oO
FUJITSU
FUJITSU SERVICES
FUJ00079946
FUJ00079946
POA Customer Service Ref: CS/PRD/122
Major Incident Escalation Process .
Version: 1.0
Process Date: 27 June 2005
Document Title:
Document Type:
Release:
Abstract:
Document Status:
Originator & Dept:
Internal Distribution:
External Distribution:
Approval Authorities:
POA Customer Service Major Incident Escalation Process
Process Definition
N/A
This describes the POA Customer Service Major Incident
Escalation Process
APPROVED
Mike Warren - Service Transformation.
Peter Thompson, Carl Marx, Tony Wicks, Mike Woolgar,
Dean Felix, Ian Daniel, Julie Welsh, Graham Mockridge, Mick
Lait, Mike Warren, Deirdre Conniss, Andy Gibson, Mik Peach,
Dave Jackson, John Flannigan, Joep Niens, Rosemary Burgess,
Nikki Hawkins, Mike Stewart, Nick Crow, Alex Kemp, Denise
Miller, Dave Wilcox, Ian Mills, Kirsty Walmsley, Ian Cooley
Dave Hulbert (POL), Richard Ashcroft (POL)
(See PA/PRO/010 for Approval roles)
Name Role Signature Date
Dave Baldwin Director Customer Services
Carl Marx Service Management Team
Manager
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 1 of 46
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Versi
ersion: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
0.0 Document Control
0.1 Document History
Version No. Date Reason for Issue Associated
CP/PEAK/PPRR
Reference
0.1 02/05 First draft — to detail the baseline Incident
Escalation process
0.2 03/05 Draft to incorporate new organisation, trigger
information and to mirror POL Incident and
Problem Management Process Profile.
0.3 05/05 Updated to reflect comments received.
0.4 09/06 Updated - second round of comments
1.0 01/07 Issued for approval.
0.2 Review Details
Review Comments by: [Date]
Review Comments to: Mike Warren
Mandatory Review
Role Name
Service Management Team Manager Carl Marx
Director Customer Services Dave Baldwin**
Business Service Delivery Manager Richard Brunskill
System Architect
Glenn Stephens
TDA Simon Fawkes
TDA David Tanner*
System Architect Mark Jarosz
Optional Review
Role Name
FS CS Service Delivery Team Manager Nikki Hawkins*
FS CS Support Delivery Team Manager
Peter Thompson*
FS CS Transformation Manager
Graham Mockridge
FS CS Business Continuity Manager
Tony Wicks**
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 2 of 46
FUJ00079946
FUJ00079946
FUNITSU SERVICES Process Date: 27 June 2005
FS CS Service Delivery Manager DataTransfer I Ian Daniel*
FS CS Service Delivery Manager Engineering Dean Felix*
FS CS Service Delivery Manager BankOnLine I Mike Stewart
FS CS Service Delivery Manager HSD Julie Welsh*
FS CS Service Delivery Manager BankOnLine I Mike Woolgar**
FS CS System Support Centre Manager Mik Peach**
FS CS Operations Manager Mick Lait*
Head of Call Centres Commercial & PS Martin Croucher
Service Delivery Manager (Ops) Tan Cooley*
Senior Consultant Deirdre Conniss**
Unix Team Leader Andy Gibson**
Northern Team Leader Dave Jackson
Operations Manager John Flanagan
Head of Data Centres Joep Niens
Technical Support Manager Rosemary Burgess
Data Networking Support Manager Oyku Tevfik
(* ) = Reviewers that returned comments
0.3. Associated Documents
Reference Version I Date Title Source
PA/TEM/001 Fujitsu Services Document I PVCS.
Template
CS/IFS/008 POA/POL Interface I pycs
Agreement for the Problem
Management Interface
CS/PRD/021 POA Problem Management I pycs
Process
CS/PRO/110 POA Problem Management I pycs
Database Procedures
PA/PRO/001 Change Control Process PVCS
CS/QMS/001 Customer Service Policy I pycs
Manual
CS/SER/023 Horizon Service Desk —I Draft
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 3 of 46
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Versi
ersion: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
Service Description
CS/PRD/074 POA Incident Management I pycs
Process
FJ/POA/NET/R Escalation Procedure I pvcs
EF/076 IS/POA
CS/FSP/002 Horizon System Helpdesk I pycs
Call Enquiry Matrix and
Incident Prioritisation
CS/PRD/122 Major Incident I pvcs
Communication Process
SMS Messaging User Guide I pycs
CS/PRD/121 SMS Major Communication I Draft
Framework Process
CS/PLA/079 Horizon Services Business I pycs
Continuity Plan
CS/PLA/080 Horizon Support Services I pycs
Business Continuity Plan
CS/PLA/O15 Horizon Systems Helpdesk I pycs
and Business Continuity
Plan
Unless a specific version is referred to above, reference should be made to the current
approved versions of the documents.
N.B. Printed versions of this document are not under change control.
0.4 Abbreviations/Definitions
Abbreviation Definition
HSD Horizon Service Desk
ISO International Standards Organisation
ITIL Information Technology Infrastructure Library
KEL Known Error Log
MSU Management Support Unit
PO Post Office
POA Post Office Account
POL Post Office Limited
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 4 of 46
FUJ00079946
FUJ00079946
FULITSU SERVICES Process Date: 27 Sune 2005
SDMs Service Delivery Managers
SDU Service Delivery Unit
SLT Service Level Targets
SMC Systems Management Centre
SRRC Service Resilience & Recovery Catalogue
SSC System Support Centre
VIP VIP Post Office, High Profile Outlet
A+G Advice & Guidance
BCP Business Continuity Plan
RFC Request For Change
KEDB Known Error Database
MBCI Major Business Continuity Incident
SCT Service Continuity Team
ocP Operational Change Proposal
0.5 Changes in this Version
Version Changes
1.0 For approval
0.6 Changes Expected
Changes
None
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence
Page: 5 of 46
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
0.7. Table of Contents
1.0 INTRODUCTION...
Ll PROCESS OWNER.
1.2 PROC! OBJECTY
1.3 PROCESS RATIONAL!
2.0 MANDATORY GUIDELINES.
3.0 DEFINITION OF A MAJOR INCIDE
3.1 INCIDENT CLASSIFICATION
3.2. INFLUENCING FACTORS..
3.3. FACTORS EXTERNAL TO POL AND POA.
4.0 TRIGGER TOPOLOGY.
4.1 TRIGGER TYPES.
4.1.1 General.
4.1.2 Branch trigger types.. 2
4.1.3 Network trigger type: 2
4.1.4 — Infrastructure Components trigger types. 2
4.1.5 Data Centre trigger types..... 3
4.1.6 On-Line Services trigger types.. 13
4.1.7 Support Capability trigger type: 13
5.0 PROCESS FLOW. 14
5.1 PROCESS DESCRIPTION...
6.0 MAJOR INCIDENT ESCALATION PROCEDURES.
6.1 TECHNICAL CONFERENCE CALL
6.2. WAR ROOM......
7.0 ROLES AND RESPONSIBILITIES.
8.0 APPENDICES...
8.1
8.2
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 6 of 46
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Versi
ersion: 1.0
FULITSU SERVICES Process Date: 27 June 2005
1.0 Introduction
11 Process Owner
The owner of this process is the POA Service Delivery Team Manager.
