Once we have these we can decom or reclaim labpuppetmaster1001/1002
|Open||None||T53494 Use Beta cluster as a true canary for code deployments (epic)|
|Open||None||T87220 Minimize differences between beta and production (Tracking)|
|Open||None||T211677 Support IPv6 in beta|
|Open||None||T211575 Enable IPv6 on tools.wmflabs.org|
|Open||None||T220306 Add IPv6 monitoring|
|Open||None||T37947 Enable IPv6 on CloudVPS|
|Open||None||T209460 CloudVPS: our ideal future model|
|Open||None||T207536 Move various support services for Cloud VPS currently in prod into their own instances|
|Resolved||Krenair||T171188 Move the main WMCS puppetmaster into the Labs realm|
|Resolved||Andrew||T232429 Create in-cloud, cloud-vps-wide cumin masters|
(I tried applying the first of those manually on the new instance, ran cumin '*' id and saw 814 hosts will be targeted so that looks good - 664 of those hosts even respond how you'd expect with uid=0(root) gid=0(root) groups=0(root), we should go through the ones that don't later in a different task)
I built a second cumin host, cloud-cumin-02.cloudinfra.eqiad.wmflabs. It's partly for backup, and partly because I wanted to confirm that the existing puppetization is sufficient. It turns out that it is! The new host just required a reboot to get keyholder on board.
-02 doesn't reach quite as many hosts since I haven't hand-tuned the exceptions, but as puppet does its work it should eventually be equivalent to -01.