mozilla :: #mdndev

9 Oct 2017
09:20interactive-examp[interactive-examples] maboa opened pull request #298: attempt to resolve conflicts in site.json (master...number-examples-resolved) https://git.io/vdgSE
10:53github[kumascript] Elchi3 closed pull request #341: Typo in GamepadEventProperties.ejs (master...patch-1) https://git.io/vdEq0
10:53github[kumascript] Elchi3 pushed 2 new commits to master: https://git.io/vdgxT
10:53githubkumascript/master d5bda3f : Typo: gampad -> gamepad
10:53githubkumascript/master 24fed40 Florian Scholz: Merge pull request #341 from xnuk/patch-1...
12:27github[kumascript] teoli2003 closed pull request #342: Update filter function paths and syntax (master...master) https://git.io/vdE3x
13:46jwhitlockgood morning #mdndev
13:56metadave heads up, partial quay.io outage: http://status.quay.io/
13:57metadaveany objections to me loading a new prod snapshot into RDS?
13:58jwhitlockno go ahead
14:00metadavethanks
14:03fscholzMoin. I'm creating a now BCD npm package now.
14:10fscholzv.0.0.9 is out https://www.npmjs.com/package/mdn-browser-compat-data
14:20jwhitlockmetadave: if it's possible, we may want to put celery* replicas to 0 during a DB load
14:20metadaveok
14:20jwhitlockI don't think the errors hurt though, can be ignored
14:21metadavethe prod DB is messed up at the moment, the backup wasn't written properly to S3 and was only 150 bytes
14:21metadavebut the prod DB had already been dropped
14:21metadaveI've also setup slack alerts to tell me when backup and restore finish
14:22metadaveI'll be running the import in ~10 mins
14:24jwhitlockok
14:26jwhitlockI think I
14:27jwhitlockI think I'll push for the BCD update, and then freeze master
14:28jwhitlockhmm last Jenkins build failed
14:32jwhitlockstopped some stalled MDN docker images from 3 days aga
14:33jwhitlockstarting new build
14:44jwhitlockok killing some more and trying again
14:59jwhitlockok master build working in Jenkins again
15:29fscholz"Merging is blocked Merging can be performed automatically with one approved review. " This is interesting. Can you report how this works for you? We might want to enable this elsewhere.
15:30anthonyfscholz it's the protected branches I think.We also did this for the Webextension. It's a really nice feature
15:30metadaveRDS restore complete, celery workers scaled back to "normal"
15:37rjohnsongood morning #mdndev!
15:39jwhitlockfscholz: it is a little bit of process, a little bit of keeping devs from making mistakes from git typos
15:39jwhitlockthanks!
15:41mdnstagepushno way, jwhitlock is pushing mdn-stage 71d210e822903813cf3b86bfaf790687d61939a2
15:41mdnstagepushwatching http://developeradm.private.scl3.mozilla.com/chief/developer.stage/logs/71d210e822903813cf3b86bfaf790687d61939a2.1507563678
15:41mdnstagepushhttps://github.com/mozilla/kuma/compare/4e037b44905bdf83b87ced92855c2872b67f26d0
15:41mdnstagepush...71d210e822903813cf3b86bfaf790687d61939a2
15:41mdnstagepushFinished: update_code (4.907s)
15:43mdnstagepushthe push is now going to the webheads!! (71d210e822903813cf3b86bfaf790687d61939a2 jwhitlock)
15:43mdnstagepushFinished: setup_dependencies (146.341s), update_info (3.134s), pre_update (154.383s)
15:45mdnstagepushFinished: update_assets (97.686s)
15:45mdnstagepushFinished: update_locales (12.421s)
15:45mdnstagepushFinished: database (3.056s), update (113.164s)
15:46mdnstagepushFinished: rsync_project (16.230s), checkin_changes (16.264s)
15:46mdnstagepushFinished: deploy_app (3.123s), restart_web (2.328s)
15:46mdnstagepushFinished: restart_kumascript (18.684s)
15:46mdnstagepushjwhitlock pushed mdn-stage 71d210e822903813cf3b86bfaf790687d61939a2
15:46mdnstagepushFinished: restart_celery (13.838s), ping_newrelic (0.369s), deploy (54.609s)
15:58rjohnsonstandups: processed email
15:58jwhitlockstage-integration-tests passed
15:59anthonyI hope to be at home in time for the dev meeting but there is no subway ... for some obscure reasons
15:59jwhitlockthere's not a lot in this push
15:59jwhitlockchris mills's work on prevnext
16:00jwhitlockfunctional and jenkins test updates
16:05safwanjwhitlock: AWS on tomorrow?