1.2
Process Objective
The key objective is to improve the overall major incident management process as follows:
1.3
Improvements within communication channels to become more effective and
streamlined
Improve accuracy of reporting against status of incident
Allowing technical teams the right amount of time to diagnose and impact an incident
Avoid unnecessary alerting of the customer
Assessing which incidents are major and which are “Business as Usual’
Clarify the need to communicate awareness of potential incidents
Demonstrate to the Post Office a more professional approach
Provision of clear defined roles and responsibilities
Defined reporting/update timelines through duration of a major incident.
Improved governance
Process Rationale
This document outlines the communication and management process and guidelines to be
followed in relation to Major Incidents impacting the live estate.
The methodology defined within this document augments the existing SMS framework
process presently deployed within the live estate.
The aim of the document is to provide a pre-defined process on which future major incident
communication and management will follow and that any parties involved in that process
provide updates /receive updates at defined intervals from inception to closure of any major
service impacts.
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 7 of 46
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Versi
ersion: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
2.0 Mandatory Guidelines
Whilst it is important to maintain a balance between:
a) Allowing the technical teams the right amount of time to diagnose and impact an
incident
b) Avoid unnecessary alerting of the customer
c) Assessing which incidents are major and which are “Business as Usual’
The following guidelines should be adhered to.
The Post Office Horizon Service Desk should be the first point of contact for
operational contact between Fujitsu and the end user.
The relevant technical teams who are monitoring and aware of a potential major
incident must page/call the Fujitsu Service Delivery Manager (Duty Manager out of
hours) as soon as possible, rather than wait. This is not limited to major incidents
alone, but must be delivered wherever a state other than Business as Usual has been
detected. The Fujitsu Service Delivery Manager must in turn communicate the
potential incident, to their counterpart for awareness and monitoring in POL.
The Fujitsu Service Delivery Manager (or Duty Manager out of hours) is responsible
for communicating both wp the Fujitsu Organisation and across (see appendix A) to
their counterpart in POL. Where this is impractical (i.e. leave, out of hours,
unavailable), the initiative should be taken to jump up the organisation. The important
fact is that the customer is informed in a timely manner and at the correct touch point.
This communication should be by voice or direct SMS. The communication should
include the date, time, name, nature of problem, severity, if service affecting, and the
owner for contact
The Fujitsu Service Delivery Manager should also initiate communication using SMS
via HSD, 08.00 to 18.30 or via SMC 18.30 — 0800.
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 8 of 46
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
3.0 Definition of a Major Incident
3.1 Incident Classification
As a general rule a Major Incident will always be an incident rated as severity level A (critical)
in the POA Customer Service Incident Management Process Details document (CS/PRD/074)
version 3.0. However not all incidents rated at severity level A qualify. This is because the
severity levels do not necessarily translate to the global business impact on POL’s business.
For example a single counter post office which is unable to transact, regardless of its business
volumes is rated as a severity A.
For simplicity, Incidents are classified into three impact levels: High, Medium and Low.
High — An Incident that has occurred with a significant and potentially prolonged adverse
impact on POL business. Typically these Incidents will initially require a significant amount of
reactive management before they can be controlled and resolved.
Medium — An Incident that has the potential to cause significant impact to POL business but
can be controlled and mitigated against through effective management.
Low — An Incident that requires business attention but if managed effectively will not have
significant impact on POL business.
3.2 Influencing Factors
It is important that a major incident is defined as such because of its business impact on the
day when it occurs, rather than simply being defined as a major incident because it appears on
a list. The following parameters will also feed into the consideration of whether a major
incident exists, as follows:
¢ Duration i.e. how long has the vulnerability to service already existed
e Impact across the estate, including consideration of whether a service is merely
degraded or actually stopped
e Time at which the event occurs in relation to the 24 hour business day
e Anticipated time before service can be resumed
e Impact to POL Branches, customers, clients or brand image
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 9 of 46
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
3.3 Factors external to POL and POA
The following factors are external to both POL and POA and represent an event or risk to be
managed by both parties to minimise the risk to POL’s business. The list is not intended to be
exhaustive:
e Adverse weather
e Fuel strikes
© Criminal or terrorist activity directly affecting the ability to deliver service e.g. ram
raids
e 3" Party Service Provision
o E.g. DVLA for on-line service
o BT where telecom supply is outside POA or POL control
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 10 of 46
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FULITSU SERVICES Process Date: 27 Sune 2005
40 Trigger Topology
This document is covering the mutual responsibility between POL and POA for sharing
information in the event of major incidents. POA however do not have responsibility for all
aspects of the POL business domain. The following diagram illustrates the trigger topology.
tae I Ouneship [essen]
=
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 11 of 46
[oo] POA Customer Service
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Process
FUJ00079946
FUJ00079946
Ref: CS/PRD/122
Version: 1.0
Date: 27 June 2005
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 12 of 46
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Feed
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
Trigger Types
4.1.1 General
The full list of triggers is documented in CS/PLA/079. Other vulnerabilities are documented in
SRRCs. The following paragraphs illustrate the types of trigger at a high level.
4.1.2 Branch trigger types
Branch major incident triggers are as follows:
e Time of day e.g. mornings 09:00 — 10:00
e Time of week e.g. Monday
e Time of year e.g. Christmas & Easter, End of month/quarter DVLA
e Scale of outage e.g. number of branches affected
e Geographical dispersion e.g. all branches in a town/city/county
e Business initiatives e.g. product launches
Duration e.g. more than an houre
4.1.3 Network trigger types
Network major incident triggers are as follows:
e Complete outage of Energis network
© Complete outage of BT network
© Complete outage of VSAT sites
4.1.4 Infrastructure Components trigger types
Infrastructure component major incident triggers are as follows:
¢ Total loss of environments providing individual on-line service capability
e Breach of access to data centres
e Breach of security
e Virus outbreak
© Loss of inter-campus links
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 13 of 46
[oo] POA Customer Service
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Process
FUJ00079946
FUJ00079946
Ref: CS/PRD/122
Version: 1.0
Date: 27 June 2005
4.1.5 Data Centre trigger types
Data centre major incident triggers are as follows:
e Network/LAN outage
e Loss of Wigan/Bootle data centre
e Breach of security
4.1.6 On-Line Services trigger types
On-Line services major incident triggers are as follows:
¢ On-line service unavailable within Data centre (not counter level)
e¢ Number of Branches not able to provide on-line services — as defined by POL
e 3" party provided service failure — Link, Fujitsu Group
4.1.7 Support Capability trigger types
Support capability major incident trigger types are as follows:
© Fujitsu supplied infrastructure
o Unable to raise OCP for “fix on fail”
©. Loss of E-mail
o Unable to access Support Capability Systems
© Access to support documentation
© Loss of Building providing support capability
o Fire
o Exclusion Zones
© Loss of key staff due to a major event
©. Terrorist attack
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence
Page: 14 of 46
FUJ00079946
FUJ00079946
[<2] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Version: 1.0
Process Date: 27 June 2005
5.0 Process Flow
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 15 of 45
POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Version: 1.0
Process Date: 27 Sune 2005
FUJ00079946
FUJ00079946
braces (SRO ‘sseigned seetie at soubor
pou ser
5.1 Process Description
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 16 of 45
O POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Versi
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
Box I Title Description Key Action owner
ee i timescales
Incident identified
Incident identified, the definition of an Incident is “Any event which is not part of the standard
operation of a service and which causes, or may cause, an interruption to, or a reduction in, the
quality of that service.” (CS/PRD/074). An Incident may be reported from within POL domain, a
supplier domain or other route
2. IReported to Service I The incident is reported into the HSD/SMC from within POL domain, a supplier domain or other
Desks: route.
