We’re going to do more than one train next week – starting Mon, 21 March 2022.

We’re aiming for four trains. 😬
____

We’ve been running the train since version 1.20, and we believe a more frequent train will be easier, faster, and safer.


There are unknowns in this process. Our goal is to surface the unknowns to increase the safety and speed of deployment.


Please reach out* if you would like to be in closer contact with our team as part of this process.

___


📅 Schedule


Join us in #wikimedia-operations IRC and #engineering-all slack for all the excitement. We’re also planning to use Google Meet sessions for pairing.


If you’re interested in joining a session (and you’ve signed an NDA) sign up (with an email that can join a meet) on the trainsperiments page on MediaWiki.

Here’s the scheduletwo trains in 24 hours to start the week:

____

🤔 Unknowns

Here’s a list of things we know we don’t know. If you can help with any of these: reach out!


  • How will more trains affect caches and site performance? (T303758)

  • Are developers going to need to be more mindful of forward and backward compatibility? (T303759)

  • How will rollback work with blockers? This one is on us, but we’ll be figuring it out as we go. (T303747)

  • Will blocker tasks be too confusing with so many trains?

  • Is this sustainable?

___


*💌 Get in touch



Tyler Cipriani (he/him)
Engineering Manager, Release Engineering
Wikimedia Foundation