Completion Report

Project Summary:

Background

The Procurement and Printing Services ColdFusion applications needed to be migrated and upgraded to newer ColdFusion versions. 

Scope

The scope included the migration of Buy@Ed to the new ColdFusion version 11 server and the upgrade from ColdFusion v9 to v11.

Also included was the upgrade of Printing Services Online (PSO) from ColdFusion v9 to v10 on the hosted windows infrastructure. 

The project also 

• implemented Bamboo automated deployment for each of these applications as part of the migration
• resolved any issues directly resulting from the migration and/or automated deployment of these applications
• decommissioned the Coldfusion datasources and applications on CF9 as they're migrated
• de-activated the CF9 scheduled tasks

Objectives

1. To migrate Buy@Ed and upgrade to ColdFusion v11 before the end-of-life in July 2016. Achieved
2. To migrate Printing Services Online (PSO) and upgrade to ColdFusion v10 in conjunction with the eITs application upgrade. Achieved

Deliverables

1. Migrated and working applications. Achieved
2. Bamboo deployment script for applications. Achieved

Benefits

This project has ensured continued support for these key applications and the underlying infrastructures.

Work undertaken as part of this project has made improvements which will allow future projects and support work to realise the following benefits:

• Simplified software releases for future projects, resulting in less time and effort being spent on routine deployments
• Improved quality and consistency of software releases, resulting in reduced re-work, downtime and general user disruption

Success Criteria

• Buy@Ed and PSO applications are now operating on their new infrastructures on ColdFusion v11 and v10 respectively with no issues.
• Old CF9 applications and datasources have been decommissioned
• All listed applications can be deployed in the future via Bamboo automated deployment

Analysis of Resource Usage:

Staff Usage Estimate: 65 days

Staff Usage Actual: 84 days

Staff Usage Variance: 29%

Other Resource Estimate: 0 days

Other Resource Actual: 0 days

Other Resource Variance: 0%

Explanation for variance:

Main reason for variance is the issue encountered with the ActivePDF components of the PSO application which took a number of different resources and additional time to resolve.

Also, due to a number of periods where resources were unavailable, both the Buy@Ed and PSO aspects of this project had to be rescheduled requiring additional project management resource for the re-planning on each occasion.

Delays also caused clashes with Business resources A/L and busy BAU periods where UAT and go-live could not be scheduled. This required further planning and extension of the project timescales and budget.

There was also an increase in the project management required due to the longer duration of the project.

Key Learning Points:

The main issue on this project related to the ActivePDF component of the PSO migration (see JIRA PPS009-6 and Issue 9 for full details). Initial investigations suggested the problem was around a newer version of ActivePDF not working with the PSO application, suggesting mappings would require to be updated in the code or potentially new APIs created. However, adding some extra error handling suggested the problem related to licensing. Support calls with the 3rd party supplier identified that licensing had been changed completely since the PSO application was built and a new license key rather than coding changes was required. This is flagged for future projects using old 3rd party software and experiencing issues during migration to consider whether licensing could be an issue and perhaps instigate earlier discussion with the supplier.

This project has highlighted that the PSO system is now old and over complicated, having been designed based on processes and technology available at the time. While the Buy@Ed migration which is newer was straight forward, the PSO migration ran into difficulties partly due to it's reliance on old 3rd party software and error handling not adequately highlighting issues. It is not how it would be delivered if it was designed now and consideration should be given to replacement otherwise any future changes will also face challenges.

Conflicts for resources with higher priority projects led to a stop / start progression of the required tasks, sometimes with different resources. This increased the duration of the tasks as each time a task was picked up again, there was a review and familiarisation with the status before moving forward. And this obviously let to the project timescales and budget increasing. While it is understood that higher priority projects will always take priority, future projects in this position should consider if there is a better way to manage resourcing to lessen the impact such as

• considering an alternative resource earlier and fully handing over
• identifying a future window where the work could be re-planned to rather than trying to make a resource available for short periods of time fitted around other projects 

Outstanding issues:

None.