Wikimedia Release Engineering Team/Checkin archive/20150421

2015-04-21

Team Business

• We Are All Remoties
  • non-bot-filled backchannel? +1000
  • https://www.youtube.com/watch?v=zZIXzAbw1XE
• Come to Jeebus
• Announcement
• IRC Cloud - Mukunda set up IRCAnywhere at https://chat.buffers.us ;)

Team Quarterly Goals

https://phabricator.wikimedia.org/maniphest/query/O9isnUt5IGLP/#R

Scrum of Scrums

https://phabricator.wikimedia.org/project/board/64/

Blocked on us: https://phabricator.wikimedia.org/maniphest/?statuses=open%2Cstalled&allProjects=PHID-PROJ-arpazvuktn2l647rb6us#R

• Do these CI tasks need to be tracked for SoS?
  • https://phabricator.wikimedia.org/T72068
  • https://phabricator.wikimedia.org/T88798

Beta Cluster stability / Staging

https://phabricator.wikimedia.org/project/board/497/?order=priority

Quarterly Priority: Green nightly builds on staging: https://phabricator.wikimedia.org/T88701

Quarterly Priority: Stable uptime metrics of the Staging cluster: https://phabricator.wikimedia.org/T88705

• Should I close T88705? (assuming that staging goes on hold?)

Migration to new labvirt hosts: https://phabricator.wikimedia.org/T96678

• Which instances should I move immediately, while everyone is watching?
  • deployment-bastion / to be done Wednesday Andrew morning
  • deployment-db1
  • deployment-db2
• What other projects should I move to the head of the line before I do a bulk 'everything else in labs' migration?
  • integration , we can unpool instances from Jenkins via https://integration.wikimedia.org/ci/
  • staging
  • cvn

Deployment Cabal

• Currently making a technical evaluation of suitability of ansible v. salt
  • Antoine: we have salt already. Might consider talking with Ryan Lane who did a nice analysis/comparaison.
  • http://ryandlane.com/blog/2014/08/04/moving-away-from-puppet-saltstack-or-ansible/
• Trebuchet is kind of terrible :\
  • Antoine: in this context does 'terrible' means 'bad' or 'awesome'? :) bad
  • Might want to use ansible/saltstak to replace Jenkins jobs

Test history

Quarterly Priority: By team test history: https://phabricator.wikimedia.org/T88706

• Positive feedback so far from James F and Jon R
• Need a home for Elasticsearch and UI: new instance in integration project
  • AGREED new instance in 'integration project with an obvious name
• Will try to finish this week:
  • Project navigation (wouldn't be "by team" without it)
  • Links to create new Phab task from failure (associate task with step and show in future builds as well)

Isolated CI instances

https://phabricator.wikimedia.org/tag/continuous-integration/board/?order=priority

Quarterly Priority: Disposable VMs - https://phabricator.wikimedia.org/T47499

• Rebuild our pool of current instances with smaller one running less jobs in paralell
  • https://phabricator.wikimedia.org/T96629 "Convert pool from a few large slaves (4X) to more smaller slaves (1X)"
  • @Andrew can we get a new instance type (between m1.small and m1.medium) for our specific usage? Wanna avoid allocating unneeded resources.
  • Andrew stated it is quite easy to create a new type and associate it to a project. Just need a Phab task.
    • Created https://phabricator.wikimedia.org/T96706

• CI checkin going on weekly basis (Tuesday 14:00 UTC or 7:00am PST)

MediaWiki Releases

Quarterly Priority: Release MediaWiki 1.25: https://phabricator.wikimedia.org/T88709