Post Masters to HSD/SMC I The incident is profiled as a potential Major Incident as outlined within this document, including
HSD to POL SCT consideration of all influencing factors, time, geographical coverage, business impact, security,
public perception, duration and relevant business initiatives coinciding at POL.
SDU to HSD/SMC . . F . .
The line of business POA Service Delivery Manager or Duty Manager (out of hours) will be
3rd Parties to HSD to POL I alerted, and is empowered to make decisions. POL SCT will also be alerted, subject to POA
SCT SDM agreement.
3rd Parties to POL SCT to
HSD
3. I Initial impact assessment I With agreement from POA Service Delivery Manager, or I T+3 HSD/SMC/POA
of incident Duty Manager out of hours, a SMS will be sent to POA and If the Incident Service
POL Management from HSD in core hours or SMC out of I ; po Delivery
. . is classified a
hours alerting to the potential existence of a Major Incident. Major Incident Manager/POA
This SMS will be sent unless expressly forbidden by the SMS Service
POA Service Delivery Manager or Duty Manager. For communication Delivery Team
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 17 of 45
FUJ00079946
FUJ00079946
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
clarity, the default position is to send the SMS once I will be invoked I Manager
discussion has taken place with the POA Service Delivery I within 3
Manager or Duty Manager and POL SCT. minutes of
The SMS message will read, “This message is to alert the classification.
potential existence of a Major Incident impacting the live
estate. A further update will follow in 15 minutes”
POL SCT will have been advised, by HSD/SMC. POL SCT
will have contacted Richard Ashcroft immediately, who may
contact the POA Service Delivery Manager or Duty Manager
directly.
The POA Service Delivery Manager will advise the POA
Service Delivery Team Manager who will in turn advise:
e Service Support Team Manager
e Service Management Team Manager T4145.
« Customer Service Director All timescalesIPOA Service
Upon initial confirmation of a Major Incident impacting the I quoted within I Support Team
live estate, the POA Service Support Team Manager I this document I Manager
manages the incident from this point forward. as viewed as
An initial impact assessment of the incident is undertaken by I Maximum, — to
members of the POA Service Team in consultation with I>@ _ improved
POL, taking into account impact on: upon wherever
possible.
e Live Service
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 18 of 45
o
FUJITSU
FUJITSU SERVICES
POA Customer Si
Major Incident Escalation Process
‘ice Ref: CS/PRD/122
Version: 1.0
Process Date: 27 June 2005
FUJ00079946
FUJ00079946
e Financial Integrity
e Business Image
The need for a Technical Conference Call will be
established as appropriate based upon the impact of the
Incident. This will involve POA Service Team (as
appropriate to the specific Incident), SMC, SSC, Core
Services, Networks, and HSD as appropriate for each
Incident. The outcome of the Technical Conference Call will
be determination of the Incident being classified as Major
(medium to high impact) or Business As Usual (low). An
initial action plan will be defined.
If the Incident is assessed as a Major Incident the POA
Service Owner and POA Service Support Manager will move
into a Technical Bridge Area on 6" floor at Bracknell.
An impact analysis will be produced referencing:
e Volume of calls received regarding the Incident
¢ Calls from 3° parties
e Support Team Analysis
e References from SRRC
e References from KEL
e Reference to Major Trigger Table
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 19 of 45
FUJ00079946
FUJ00079946
[<2] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Version: 1.0
Process Date: 27 June 2005
A further SMS will be sent, this will read “A Major Incident is
confirmed, impacting the live estate, POA working group in
attendance, further update in 30 minutes”
lf the outcome of the Technical Conference Call is that the
Incident is determined Business As Usual (low) then an SMS
communication will be sent stating that the Incident is not a
Major Incident.
From this point forward, with the exception of box 26, which
is the responsibility of the Service Support Team Manager,
SMS communication, timing and delivery requests, becomes
the responsibility of the POA Service Delivery Manager. 30-
minute updates should be the norm wherever possible.
If the incident is low impact go to box 12. If the incident is medium to high impact proceed to box 4(Communication box 23):
4. I Appoint working group IThe POA Service Support Team Manager appoints aI T + 35. POA _ Service
and hold conference call working group specific to each incident to participate in the Support Team
management of the incident. The working group will be Manager
made up of POA Service Support Manager, Service Delivery POL
Manager, Service Team Manager, Support Team, Business
Continuity Manager, 3 party Account Manager along with
relevant POL business/technical managers as appropriate to
each incident. Request to join the working group will be
made by SMS via HSD in core hours or SMC out of hours.
Service
Manager
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 20 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
Communication will be via a conference call to be convened
and chaired by POA Service Support Manager, this may
include 3 party representatives subject to domain.
The POL Service Manager with input and assistance from
POA will chair POL domain incidents.
5. I Agree actions to further I Conference Call 1: An action plan is formulated and I T +45. POA _ Service
assess impact agreed. Time of next conference call is agreed. Support Team
Manager
POL Service
Manager
6. I Follow up call and final I Conference Call 2: Following feedback on actions further I T + 60. POA _ Service
assessment — route A or B I assessment of the situation takes place. A decision is then Support Team
taken on whether to manage the incident within the Manager
appointed working group [route A] or escalate for War Room POL Service
direction [route B]. Manager
Please note: Depending on the severity of the incident the
decision to escalate may have already taken place.
If route A is chosen go to box 21. If route B is chosen proceed to box 7:
7. I Escalate for War Room I If the appointed working group are unable to provide aI Timescale POA _ Service
direction timely resolution to the incident it is escalated for MAJOR I dependant on I Support Team
INDICENT ESCALATION GROUP direction via the I impact and I Manager
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 21 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process ;
Version: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
establishment of a War Room. nature ofIPOL Service
incident. Manager
The nature of the incident determines which POA Service
Team members and POL Managers are involved in the War
Room but it would include all or some of the following:
e POL General Manager IT
e POL Head of Technical Services
e POA Service Support Manager (War Room
Chairman)
« POA CS Director
e 3rd party Executives
e Appointed working group representatives as
appropriate
The purpose of the War Room is to:
Provide appropriate direction on Incident resolution
Provide added impetus to restoration of service ASAP
Involve 3" party Executives
Define communication intervals to Key Stakeholders
Provide focused Incident Management in line with the
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 22 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
e impact and severity of the Incident.
