During our USAF LOGMOD and BaS&E contract, SJ’s DevSecOps Model included a highly collaborative CI/CD development environment with Jenkins CI Server that increased the speed and accuracy of software changes, using a proven Agile CI/CD pipeline process integrated with Jira. We enhanced this capability with the deployment of our SJ GovCloud environment. From 2018 to 2020, we maintained SJ GovCloud, an AWS platform, for LOGMOD and BaS&E software development and testing that replicated the government’s AWS Cloud One environment. The Government’s PMO testers and security teams used SJ’s GovCloud to conduct pre-Release testing and security scans to streamline the testing and security timeline.

Our CI/CD process included a Jenkins application-build pipeline and automated build process. When SJ developers pushed code to the Bitbucket repository, Jenkins automatically ran a series of jobs that reduced both build time and error rate. SJ configured Jenkins to automatically source code from the repository, then run a build of the source code using Maven. Jenkins was also set up to run unit tests, package the build into a deployable artifact, and send the artifacts for release to the Nexus artifact release repository. An added feature of the process was the initiation of release code quality and security scans.

The SJ Team’s testing approach was designed to engage both test engineers and stakeholders early and often in the process and to facilitate acceptance through automation. We provided a dedicated test engineer for each of the LOGMOD and BaS&E teams. They participated in the entire process from creation and refinement of User Stories through the development of code and stakeholder acceptance. Our approach streamlined development through SonarQube Scans, security scans and unit testing to validate the code during code merges, followed by test engineers’ functional and interoperability tests. Our approach also enabled the PMO to perform tests within the SJ GovCloud environment to reduce the timeline for Product Owner acceptance. This approach increased User Story velocity leading to swift and consistent capability delivery. For traceability, tests were tracked as tasks within Jira and linked directly to respective User Stories. We developed our tests incrementally which increased coverage of the system. Our approach provided 100% test coverage for all Releases from both manual and automated tests.

During development, we validated code using developer-written JUnit tests, that  automatically executed pull requests or merges into Bitbucket. This ensured only thoroughly tested and passing code merged into the branch.  We wrote test scripts using  manual and automated test development tools contained within Jira, giving each testing task a unique identifier linked  to a User Story. These test scripts are recorded as tasks that are part of the permanent record of test activity archived within Jira.

Once the code was stable following testing, SJ’s test engineers created automated testsfor regression testing to be run during future builds. We created unit tests for new code Features before we developed the code. We integrated regression tests into the CI/CD pipeline and initiated a new merge of code. We automated regression testing using automated tools such as Microfocus UFT and Ranorex or by using manual test scripts managed by tools that integrate with Jira such as TestRail or Zephyr for Jira (Z4J). We deployed a robust automated testing approach for the maximum benefit in terms of quality and velocity.

We tracked tests as tasks within Jira and linked them directly to their respective User Stories, providing  transparency and visibility to the PMO. Test results were displayed in the User Story when using TestRail. When using Z4J, test results were displayed in the test. TestRail also integrated with the automated Ranorex Studio. We made test scripts and test results available via Jira to all stakeholders with access to view in real time.

In our CPARS, the Government noted that, “SJ recommended and implemented script cost saving settings for Cloud 1. This provided the Government with recommendations for settings regarding cost savings measures within the BaS&E AF CCE Installation Guide. The resulting implementation of these measures will reduce the government’s costs associated with the CCE.SJ was instrumental in the successful migration of both BaS&E and LOGMOD from the GCSS-AF environment to the Common Computing Environment (CCE) Amazon Web Services (AWS) cloud [now known as Cloud One]. In a task for which they were not even the prime integrator, SJ closely guided the cloud integrator with code suggestions related to security authorizations along with crucial health-check coding.”