership, monitoring, tracking and communication
• Resolution & recovery of incidents not routed to other support groups
• Closure of incidents
6.7.3 Second line support
• Handling service requests
• Incident investigation and diagnosis
• Detection of Problems and assigning them to problem management
• Resolution & Recovery of assigned Incidents
IBM Belgium
Page 20 of 31
IGS non-Confidential
7 Problem Management
7.1 Mission Statement and scope
7.1.1 Goal
The goal of Problem Management is:
• To minimize the adverse impact of Incidents and Problems on the business, caused
by errors in the IT infrastructure and to prevent reoccurrence of Incidents related to
these errors.
• To find the root cause of Problems and to initiate corrective action.
• Permanently reduce number & severity of Incidents & Problems by identifying
ameliorations in the infrastructure.
2 components:
• Reactive: identify & solve problems in response to one or more incidents
• Proactive: analyze trends of incidents and identify & solve problems before they
occur.
7.1.2 Scope
• Problem Control: handles problems in an effective way. Identify root cause (CI at
fault), and provide SD with information and work around. Similar to Incident Control,
bur more careful managed as reoccurrence should not happen.
• Error Control: Progresses from Known Errors until elimination by implementation of
a Change.
• Proactive Problem Management: Analysis of trends from incident records provides
view on potential problems b4 they occur
7.1.3 Definition
• Problem: Underlying unknown cause of one or more incidents
• Known Error: A Problem that is successfully diagnosed and for which a Work-
around has been identified
7.2 Activities
7.2.1 Problem Control
Reactive: Find the root cause of a problem and come to a Known Error (sides with Incident
Management).
• Problem identification and recording: if incidents cannot be matched against
Known Errors & Problems, or are recurring. Also when major Incident occurs.
• Problem Classification: determine effort required to detect & recover failing CI.
Impact on service levels assessed. CMDB helps! Determine: category, impact,
urgency, priority.
IBM Belgium
Page 21 of 31
IGS non-Confidential
• Problem Investigation and diagnosis: leading to a Known Error (which includes a
workaround for the associated Incident). Find underlying cause. Procedural errors
don’t become Known Errors. These problems are closed with appropriate
categorization code.
7.2.2 Error Control
Reactive: solve the Known Error (sides with Change Management)
• Error Identification and recording: faulty CI is detected, and Known Error status is
assigned.
• Error assessment: initial assessment of means required to solve the problem and
raise of an RFC. The actual work done is under control of Change Management.
Also the decision can be that some errors are not solved because too expensive e.g.
• Recording error resolution: solution for each Known Error should be in the PM
system, made available for incident matching
• Error Closure: After successful implementation of the Change, the error is closed
together with all associated Incident records. Potentially interim status given.
• Monitoring Problem and error resolution progress: Change Management is
responsible for implementing RFC’s, but Error control is responsible for monitoring
progress in resolving Known Errors and the continuing impact of the Problem. Hence
close interaction between both (PM might raise CAB). Monitoring against SLA
should happen.
7.2.3 Proactive Problem Management
Identify Problems and Errors b4 Incidents occur.
• Trend Analysis: identify ‘fragile’ components and their reason. Requires availability
of sufficient historical data.
• Targeting support action: towards Problem areas requiring most support time, or
causing most impact to the business (volume of incidents, n° of Users impacted, cost
to the business).
• Providing information to organization: Providing insight in effort & resources spent
by organization in diagnosing & resolving Problems & Known Errors to management.
Also information on Work-around, permanent fixes and status information should be
given to Service Desk.
7.3 Benefits
• Improved IT service quality: removal of structural errors improves the quality of
service.
• Incident volume reduction: Problems generating Incidents are removed
• Permanent solutions: Problems resolved stay resolved
• Improved knowledge on organization: PM learns from past experience, using
historical data to identify trends
• Higher first time fix rate at Service Desk: because incident & problem work around
data is available
7.4 Problems
• Absence of Incident Management and hence absence of historical incident data will
make it difficult to identify Problems.
• Failure linking Incident records with Problem records will make it difficult to move
from reactive PM to proactive PM
IBM Belgium
Page 22 of 31
IGS non-Confidential
• Lack of management commitment makes it difficult to dedicate staff on finding
problems
• Undermining Service Desk role if requests come from multiple sources.
• No knowledge base restricts benefits of the process
• Lack of prioritization of Problems that need to be tackled will take business critical
problems take longer to be solved.
7.5 Planning and implementation considerations
• PM relies on established Incident Management process. If few resources are
available, implement first Problem Control and Error Control. After that Proactive
PM.
• Concentrating on top 10 Incidents previous day can be efficient for PM as 20% all
Problems cause 80% of service degradation.
• Ensure there is staff available to spend at least part time exclusively on working on
underlying Problems in the infrastructure
• IM & PM should cooperate very well but should not fall under responsibility of the
same person due to conflicting interests.
