|
Hi All,
Using RUP enables preventive testing activities. The Test Strategy and Test Plans are released at each iteration or build cycle until customer delvery (mean GO for transition [if any] or production level). After delivery the change management and control process will take place. So to start an answer to your question - I used 3 main activities :
1) Validation activities (preventive testing activities + first test preparation steps))
1.1 - At inception, requirements must be reviewed to test their effective correctness and completness; this ensure their implementability: put in place a Qestion&Answer sheet, a defect report and quality recommendations (exchanged with project managers and business or functional experts). When requirements are validated we plan for test preparation, risk analysis and test requirements... (+ At each iteration, the impact analysis is checked against the last delivery).
1.2 - At Elaboration, all macro and micro design deliverables are reviewed to test requirements coverage and the testability of the specifications (what will be coded - specs mut be clear and ready to implement). The same activities as for requirements + full coverage (ie review => Q&A => defect report => fix deliverables => validate).
Given free from defects and validated deliverables (freezed, but subject to change - see change management), the construction can start and the test team can continue with strategy, planning, test plan and test campaigns to be ready for test execution.
2) Verification activities (be prepared to detect implemented defects + test planning then execution))
2.1 - During Construction, the preparation must be completed and all test deliverables validated by stakeholders (Test Strategy, Test plan) , Test Campaigns ready and Test Requirements in place. Other tasks are suitable: TTRM (Test To Requirements Matrix) based on macro design and Requirements specification, tune Risk matrix, test the reporting framework, ... If the oportunity is given by the planning, conduct some sanity checks to anticipate blocking defects. And if its part of the process, the code can be reviewed to verify conformance to a required coding standard.
2.2) At each delivery point chek entry and exit criteria for each runed test phase: smoke tests, integration tests, system tests, acceptance tests. Manage defects and maintain a defect resolution plan and follow patches delivery as planed. Rise any iminent risk (to planning - a test requirement may be unavailable, potential damage if a decision to fix a defect is taken, ...). May be 2 or more runs were required (1st run 100% of test plan, 2nd run re-test fixed defects and test potential regressions).
The full iteration scope was tested, all risks evaluated and a mitigation plan and a defect resolution plan were validated by stakeholders. And if the deliverable is elected so sane to go forward, remember that the test preparation activities for the coming test phases (end user / business testing, performance testing, user guide testing, technical operating guide testing) were under preparation in parallel with verification activities.
Also remember that other teams were involved in support of testing activities : architectects for scenarios, scripting specialists for performance tests, DBAs for data, infra for test environments, configuration manager for deployments, ....
3) Acceptance and observation periode: All sever and most of major defects were resolved (mean fixed and re-tested passed), still few major and some minor (eh! planning arbitration) - We met the criteria to enter a transition phase (wow, oh yes!)
3.3 - At Transition, acceptance testing by end users is ongoing and we are involved in reproducing finding (defect or not defect). But at this stage we are in confidence, the end-user had visibility during the past test phases and is confident on the build. Any way, a chance to reveal somhiden defects still there because the integration environment moved to acceptance environment and was upgraded and is now under end-user governance. So we moved our re-test activities to a support environment (new one).
I'will understand if my answer is not worthy, because of a your project (size), type of planning strategy, available ressources, implemented processes and type of your project organization.
Any way I implemented this with variance for diferent projects.
Cordially yours
Saïd Sakhraji
-------------
The ultimate judgment of progress is this: measurable results in reasonable time.(Robert Anthony)
|