New Test Optimization Module Powered by Machine Learning
A new test optimization module is now available in Continuous Delivery Director 7.1 with built-in machine learning capabilities which allow you to achieve more effective, continuous testing. The new capability allows the users to integrate continuous testing in the continuous delivery pipeline and tests the new code as soon as a new build is deployed. The new module significantly improves your agile releases by reducing test cycle time and allowing faster iterations between development and testing.
Continuous Testing Orchestration
Use the new module to:
- See executed tests from multiple frameworks in one place and have a consolidated view on the release quality and test activities that were performed
- Orchestrate test execution during the release process
- See test results by release/application/build, and find the root cause of failures.
Reduce Test Cycle Time and Identify Gaps in Test Plan
Orchestrating an extensive set automated tests can delay releases and feedback to developers.
The new capability help developers to mitigate this by:
- Prioritizing which tests should be executed against each build that is under test
- Identify gaps in the test plan and help to ensure that all changes are properly tested
Release Quality Report
This report helps you gain confidence in the quality of your release. The report shows quality information such as:
- Uncovered files - which classes that changed during the release that have no test coverage. This will help you understand that you tested the changes properly, without any gaps.
- Flaky tests - the number of tests that are flaky in your release. It can indicate the amount of 'buggy' code.
- Test Distribution - how your test execution went. Success vs. failure of tests during the release, the number of test suites executed in every test cycle, time to the first failure.
'On Error' Phase
You can now determine what to do in case an error occurs in your phase. Define in advance the list of tasks you want to run if something goes wrong in your phase, and then, in case of an error, these tasks will start running automatically. This is a very straightforward way to design your rollback or remediation process.
Activity Log
Easily perform root cause analysis and gain insight into events that occurred in CDD. CDD now has a central place where activities of all kinds are logged. Visit this page whenever you want step-by-step analysis of historic events, looking for errors or configuration details. The activity log allows easy filtering to find just what you're looking for.
Email Template
Edit the email notifications sent automatically by CDD as a result of different events such as task failure. Use the capability of editing the email templates when you want to add more information to the email like who to contact or when you want to send notifications in different languages.
'Mark as done' Automatically
Marking a release as done has important implications for reporting and licenses. Previously, you had to manually mark a release as done, but this was often forgotten.
You no longer have to remember to mark a release as done. Instead, just configure the phase that will automatically mark the release as done; once this phase is executed successfully, the release will automatically be marked as done.
Additionally, releases now have statuses. The default status is Success but if a release turns out to have problems you can change its status to reflect this. The status lets you retrospectively look at your releases to understand where and why you had problems in your release process.
Work Items Report - Enhanced Report
Work Items Report - We re-wrote this report (formerly known as the Content Report) to give you better information on work items in the CD pipelines. This report allows you to see all work items under a track or under a feature/epic/.. spanning multiple releases.
Releases Over Time Report - New Report
Releases Over Time Report - This is a new report in which you can see all releases that ended in the selected time range by status. You can see more information about each release and easily drill into it.
Integration User
To prevent loss of access if employees leave the company, you can now use a dedicated local user, known as an integration user, to manage integrations. This option is also available when using LDAP or SAML.
Track Improvements
You've got better control over your milestones through the track when you configure them as approval-driven. An approval-driven milestone in a track prevents phases assigned to it from running until the milestone is approved.
Enhanced Plugins
- CDD for Jenkins - now supports Jenkins pipeline and the ability to overwrite CDD integration per project.
- Jenkins plugin in CDD - better support for CloudBees with Query Tree Parameters.
- JIRA - now supports 'Planned vs. Actual' status for work items.
------------------------------
Assaf Shlomi
Principal Product Owner | Continues Delivery Director
Broadcom
assaf.shlomi@broadcom.com------------------------------