8. I War Room drive working I War Room provide the appropriate direction on the incident I Timescale POA _ Service
group actions resolution priorities. dependant onI Support Team
impact and I Manager.
nature of War Room.
incident.
9. I Actions delivered to I Plan developed to resolve the incident with POL and other I Timescale POA _ Service
manage incident within I support teams as appropriate. dependant on/I Support Team
POA Communication to box 26. impact and Manager.
nature "I War Room.
incident.
10. I Incident ‘managed down’ I Action agreed using standard technical work procedures I Timescale POA _ Service
to control of Working I across the estate. dependant onI Support Team
Group level Communication to box 27. impact and I Manager
nature of
incident.
11. I Incident resolved within I Verify incident is now resolved and can be closed. Timescale POA _ Service
working group Communication to box 28 dependant onI Team Support
. impact and I Manager
nature of
incident.
Low impact incident [continued from box 3]:
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 23 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
12.IManage as Incident I The Service Desk and Support Teams manage low impact incidents using existing Incident
Management Process I Management Process (CS/PRD/074) and Problem Management Process (CS/PRD/021).
(CS/PRD/074) and
Problem Management
Process (CS/PRD/021) by
Service Desk and Support
Teams
13.I Incident opened and I On notification of the Incident an Incident record would be I T + 15. HSD (IMT and
ownership assigned opened by HSD, and should no KEL already exist, Service
agreement reached on the appropriate team to be assigned Delivery
the Incident for investigation. Manager if
appropriate)
14. I Further root cause I Investigation into the Incident is undertaken and a specific I Timescale POA
analysis cause identified. dependant on Support Teams
impact and
nature of
Please note: This may mean that the Incident is escalated I incident.
to ‘Major Incident’ status. If this is the case return to box 3.
15. I Agree proposed solution A solution to the problem is agreed. This may be solely I Timescale POA
within POA or within 3% party domains, as appropriate. I dependant on
Follow known KEL’s wherever possible. impact and Support Teams
Closure criteria are clearly defined. nature of
incident.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 24 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
16.I Devise action plan toIPOA Support Team work together to develop a plan for I Timescale POA
deliver solution Incident resolution. dependant on Support Teams
The action plan should include details of the specific impact and
solution including: nature oO
incident.
« Resources
e Timescale
e Service Impact Assessment
¢ Defined Communication Plan
e Defined Regression Plan
¢ Go/No-Go Decision
e POACS CP
17. I Deliver action plan Action plan is presented to appropriate management for I Timescale POA
agreement. An OCP will be raised and managed as I dependant on
described in Change Process (PA/PRO/001) impact and Support Teams
nature of
incident.
18. I Deploy solution Once approved, the solution is scheduled and implemented I Timescale POA
to agreed timescales, delivering the approved OCP/CP or in I dependant on Support Teams
the case of a Software fix via Release Management. I impact and
Regression testing and assurance is accepted at this point. I nature of
incident.
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 25 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
19. I Monitor solution The solution is monitored to ensure successful I Timescale POA
implementation. Post implementation appropriate monitors I dependant on Support Teams
with the capacity to monitor through Tivoli and System Logs I impact and
will be in place. nature of
incident.
20. I Close Incident Incident is resolved and the Incident record closed. Timescale HSD/POA
dependant on
impact and Support Teams
nature of
incident.
Route B [boxes 21 and 22]
21.I Manage within working I If the incident does not require senior management direction I Timescale POA _ Service
group the existing POA Service Management Team can manage it I dependant on I Delivery
through to resolution. impact and I Manager
The POA SDM will initiate SMS messages updated every 30-I Nature of I HSD/SMC
minutes. incident.
22.I Resolve within working I Verify incident is now resolved and can be closed. Timescale POA _ Service
group dependant on I Delivery
impact and I Manager
A Major Incident Review now takes place as box 29. nature of
incident.
Communication [boxes 23 to 28]
23. I Communicate incident to Details of the major incident are sent via SMS to the Key I T + 20. Service
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 26 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
Key Stakeholders Stakeholders in POL Ltd and POA. An SMS text alert Delivery
message is sent to advise that a major incident has occurred Manager via
and that further detail will be sent out. HSD
24. I Communicate incident I An update on current situation/status of the incident is sent I Timescale POA _ Service
update to Key I via SMS to the full range of POA SMS groups including I dependant on I Delivery
Stakeholders POL. impact and I Manager
nature of usp/smc
incident.
25. I Communicate incident I An update on current situation/status of the incident is sent I Timescale POA _ Service
update to Key I via SMS to the full range of POA SMS groups including I dependant on I Delivery
Stakeholders POL. impact and I Manager
nature =f I HSD/SMC
incident.
26. I Communicate incident I An update on current situation/status of the incident is sent I Timescale POA _ Service
update to Key I via SMS to the full range of POA SMS groups including I dependant onI Support Team
Stakeholders POL. impact and I Manager
nature =f I HSD/SMC
incident.
27. I Communicate incident I An update on current situation/status of the incident is sent I Timescale POA _ Service
update to Key I via SMS to the full range of POA SMS groups including I dependant on I Delivery
Stakeholders POL. impact and I Manager
nature of usp/smc
incident.
28. I Communicate incident I An update on current situation/status of the incident is sent I Timescale POA _ Service
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 27 of 45
FUJ00079946
FUJ00079946
-e) POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process Version:
ersion: 1.0
FUAITSU SERVICES. Process Date: 27 June 2005
update to Key I via SMS to the full range of POA SMS groups including I dependant on I Delivery
Stakeholders POL. impact and I Manager
nature =f I sp/sMC
incident.
29. I Major Incident Review A review of the Incident including consideration of: Within 24IPOA Service
* Lessons learnt hours of the ISupport Team
Incident Manager
¢ Incident definition closure
« What went well
¢ Timeline
e Changes required to infrastructure
« A review of the Major Incident Communication
Process
e Root Cause Analysis * if known at this point
e Business impact
e Action plan
e Service Improvement Plan update
30. I Major Incident Record A written report detailing the agenda items of the Major POA _ Service
Incident Review, for distribution to relevant POA, POL and Support Team
3" party stakeholders. Manager
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 28 of 45
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
6.0 Major Incident Escalation Procedures
61 Technical Conference Call
This is a technical conference for experts to discuss and analyse the incident enabling an
appropriate action plan to be formulated to restore the service to POL without delay. The
Technical Conference Call will baseline the anticipated response, covering resolution, time and
resources required.
The Technical Conference Call will be incepted at T+15.