7.6 Reporting
• Number of RFC’s raised and impact on services
• Time worked on investigation & diagnosis per problem type
• Number of Incidents linked to Problem b4 it is closed.
• Plans of resolution of open Problems relating to resources to be used
• N° of Problems/Errors split by status, service, impact, category, user group
• Elapsed time on closed/open problems
• Expected resolution time on open problems
7.7 Roles
7.7.1 Problem Manager
• Develop & maintain the process
• Review efficiency & effectiveness process
• Produce management information
• Managing problem support staff
• Allocating resources for support staff
• Developing & maintaining Problem & Error control systems.
7.7.2 Problem Support
• Identify problems
• Investigate problems until resolution or error identification
• Raising RFC’s
• Monitor progress on resolution of Known Errors
• Advice IM on Work-around for incidents related to Problems or Known Errors
• Identify root cause
• Identify trends and potential problem sources
• Prevent replication of problems to other systems.
IBM Belgium
Page 23 of 31
IGS non-Confidential
8 Change Management
8.1 Goal, scope and definitions
8.1.1 Goal
Change Management ensures that standardized methods and procedures are used for
efficient and prompt handling of all changes. To minimize the impact of Change-related
Incidents on service quality and to improve the day-to-day operations of the organization.
8.1.2 Scope
• Responsible for managing change processes involving:
o Hardware
o Software
o Documentation & procedures associated with IT infrastructure in scope
• Development is not under CM but under Release Management, but should have its
own project based CM.
• Change Management produces approval for proposed Changes based on input from
CAB and Configuration Management impact data.
• CM is not about handling Service Requests (Incident Management), and not every
Problem leads to a Change.
8.1.3 Definitions
• Change: process of moving from one defined state to another
• Change Advisory Board (CAB): Body which approves changes and assists Change
Management in assessment and prioritizations of changes. CAB members can be
different depending on which changes to approve. Composition should reflect user,
customer and service provider view. CAB members should be:
o Change Manager
o Customer
o User manager/group representative
o Experts, technical consultants
o Services staff
o Contractors
o Advisory Body.
Final decision is responsibility of management.
• CAB/EC: CAB Emergency Committee. Convened for urgent changes, or when
emergency decisions have to be made.
• Forward Schedule of Change (FSC): list of all Changes that have been authorized
for implementation over a previously agreed (with the business) period of time. Input
for CAB. Agreed with Service Desk, Availability Management, SLA Management.
• Projected Service Availability (PSA): Details of Changes to agreed Service Level
Agreements, and service availability because of the currently planned FSC. Input for
CAB. Agreed with Service Desk, Availability Management, SLA Management.
• Post Implementation Review (PIR): assessment implemented changes after
predefined period of time.
IBM Belgium
Page 24 of 31
IGS non-Confidential
8.2 Activities
• Change logging and filtering: practical feasible changes only. Use standard forms.
Everybody should be able to raise requests (potentially after manager approval).
• Allocation of priorities: order in which to do changes. Risk assessment. Based on
impact and urgency; immediate, high, medium, low
• Change categorization: prior to approval, category based on impact in environment
(risk to the business).
• CAB meetings: to be prepared in advance with FSC, RFC’s, once or twice a year.
Not necessarily face-to-face meetings. Impact assessment on business &
infrastructure. Impact not implementing. CAB provides recommendations.
• Change approval: 3 principal approval groups: financial, business & technical
organizations. Formally Change authority approves changes.
• Change scheduling: best, one change at a time, otherwise manage concurrent
changes and the impact on each other. Changes best as packaged Releases.
FSC’s help in planning.
• Change building, testing & implementation: CM coordinates (supported by
Release Management). Back-out instructions! Testing includes:
o Performance
o Security
o Maintainability
o Supportability
o Reliability/availability
o Functionality
• Urgent Changes: Kept at absolute minimum. Try to test anyway, even after going life
it’s useful to test. Too many urgent changes that go wrong indicated problems with:
o Problem analysis
o Remedy testing
o Solution implementation
• Change Review: after predefined period of implementation. Check whether:
o Change had desired effect
o Users/Customers are content with result
o Implementation plan worked correctly
o Change was implemented within time and budget
o Back-out plan, if implemented, worked correctly.
• Review Change management process: for efficiency and effectiveness.
o Large change backlog: under resourced CM process
o High number of unsuccessful changes: bad CM assessment and/or change
building
o Review could point to problems in other processes, CI’s or
documentation/training users/staff
8.3 Benefits
• Better alignment of IT services to business requirements.
• Increased visibility and better communication of Changes
• Improved Risk Assessment
• Reduced adverse impact of Changes on service quality and SLA
• Less backed out failed changes
• Better Problem and Availability Management due to management information on
accumulated changes
IBM Belgium
Page 25 of 31
