Closure Review
Project Summary
The USG Application, Infrastructure & Database End of Life Compliance project work identified an urgent need to ensure the Edinburgh Global Directory
remains on supported versions of supporting software. The project was needed because the Apache tomcat version is out of security support in June 2017.
Scope
This project was to update to a current supported version of Apache tomcat. It was hoped that a simple update to Tomcat 8 was all that would be required.
Tomcat 8 is supported but this relies upon Java8 which is no longer being updated by Oracle. The project team and project sponsor agreed it was desirable and would provide an extended period of support, to migrate to Tomcat9/Java8. This in turn necessitated some work on the spring framework of the application to make it compatible with Tomcat 8.
Objectives & Deliverables
| No. | Description |
Priority (MoSCoW) |
Achieved (Y/N) |
Notes |
|---|---|---|---|---|
| O1 |
The Edinburgh Global team to be able to continue using the Edinburgh Global Directory to record collaborations in the same way as currently, and to be able to view previous collaborations. |
M | Y | |
| D1 |
Suitable development environment (to enable future projects to work on a consistent environment version) |
S | Y | |
| D2 |
Upgraded development environment (to check for any potential major issues before moving to test, and to enable future projects to work on a consistent environment version) |
S | Y | |
| D3 | Suitable TEST environment (to enable confidence in the upgrade on LIVE) | M | Y | |
| D4 | Upgraded TEST environment (to enable confidence in the upgrade on LIVE) | M | Y |
|
| D5 | Upgraded LIVE environment with a current supported Apache Tomcat version. | M | Y | |
| O2 | To retain all of the data entered into the Edinburgh Global Directory as it is in LIVE prior to the upgrade. | M | Y |
TEST data refreshed from LIVE |
Benefits
The benefit is that Edinburgh Global will continue to be able to use the Directory to record collaborations. This will enable the Edinburgh Global team to
understand where collaborations and visits occur, and enable them to plan accordingly. This will support the Edinburgh Global team in their goals of
building strategic University partnerships with governments, business and industry, supporting strong and vibrant global communities, and supporting
global citizenship and enhancing our exchange with the world.
Success Criteria
All the success criteria were met.
| Success Criteria |
Achieved (Y/N |
|
|---|---|---|
|
The Edinburgh Global Directory continues to work as it does currently, enabling the Edinburgh Global team to add collaborations and visits and search current data. |
Y | |
|
The data held prior to the upgrade in LIVE is retained and accessible in the same way as it is prior to the upgrade. |
Y | |
Analysis of Resource Usage:
Staff Usage Estimate: 20 days
Staff Usage Actual: 19 days
Staff Usage Variance: 5%
Other Resource Estimate: £0
Other Resource Actual: £0
Other Resource Variance: 0%
Outcome
The project was achieved within budget.
Explanation for variance
The duration of the project was extended due to the requirement to upgrade the Java version and Spring Framework.
Key Learning Points
Contingency - time and budget
Although it was expected that only an update to tomcat was required, there was a contingency budget available in order to carry out any additional work should the application require updating in order to remain within support. This contingency was used in order to update Java and the Spring Framework, which work was not originally planned for.
Testing
The project made use of the pilot Testrail service. The project team found this easy to set up. The business users found it easy to use. The project manager found it useful to monitor testing progress, without the need to rely on email or telephone exchanges with testers.
As there was no cost for using Testrail in the pilot, the programme manager suggests that the expected cost for Testrail use may not be justified for such a small project in the future as there were only 13 use cases to test, and the whole test period duration was less than a month. Testrail is a suitable product for monitoring testing progress, and would be more highly valuable in a more complex testing scenario. Consideration should be given to the applicability of using Testrail to the size of project in future, as well as the ability to extract the data as the users are unlikely to need to test again for 3-5 years.
Outstanding Issues
There are no outstanding issues.