Telephone number for the Technical Conference Call is } the Chairman will
enter the call prior to the attendance of other callers and enter a designated PIN, allowing
direct entry for subsequent callers.
Participants required on the call will be contacted via SMS as appropriate. The Service
Support Manager will initiate the Technical Conference Call, with information passed onto the
War Room if deemed appropriate.
62 War Room
The purpose of the War Room is to provide a focused area from which strategic decisions can
be made regarding a Major Incident confirmed a MBCI.
Attendance will be mandatory from the following or their designated representative:
¢ POA Customer Services Director
e POA Service Management Team Manager
e POA Business Continuity Manager
e POA Service Delivery Manager (Business line specific)
© POL Head of Technical Services
© POL Service Continuity Manager
e POL Service Delivery Manager
e 3° Party Account/Service Delivery Manager
Actions within the War Room include:
e Agreement of Containment Plan
e¢ Documentation of all agreed actions with owners, and timescales
e Consistent management of the major incident across all involved locations
¢ Co-ordinate meeting times and locations
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 29 of 45
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
In the event of a major incident requiring a War Room to be incepted, it is envisaged that this
will be in place at T+60. Participants required in the War Room will be contacted via SMS as
appropriate.
Telephone number for the War Room working group is} the Chairman will
enter the call prior to the attendance of other callers and enter a designated PIN, allowing
direct entry for subsequent callers.
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 30 of 45
FUJ00079946
FUJ00079946
cO POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUATTSU SERVICES Process Date: 27 June 2005
7.0 Roles and Responsibilities
This section defines the roles and responsibilities individuals and teams have with regard to the
Major Incident Escalation Process.
Role Responsibilities
HSD (IMT) I Log calls received from Post Masters and from the Post Office Service Continuity Desk
Core Hours Place and progress calls with support for investigation.
PMc Out of Notify Post Office Account Service Delivery Manager (Duty Manager, out of hours) of
I Mr (to incidents impacting /with potential to impact the live estate
Core Escalate
Hours) Central point of contact for progress updates and deployment of SMS messages.
P prog! iP ploy ig
POA Respond to Incident or system faults.
Service Diagnose and impact the incident.
Teams
Attend Technical Bridge meetings & carry out service recovery tasks within the agreed
action plan in order to recover the service to the customer.
POA : POA TDA’s and Development Managers involvement will be dependant upon the level,
Technical verity and potential business i ff the Major Incident under review and will be via th
Desion severity and potential business impact of the Major Incident under review and will be via the
8 Technical Conference Call.
Architectes
POA Dev
Managers
POA Service Delivery Managers and Duty Managers maintain the same level of
Service empowerment as previously exercised. Responsibility, particularly out of hours, lies
Delivery with the Duty Manager to decide incident resolution routes most suitable for the
Manager delivery of the highest level of customer service.
Initial impact assessment
Management of team resources involved in the incident including the need for 3" party
involvement.
Notify the POL Service Delivery Manager of initial findings and if inconclusive and requiring
additional parties to be involved, instruct that a Technical Bridge be initiated.
Inform POL of issues with regard to the live estate
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 31 of 45
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FULITSU SERVICES Process Date: 27 Sune 2005
Communication Co-coordinator during a major incident
Communication to POL Business Continuity Manager and POL Service Delivery Managers.
Agreeing content of Communication updates.
Appointment of a single problem owner to coordinate investigation.
Core Management of Core team resources involved in the incident
service Attend Technical Bridge meetings & carry out service recovery tasks within the agreed
Te action plan in order to recover the service to the customer.
Managers Ensure that Core Services Senior Management is notified of the major incident and the
impact presently in relation to the live estate within the organisation (Escalation Procedure
IS/POA FJ/POA/NET/REF/076).
POA Appointment of a single problem owner to coordinate investigation.
Service Initiates the technical bridge meeting if required subject to the severity of the incident and
Support . . . oe
Team number of technical support teams required to resolve the incident.
Manager Calling all appropriate parties together in a 'War Room’ at the outset to establish situation
Or and develop action plans
: Ensuring effective communication lines
Most Senior
POA Ensuring Major Incident management is followed
Manager Ensuring War Room meetings include updates on actions agreed at start of each new
present. session.
Developing Communicating Action Plan
Developing Containment Action plan.
Developing Recovery Action Plan.
Chairman of the Technical Bridge.
Identifying Technical Expects required for investigation.
Management /co-ordination of 3" party Support Managers, escalation to 3rd party Account
Managers and 3 party 3/4" line technical experts, if required throughout the duration of
the incident
POA Business Continuity Manager is the contact for coordinating Executive decisions and
Business communication directives from Exec.
Continuity Attending ‘War Room’ Major Incident meeting
Manager
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 32 of 45
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FULITSU SERVICES Process Date: 27 Sune 2005
Ensuring representation in meeting to agree high-level action plans.
Notifying the Post Office Business Continuity Manager on the status of the incident and
appropriate action plan to be agreed and subsequently implemented.
POA Head of I Ensures appropriate Service Management resources are available to manage the incident and
Service return the service to normal operation.
Management I _ .
Informs counterpart in Post Office of incident and progress
Escalates to POA Customer Service Director
POA Liaises with POL IT Director and POL General Manager (Network Support Services)
Customer Escalates to POA Account Director if appropriate
Service
Director
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 33 of 45
FUJ00079946
FUJ00079946
co POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
PUATSU SERVICES Process Date: 27 June 2005
8.0 Appendices
8.1. Escalation Communication Protocol
The primary principle:
Up”
and
“Across
Escalation protocol:
Fujitsu; Service Owners SCT / Problem Managers
Service Delivery Team Manager Service Continuity Manager
Head of Service Management Supplier and Service Performance Manager
Customer Service Director Head of Network Support / IT Director
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 34 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process ;
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
8.2 Major Business Continuity Incidents (MBCI)
Sub-System I Location I Loss of I Loss of I Loss of Services Core Hours I Predicted Non-Core Hours
Component I of Failure I Capacity Resilience Impact Recovery Time I Impact
Data-centre Bootle Data-I 50% loss _ of I Generally total loss I 1, NBS service whilst 1, No NBS service I 1, NPS —1 hour 1, No NBS service
failure centre infrastructure of resilience. NPS is failed over. whilst NPS is failed whilst NPS is failed
capacity 2, APS — Quantum I over. over. Approximately
Loss of EMC Data-
duplication
emergency, TES-QA
access and OBCS
database whilst
Database server failed
over
3, Delay in Bootle
outlets connecting into
Wigan.
4, Loss of POLFS
Production and
Development services.
2, No OBCS ‘foreign
encashment service
until Database server is
failed over.
3, No online services
until ‘Bootle’ outlets
connect into Wigan.
4, No POLFS
production service until
fail-over of the Wigan
POLFS QA-TEST
2, OBCS service 60
to 90 minutes for fail-
over.
3, Bootle outlets
should connect to
Wigan within 1
minute.
4, fail-over of the
Wigan POLFS QA-
TEST server to run
the Production
service takes
approximately 48
hours.
