Closure Report
Project Summary
The project was first initiated as a two phase approach:
- Phase 1 to deliver short-term upgrade to infrastructure
- Phase 2 to evaluate and implement move to new Web version of ThankQ.
Phase 2 was descoped from this project. This work is now carried out as part of a procurement exercise in a new project: USG014 Development and Alumni CRM Procurement.
A clean installation was first considered and planned to deliver the infrastructure upgrade. Technical teams later advised that it would be possible to clone the original servers onto the new servers and to run an in situ upgrade for the operating system and database. This method was untested, but could save time and resources, and pose less risk of issues arising after the upgrade as the configuration would remain unaltered. After confirmation from ITI that this was possible, this method was chosen to upgrade both servers.
Applications Management and Development Technology took the lead in overseeing the upgrade to TEST using this method, the communication to the business and the final upgrade to LIVE.
The upgrade on TEST and LIVE was completed successfully in August 2019.
Objectives and Deliverables
|
Description |
Priority |
Outcome |
|
O1 To ensure that the application server components are upgraded to versions that will remain supported until the end of the procurement period (24 months), or to the latest version that is supported by ThankQ |
||
|
O1. D1. Provide a new virtual machine for the application with the following specification: Specification as per the "thankQ_hardwareandsoftwarerequirements_2018" document. Running on Windows Server 2016. This will in turn upgrade the following:
|
M |
De-scoped as no longer required (1) |
|
O1.D2. Our current version of the ThankQ application installed and configured by The Access Group. |
M |
Achieved (2) |
| O1.D3. Upgrade the MyEd Alumni portal channels. | Achieved | |
|
O2 To ensure that the ThankQ is running on a supported database and operating system |
||
|
O2.D1 Provide a new virtual machine for the database Specification as per the "thankQ_hardwareandsoftwarerequirements_2018" document. Operating system: Windows Server 2016 Database: SQL Server 2016 Standard Edition |
M |
De-scoped as no longer required (1) |
|
O2.D2. Access Group to perform database migration & ThankQ app repointing from current SQL 2008 database |
M |
Achieved (2) |
|
O3 To ensure that old versions of software and infrastructure are decommissioned |
||
|
O3.D1 Decommission legacy infrastructure and components on both database and application tier if they are unused |
M |
De-scoped as no longer required (1) |
- The operating system and database were upgraded in situ. New virtual machines were not required and therefore decommissioning was also not required.
- This work was carried out by Dev Tech. Access Group were available to support Dev Tech in the migrations.
Analysis of Resource Usage:
Staff Usage Estimate: 55 days
Staff Usage Actual: 43 days
Resource Variance: 22%
Explanation for variance
Scope
Following a review of the current contract for ThankQ and the proposed move to the new cloud based ThankQ with Enterprise Architecture and Procurement at the start of the project, it was established that an open procurement was required for this next phase. The current NCA (Non Competitive action) for ThankQ was extended for 2 years to allow the continued use of ThankQ during the period needed to procure a new solution. As a result,
- a new project, USG014 Development and Alumni CRM Procurement, was started in April 2019 to support the procurement
- the move to web based ThankQ was taken out of scope by Development and Alumni to avoid the burden of moving application twice in two years, one to the web based version of ThankQ, and one to the new procured solution which may not be ThankQ
See PICCL 1, PICCL 2 and PICCL 4.
Time
This project ran over an extended period of time because of several factors:
- Discussions over scope as noted above before the project could be planned.
- Identifying a suitable window for all technical teams, supplier and business partner to implement and test the upgrade. Initial scheduling was unachievable as the dates had not been validated by all. A change to HIGHER priority was approved in June to support this.
- Extended planning phase to support technical investigation to identify and recommend an upgrade path.
- Laterly lack of availability of Programme Manager to cover the closure the phase.
See PICCL 3, PICCL 5 and PICCL 6.
Resources
The project is under the agreed budget of 55 days due to:
- Effort spent in the earlier part of this project used for the procurement discussions have been transferred to the new project: USG014 Development and Alumni CRM Procurement.
- The change to a quicker upgrade method, i.e. running in-place upgrades rather than cloning the servers.
See PICCL 7.
Key Learning Points
- Involve all technical colleagues earlier on to inform planning and estimation.
- Obtain formal approval on important milestone dates through agreed channel. The initial implementation plan included dates which had not been formally validated or signed off by all stakeholders. The dates were first communicated via a meeting invite, and not seen by all stakeholders. Re-planning was needed to identify dates that could be achieved by all technical teams.
- Identify and record any risk of financial penalty when working with suppliers. The supplier was booked on specific dates to support the upgrade. These costs are covered directly by Development & Alumni. Dates had to be rescheduled at short notice. We narrowly avoided a financial penalty as we were inside the notice period but were lucky enough that the supplier could fit in other work and didn’t charge for the booking. This risk had not been captured, see lesson learned 1.
- TADs and other documentation to be updated in project and support, and shared with business partner. The service documentation for ThankQ that Development & Alumni had at the start of the project was last updated in 2014, see lesson learned 2.
- A lot of time was spent at the start of the project finalising the scope. The project proposed two phases. Risk shoud be raised and monitored on dependencies between the phases and the impact that any delay on one may have on the other.
- Review priority and escalate earlier issues and resource conflicts.
Outstanding Issues
None.