Wikimedia Release Engineering Team/Checkin archive/20150428



Team Business[edit]

For multiple timezones meetings

Team Quarterly Goals[edit]

Scrum of Scrums[edit]
Blocked on us:

Last week notes

Do these CI tasks need to be tracked for SoS?

Beta Cluster stability / Staging[edit]
Quarterly Priority: Green nightly builds on staging:
Quarterly Priority: Stable uptime metrics of the Staging cluster:
  • Unstable last week due to mgiration to new labs hardware yeah!, scap, disk full, mw stacktracing...
  • Request from Andrew: If there are any instances that can be easily destroyed + recreated, please do (eg in staging?)
  • Scap $TMPDIR craziness causing scap failures

Deployment Cabal[edit]

Took a step back from fighting over salt/ansible/trebuchet/scap and decided to make a common model for ideal deployments.

  • Once we have a shared vision, it's easier to decide what tools to use
  • Will define clear steps and requirements for each of those steps
  • Will take existing service deploys, MW code deploys, trebuchet-based deploys all into account
  • might eventually contain details about this grand vision

Test history[edit]

Quarterly Priority: By team test history:

Isolated CI instances[edit]
Quarterly Priority: Disposable VMs -

MediaWiki Releases[edit]

Quarterly Priority: Release MediaWiki 1.25:


  • TBD, Kevin Smith is doing stuff
  • Ops Access Request
    • T87467

Other Work[edit]

  • Search team will be rolling restarting Elastic soon in beta/staging/prod for some plugin updates. Unlikely to impact much, just FYI.



  • Željko May 1 (Croatian holiday)
  • Antoine: observing french holidays: Fri May 1, Fri May 7-8, Thurs May 14
  • Dan in France the week before offsite (May 11-15, leaving evening of Thursday May 7)
  • Week before hackathon: Team offsite in France -
  • May: Hackathon in Lyon, France
  • Chad vacation after offsite/hackathon (through 5/31)
  • Andrew Bogott: May 20-27, hackathon &c. (approximate), July 18-27, mostly unreachable