Note full fail-over of
the supporting
servers and services
2 hours between
18.00 and 08.00
2, No OBCS service
for 120 to 150
minutes.
3, Bootle — outlets
should connect to
Wigan within 1
minute.
4, Fail-over of the
Wigan QA-TEST
server to run the
POLFS Production
service takes
approximately 48
hours.
5, Potential delay to
Ref Data drops.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 35 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
to Wigan, e.g., KMS, I 6, Potential delay to
ACS, OCMS servers I Harvesting.
etc will take A -
wl ax I 7, Potential delay of
approximately five] kp Client fle
hours.
transfers.
Data-centre Wigan Data-I 50% loss of I Generally total loss I 1, Delay in Wigan I 1, No online servicesI 1, Wigan outlets} 1, Wigan outlets
failure centre infrastructure of resilience outlets connecting into I until ‘Wigan’ outlets I should connect to I should connect to
capacity Bootle. connect into Bootle. Bootle within 1] Bootle within 1
. Loss of the OMDB and minute, minute.
Loss of EMC Data-I 7.7 . 63 e - -
‘ Tivoli monitoring. tent 2, fail-over of the I 2, fail-over of the
duplication 2, The Tivoli master I Tivoli infrastructure I Tivoli infrastructure
tren the POLFS QA- OMDB. ‘0 gateways and takes approximately I takes approximately
Service require at"! 3 hours during core I 4 hours during non-
over to Bootle
hours core hours
3, Potential delays
for the distribution of
software via Tivoli.
Inter-Campus Both Links 100% loss of I Total Loss of EMC I Both Data-centre I 1, Loss of NBS service I Recommendation: Recommendation:
Link Inter-Campus I Data-duplication I operating in isolation. I to Wigan. 1,Manually perform I 1,Perform a
Link capacity No cross campus data I 2, Maestro running on I a closedown of the I closedown of the
synchronisation. the Bootle Database I Wigan LAR and I Wigan LAR and
server loses ISDN routers to force I ISDN routers to force
ae all Wigan connected I all Wigan connected
The: abiity ° schedule outlets to Bootle. I outlets to Bootle.
work on any £4" I This would take I This would take
maestro agents. z
approximately 30 I approximately 90
minutes. minutes.
2, Failover the I 2, Fail-over the
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 36 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
Tivoli infrastructure I Tivoli infrastructure
(Tivoli master TMR, I (Tivoli master TMR,
PO gateways andIPO = gateways and
OMDB) this takes I OMDB) this takes
approximately 3 I approximately 4
hours during core I hours during non-
hours core hours.
3, Potential delays
for the distribution of
software via Tivoli.
Database Server I Bootle No loss of I Total loss of I 1, Loss of OBCS I 1, No OBCS ‘foreign I 1, Database server I 1, Database server
Database server I Database __ server I service whilst Database I encashment service I fail-over to Wigan I fail-over to Wigan
capacity resilience server manually failed I until Database server is I takes approximately I takes approximately
over. failed over. one hour within core I two hours during non-
2, Loss of TES QA hours. core hours.
functionality while the
server is manually
failed over.
3, Loss of Quantum
emergency file delivery
to the counters until the
server is manually
failed over.
NPS (Dual I Bootle No loss of NPS I Total loss of NPS I Loss of NBS service I 1, No NBS service I 1, NPS _ fail-overI 1, NPS __ fail-over
Node) capacity. (Le. I resilience whilst NPS manually I whilst NPS is failed I takes approximately I takes approximately
the NPS is only failed over over. 1 hour during core I 2 hour during non-
run from 1 data- hours. core hours.
centre at a
time.)
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 37 of 45
Fe) POA Customer Service
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Process
Ref:
Version:
Date:
CS/PRD/122
1.0
27 June 2005
FUJ00079946
FUJ00079946
EMC Disc Array I EMC Disc I 50% loss of
Array at I EMC Array
Bootle Data-I capacity
centre
Total loss of EMC
Data-duplication.
Total impact on
services running
from Bootle
This equates to a loss
of Bootle Data-centre
(See entry above)
This equates to a loss
of Bootle Data-centre
(See entry above)
Recommendation:
Consider performing
a controlled fail-over
of Bootle Data-centre
services to Wigan.
See loss of Bootle
triggers above for
impact to services.
Recommendation:
Consider performing
a controlled fail-over
of Bootle Data-centre
services to Wigan.
See loss of Bootle
triggers above for
impact to services.
EMC Disc I 50% loss of
Array — at I EMC Array
Wigan Data-I capacity
Total loss of EMC
Data-duplication.
Total impact on
services running
from Wigan
This equates to a loss of
Wigan Data-centre (See
entry above)
This equates to a loss of
Wigan Data-centre (See
entry above)
Recommendation:
Consider performing
a controlled fail-over
of Wigan Data-centre
services to Bootle.
See loss of Wigan
triggers above for
impact to services.
Recommendation:
Consider performing
a controlled fail-over
of Wigan Data-centre
services to Bootle.
See loss of Wigan
triggers above for
impact to services.
Total loss of
internal — campus
LANs and _ Inter-
Campus Link
resilience
No loss of services.
Wigan CAT 2 reduces
the Master TMR and
Tivoli monitoring
capability
1, Very limited impact
to on-line services
(during fail-over.)
1, Automated _ fail-
over very limited
impact
1, Automated — fail-
over very limited
impact
centre
Catalyst Switch I Single 50% loss of
Switch internal campus
either Data-I LANs.
centre. (resilience) and
inter-campus
link bandwidth
Both Total loss of
Catalyst internal campus
switches at I LANs and total
Bootle Data-I loss of _ inter
centre. campus link
bandwidth
Total loss of Bootle
internal — campus.
LANs and _ Inter-
Campus Link
resilience
This equates to a loss of
Bootle Data-centre (See
entry above)
This equates to a loss of
Bootle Data-centre (See
entry above)
Recommendation:
Consider performing
a controlled fail-over
of Bootle Data-centre
services to Wigan.
Recommendation:
Consider performing
a controlled fail-over
services to Wigan.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 38 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
See loss of Bootle I See loss of Bootle
triggers above for I triggers above for
impact to services. impact to services.
Both Total loss of I Total loss — of I This equates to a loss of I This equates to a loss of I Recommendation: Recommendation:
Catalyst internal campus I Wigan internal I Wigan Data-centre (See I Wigan Data-centre (See
switches at
Wigan Data-
centre.
LANs and total
loss of inter-
campus link
bandwidth
campus LANs and
Inter-Campus Link
resilience
entry above)
entry above)
Consider performing
a controlled fail-over
of Wigan Data-centre
services to Bootle.
See loss of Wigan
triggers above for
impact to services.
Consider performing
a controlled fail-over
of Wigan Data-centre
services to Bootle.
See loss of Wigan
triggers above for
impact to services.
Correspondence I Loss of three
75% loss of
Total loss of
Probable impact _ to
Probable impact _ to
NT SLA is 8 Hours.
