Quality Assurance SIG/20190215
Attendees: Greg, Zeljko, Elena, Erik, JR, Edward, Anthony, Dom
* Welcome Dom
** Code Stewardship Reviews
*** This is a process that we use to help address un/under funded code that is deployed to production.
*** Under review: UserMerge extension, CodeReview extension, Graphoid
*** More info: https://www.mediawiki.org/wiki/Code_stewardship_reviews and https://phabricator.wikimedia.org/project/board/3144/
** Code Health Metrics WG
*** SonarQube test instance: https://sonarcloud.io/organizations/wmftest/projects (will self-host in future). Currently doesn't include Code Coverage data as we are still figuring out how to get that data in from existing systems.
*** More info about Code Health: https://www.mediawiki.org/wiki/Code_Health_Group
** Example of why it's important to sunset things - UserMerge
** Who is currently testing mobile web? Edward is
** bulk of time spent try to shore up approach to communication with team re testing
** Would love to know how people use Phabricator to manage the testing effort and the results.
** Very vast and open. There's not a lot of QA process in place for his team.
** Selenium test project update
*** no apparently correlation between presences of browser based tests and reduced incidents
** Code Health Metrics WG spike update
** Elastic 6 update.
*** It would be nice to have a test environment to which a copy of the production GETs get routed. This would help verify that the upgrade won't break under production load. Currently this is done in production. It's not ideal, but it's been working. We don't upgrade often enough for warrant spending time/energy on this, but it would probably be valuable to have.
** Editing work on Android.
*** Some gamefication features coming done.
**** Elena: Actual games?
**** Anthony: serves up content that needs more info and then there are some rewards.
**** more info: https://docs.google.com/document/d/1Y-bWokzaoRGMssNae4FhOsOs5owMa-YcruYvrCAEgUI/edit?usp=sharing
Mid-term planning is the focus
** What is the update cycle for the various testing environments?
*** Beta cluster updates about every 10 minutes
*** Services: managed and updated manually as needed by the teams that manage the services in production
*** Greg: Test wikis are updated as part of the weekly train deployment
Elena: should we have another meeting outside of the SIG to talk about Testing processes and such.
** JR to find some time to meet and discuss
JR: Would it make sense to create some overview materials or even have live sessions every once and a while to describe some of our key SD processes. For example, knowing how the train works at a high-level.
** JR to create a document to brainstorm on the topic.