Configure a production deployment system


2 years ago
11 months ago


(Reporter: jwhitlock, Unassigned)




(Whiteboard: [specification][type:feature][bc:infra][bc:milestone=motorbike])



2 years ago
What problem would this feature solve?
A production infrastructure is needed to serve MDN traffic. Automated, reliable, and repeatable deployments are needed to smoothly push new features and bug fixes to production.

Who has this problem?
All visitors to MDN

How do you know that the users identified above have this problem?
MDN received 13 million visitors in December 2015. The BC production infrastructure is expected to serve 1 million requests per day.

The MDN deployment pipeline allows developers to deploy new code to production without IT assistance. A similar deployment system is needed when BrowserCompat goes to production.

How are the users identified above solving this problem now?
During the current phase of development, a "hobby-tier" deployment is used on Heroku, which has been sufficient for the beta testing load.

The Heroku instance is automatically deployed when master is updated and the TravisCI tests are "green".

Do you have any suggestions for solving the problem? Please explain in detail.
Other Mozilla project use Jenkins as the coordinator for testing and deploying releases. Plugins can be added and configured to create the release process.

Is there anything else we should know?
A sample production infrastructure is documented at:


2 years ago
Blocks: 1242437
Severity: enhancement → normal
Depends on: 1246975
Whiteboard: [specification][type:feature] → [specification][type:feature][bc:infra][bc:milestone=motorbike]
Keywords: in-triage

Comment 1

11 months ago
The BrowserCompat project is canceled.  See for current effort. Bulk status change includes the random word TEMPOTHRONE.
Last Resolved: 11 months ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.