NT SLA is 8 Hours.
Servers Corresponde I Correspondence I Correspondence online services I online services
nce Servers I Server capacity I Servers resilience I connected in that I connected in __ that
in any I for that cluster I in that cluster cluster, depending I cluster, depending I (Up to 8 Hours if all I (Up to 8 Hours if all
cluster upon transaction I upon transaction I 3 failed I 3 failed
(Potential volumes. volumes. simultaneously). simultaneously).
MBCI)
Loss of all I Total loss of I Total loss of I Total loss of online I Total loss of online I NT SLA is 8 Hours. I NT SLA is 8 Hours.
four Correspondence I Correspondence services at outlets I services at _ outlets
Corresponde I Server capacity I Servers resilience I connected to that I connected to that
nce Servers I for that cluster I in that cluster cluster. cluster. (Up to 8 Hours if all I (Up to 8 Hours if all
in any 4 failed I 4 failed
Cluster simultaneously). simultaneously).
NBX Routing I Loss of I Total loss of I Total loss of NBX I Total loss of NBS I Total loss of NBS I NT SLA is 8 Hours. I NT SLA is 8 Hours.
Agents Bootle & I NBX routing I routing resilience I service for branches I service for branches
Wigan capacity for the I for the two clusters I connecting to the two I connecting to the two
Routing two clusters. failed clusters, i.e, I failed clusters, ie, I (Up to 8 Hours ifI (Up to 8 Hours if
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 39 of 45
Fe)
FUJITSU
FUJITSU SERVICES
POA Customer Service
Major Incident Escalation Process
Process
Ref:
Version:
Date:
CS/PRD/122
1.0
27 June 2005
FUJ00079946
FUJ00079946
Agents for
approximately 50% of
approximately 50% of
both Routing Agents
both Routing Agents
two Clusters outlets. outlets. fail simultaneously). I fail simultaneously).
(1&3 or
2&4)
NBX Loss of all I Total loss of I Total loss of NBX I Total loss of NBS I Total loss of NBS I NT SLA is 8 Hours. I NT SLA is 8 Hours.
Authorisation Authorisatio I NBX capacity I resilience for that I service for the failed FI I service for the failed FI
Agent n Agents for I for that FI FI (e.g. CAPO. LiNK or I (e.g. CAPO. LiNK or
any FI (e.g. A&L) A&L) (Up to 8 Hours if the I (Up to 8 Hours if the
CAPO. NBX_ Authorisation I NBX Authorisation
LiNK, Agents fail I Agents fail
A&L) simultaneously). simultaneously).
NBX Data-I Major fault I 50% loss of I Total loss of NBX I Potential total loss of I Loss of NBX service I Unknown. Recovery I Loss of NBX service
centre LAN I affecting NBX capacity _ I resilience NBX service. until LAN and NBX I time is dependant I until LAN and NBX
and/or Firewalls I NBX infrastructure in the I upon fault conditions I infrastructure in the
internal data-centre with the I and diagnosis that I data-centre with the
LAN within
either
Wigan OR
Bootle
fault is shutdown and
the NBX services are
manually stopped and
switched to the
operational data-centre.
failure is restricted to
one data-centre.
From time ~— of
determining that
NBX service manual
fail-over is required
allow approximately
15 minutes.
fault is shutdown.
Unknown. Recovery
time is dependant
upon the fault
conditions and
diagnosis that failure
is restricted to one
data-centre.
NBX WAN
Network to one
or more FI.
Major fault
affecting
NBX
external
WAN from
50% loss of
NBX
bandwidth
capacity to the
FIG)
Total loss of NBX
WAN resilience to
the FI(s)
Potential total loss of
NBX service.
Loss of NBX service
until external firewalls
and NBX infrastructure
in the data-centre with
the fault is shutdown
Unknown. Recovery
time is dependant
upon fault conditions
and diagnosis that
failure is restricted to
Loss of NBX service
until the WAN and
NBX _ infrastructure
in the data-centre
with the fault is
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 40 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
either and the NBX services I one data-centre. shutdown.
Wigan OR are manually stopped
Bootle and switched to the Unknown. dependant
data-centre from which I From time of upon fault conditions
the NBX WAN_ is I determining that fi A
operational NBX service manual I 24 diagnosis that
failover is required failure is restricted to
aikover is requited I ie WAN ot external
allow approximately I firewalls in only one
15 minutes.
data-centre.
DCS/ETS Loss of both I Total loss of I Total loss of I Total loss of DCS/ETS I Total loss of DCS and I NT SLA is 8 Hours. I NT SLA is 8 Hours.
Authorisation
Agent
DCS/ETS
Authorisatio
n Agents for
a cluster
DCS/ETS
capacity for that
cluster
DCS/ETS
resilience for that
cluster
service for branches
connected to the failed
cluster.
ETS service for one
cluster, approximately
25% of Branches.
(Up to 8 Hours if the
DCS Authorisation
Agents fail
simultaneously).
(Up to 8 Hours if the
DCS Authorisation
Agents fail
simultaneously).
DCS/ETS _Data-
centre LAN
and/or Firewalls
Major fault
affecting
DCS
internal
LAN within
either
Wigan OR
Bootle
50% loss of
DCS/ETS
capacity
Total loss of
DCS/ETS
resilience
Potential total loss of
DCS and ETS services.
Loss of DCS and ETS
services until LAN and
DCS infrastructure in
the data-centre with the
fault is shutdown and
the DCS services are
forced to switch to the
operational data-centre.
Unknown. Recovery
time is dependant
upon fault conditions
and diagnosis that
failure is restricted to
one data-centre.
From time of
determining that
DCS services require
failover to the
alternative data-
centre allow
Loss of DCS and
ETS services until
the LAN and DCS
infrastructure in the
data-centre with the
fault is shutdown.
Unknown. Recovery
time is dependant
upon the ‘fault
conditions and
diagnosis that failure
is restricted to one
data-centre.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 41 of 45
FUJ00079946
FUJ00079946
2 POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Version: 1.0
Process Date: 27 June 2005
approximately 15
minutes.
DCS/ETS WAN I Major fault I 50% oss of I Total loss of DCS I Potential total loss of I Loss of DCS service I Unknown. Recovery I Loss of DCS and/or
Network to one I affecting DCS and/or I and/or ETS WAN I DCS and/or_~— ETS I until external firewalls I time is dependant I ETS service until the
or more FI. DCS and or I ETS bandwidth I Tesilience to I service(s). and DCS and/or ETS I upon fault conditions I WAN and DCS
ETS capacity to I Streamline and/or infrastructure in the I and diagnosis that I and/or ETS
external Streamline E-Pay. data-centre with the I failure is restricted to I infrastructure in the
WAN from I and/or E-Pay. fault is shutdown and I one data-centre. data-centre with the
either the DCS/ETS services fault is shutdown.
Wigan OR are manually stopped ;
Bootle and forced to switch to I From time —_ of I Unknown. Recovery
time is dependant
upon fault conditions
and diagnosis that
failure is restricted to
the data-centre from I determining that
which the DCS and/or I DCS/ETS service
ETS WAN is I requires forcing to
operational fail-over to the the WAN or external
alternative data-I ‘
firewalls in only one
centre allow
+. data-centre.
approximately 15
minutes.
Energis ‘Switch I Loss of any I Total loss of I Total loss of I Total loss of online I Loss of on-line services I Reconfiguring Reconfiguring
Exchange’ Energis capacity to all I resilience. services to all outlets I to all Branches on the I Branch connections I Branch connections
Point of I outlets (ADSL) (ADSL) connected via I unavailable BAS I via BAS routers at I via BAS routers at
Presence connected via that Energis site (e.g. I routers. alternative Energis I alternative Energis
(Eg., that Energis multiple loss of BAS Points of Presence I Points of Presence
Watford, site (e.g. Routers) takes approximately I takes approximately
Birmingham I multiple loss of 24 hours. 24 hours.
, Kersley) in I BAS Routers)
a disaster.
Post Office I Post Office I Total loss of I Total loss of I Major impact to: No impact to Branch I No impact to online I No impact to online
Limited —I Limited —I capacity until I resilience. online services. services. services.
©Copyright Fujitsu Services Ltd 2005 Commercial in Confidence Page: 42 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process .
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
Northern Data-I Northern SunGard at OpTIP, LFS, SAPADS, I LFS, SAPADS and OpTIP, LFS,
centre Data-centre I Hounslow fully EDG, POL TES, I POLFS services Fail to the NDC SAPADS, EDG, POL
active POLFS. adversely affected. all-over to the TES, POLFS services
DR site takes
approximately 48
hours.
adversely affected.
Fail-over to the NDC
DR site takes
approximately 48
hours.
POLFS (SAP) I POLFS Total loss of I Total loss of I Major impact to POL I No impact to online I Fail-over of the I Fail-over of the
Production Production capacity until I resilience. Financial Services. services at Branches. Wigan POLFS QA-I Wigan POLFS QA-
Service Service at I POLFS QA- TEST service to] TEST service to
Bootle. TEST service is become the POLFS I become the POLFS
activated as the Production Service I Production Service
DR Production takes 48 Hours takes 48 Hours
service
DVLA Web I Loss of both I Total loss of I Total loss of I Total loss of DVLA I Total loss of DVLA I NT SLA is 8 Hours. NT SLA is 8 Hours.
Servers DVLA Web I DVLA capacity I DVLA resilience online service for all I online service for all
Servers, i.¢.,
at both Data-
centres
Branches
Branches
(Up to 8 Hours if the
DVLA Web Servers
fail simultaneously).
(Up to 8 Hours if the
DVLA Web Servers
fail simultaneously).
WAN Network
Connection
DVLA
to
Major fault
affecting
DVLA
external
WAN from
either
Wigan OR
50% loss
bandwidth
capacity
DVLA.
of
to
Total loss of AN
Network resilience
to DVLA
Potential total loss of
DVLA service(s).
Total loss of online
DVLA I service until
external firewalls
and/or DVLA WAN
infrastructure, for the
specific data-centre
with the fault, _ is
Unknown. Recovery
time is dependant
upon fault conditions
and the diagnosis
that failure is
restricted to one data-
centre.
Unknown. Recovery
time is dependant
upon fault conditions
and diagnosis that
the failure is
restricted to the
WAN _ or__ external
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 43 of 45
FUJ00079946
FUJ00079946
[oe] POA Customer Service Ref: CS/PRD/122
FUJITSU Major Incident Escalation Process A
Version: 1.0
FUNITSU SERVICES Process Date: 27 June 2005
Bootle shutdown and the firewalls in only one
DVLA I services are From the time of data-centre.
forced to switch to the am
data-centre from which I #etermining that
the DVLA WAN is still oh SeTVIC®
operational requires forcing to
failover to the
alternative data-
centre allow
approximately 15
minutes.
PAF Web I Loss of two I Total loss of I Total loss of PAF I Total loss of PAF I Total loss of PAF I NT SLA is 8 Hours. NT SLA is 8 Hours.
Servers PAF Web I PAF capacity resilience online service for all I online service for all
Servers, i.e., Branches Branches
both servers
at each Data-
(Up to 8 Hours if the
PAF Web Servers fail
(Up to 8 Hours if the
PAF Web Servers fail
centre. simultaneously). simultaneously).
Data-centre Router 50% loss of I Total loss of I Total loss of online I Total loss of online I Allow 15 minutes for I Allow 75 minutes for
Outlet Routers I failures at I Router capacity I Router resilience services for I services for I manual shutting I manual shutting
(ISDN, LNS, I one data- approximately 50% of I approximately 50% of I down of Routers and I down of Routers and
LAR) centre, Branches Branches. forcing of Branches I forcing of Branches
Wigan or across to the I across to __ the
Bootle Shutdown any faulty I attemative —data-I alternative data-
Routers at the one data-
centre centre
centre to force
Branches across to the
alternative data-centre.
Note sofiware drops
and Ref. Data drops
could be affected
during non-core
hours.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 44 of 45
Fe)
FUJITSU
FUJITSU SERVICES
POA Customer Service Ref:
Major Incident Escalation Process .
Version:
Process Date:
CS/PRD/122
1.0
27 June 2005
FUJ00079946
FUJ00079946
Horizon Support I Loss of I Loss of I Total loss of I None or limited impact I Invoke DR and I STE14 and I STE14 and
Building (¢.g., I primary site. I approximately I resilience as each I to Services whilst DR I relocation to the I Bridgeview Bridgeview
BRAOI, STE09, I E.g. BRAOL, I 66.66% site has one I to the alternative site is I alternative site. relocation relocation
IRE11) STE09, capacity. (Le. I primary Disaster I invoked approximately 30 I approximately 90
TRE11 DR sites I Recovery location, minutes. BRAOI I minutes, non-core
generally small I eg., FELO1, relocation to FELO! I hours BRAOL
that original I STE14 and approximately 2 I relocation to FELO1
office area.) Bridgeview. hours for Hot standby I approximately 3
equipment and I hours for Hot standby
approximately 1 I equipment and
week for cold standby I approximately 1
equipment week for cold standby
equipment.
People (e.g., I Any Team Dependant Dependant —_upon I Potential impact to both I Consider reallocation I Probably >48 hours. I Consider reallocation
SOS, SMC, SSC,
POA SI, POA
Cs)
upon the nature
of a disaster
the nature of aI online and __ offline
disaster services.
of resources across
teams as appropriate.
of resources across
teams as appropriate.
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence Page: 45 of 45
[oo] POA Customer Service
FUJITSU Major Incident Escalation Process
FUJITSU SERVICES
Process
FUJ00079946
FUJ00079946
Ref: CS/PRD/122
Version: 1.0
Date: 27 June 2005
©Copyright Fujitsu Services Ltd 2005
Commercial in Confidence
Page: 46 of 46