16:05safwanhow long the downtime will?
16:05safwanI mean
16:05safwanthe maintainance mode
16:06jwhitlockyes, tomorrow. from the front page - MDN will be in maintenance mode on Tuesday October 10, from 9:30 AM to 1:30 PM Pacific (4:30 PM to 8:30 PM UTC) as we move data centers.
16:14sheppyooh
16:14sheppyThe time has come, eh? :)
16:15anthonyI will pray / cross fingers/practice witchcraft for you
16:15rjohnsonanthony: lol
16:17jwhitlockpush includes error handlers for untrusted domain, but shouldn't be used in SCL3
16:17jwhitlockpush song is Won't Back Down - https://www.youtube.com/watch?v=nvlTJrNJ5lA
16:18rjohnsonrest-in-peace mr. petty
16:41jwhitlockrjohnson: that worked https://mdn.mozillademos.org/es/docs/Web/Guide/HTML/Forms/My_first_HTML_form/Example$samples/A_simple_form
16:42rjohnsonjwhitlock: great! thanks for checking!
16:43jwhitlockstandups: deployed BCD v0.0.9 to production. That's probably the second-to-last push to SCL3.
16:47fscholzThanks jwhitlock!
16:48rjohnsonalmost makes me feel something for scl3, almost. :)
16:51sheppyrjohnson: hah
17:07metadavescaling down replicas before DB import added to deployment plan
17:07metadave(and post-import scaling up as well)
17:08github[kumascript] teoli2003 opened pull request #343: Integrate mdn-data (master...mdn-data-integration) https://git.io/vd2S4
17:11fscholzteoli++
17:11teoliNot 100% of sure nothing is missing there
17:12fscholzGlad to see this!
17:13sheppyYes, this is a great beginning!
17:14metadavejwhitlock rjohnson are you ok with the wording here for moc/webops notification: https://gist.github.com/metadave/cab96fe7ba35db2465d513bfdfee60f8
17:14jwhitlock
17:14metadavethank you
17:15jwhitlockwait a sec
17:15sheppymetadave: I suggest "MDN will be in maintenance mode for between 1 - 4 hours..."
17:15rjohnsonmetadave: yep!
17:15sheppy(adding "for" will be clearer that it's a duration and not some kind of weird time thing)
17:15jwhitlockmetadave: maybe they don't know what MM is
17:15sheppyActually just "for 1-4 hours"; between isn't needed.
17:15sheppyjwhitlock: Call it "read-only mode" instead?
17:15jwhitlockyeah I like that
17:16metadaveok
17:16metadavethanks for the feedback!
17:16rjohnsongood point, i like "read-only" better as well
17:16metadavegist updated
17:18* metadave posts
17:32* anthony broke vidyo again
17:50anthonyWell Vidyo works better on my phone, maybe I should just do it on my phone and plug it on the tv
17:53fscholzMDN on TV! season 1 trailer is coming soon!
17:56anthonyMDN on TV, best MDN
18:01Sphinx<flood>(I hope Sean Bean is not starring)</flood>
18:34rjohnsonstandups: 1:1 w/jwhitlock
18:34rjohnsonstandups: mdn dev bug triage w/jwhitlock,sheppy,fscholz
18:35rjohnsonstandups: reviewed/merged https://github.com/mozmeao/infra/pull/569
18:35rjohnsonstandups: reviewed https://github.com/mozmeao/infra/pull/570
18:37rjohnsonstandups: worked on https://github.com/escattone/infra/tree/rj-db-migration-job and https://github.com/escattone/kuma/tree/aws-deployments-1406546
18:39rjohnsongoing to grab some lunch
20:24metadavejwhitlock|away: when you&#39;re back, we had some questions related to the &quot; Rebase maintenance_mode branch on current master&quot; and &quot;Deploy maintenance_mode branch on current master&quot; rows in the prod release plan
20:26jwhitlockmetadave: headed out again in 2 minutes
20:26jwhitlockthe way to put SCL3 into maintenance mode is to push kuma with MM as the default
20:27jwhitlockso I have a branch that does that https://github.com/mozilla/kuma/tree/maintenance_mode
20:27jwhitlockI rebase it on the current master, then push it to SCL3
20:27jwhitlockcan explain more later
20:27metadaveok
20:28rjohnsonjwhitlock: thanks, that clears that up, we just didn&#39;t know what that branch was for
20:31rjohnsonstandups: mdn to aws final sync-up prior to go-live tomorrow w/ben, meta.dave, jg.mize
20:44metadaveping rjohnson
20:44metadaveI&#39;m in my Vidyo room
20:46rjohnsonmetadave: see you there!
21:25rjohnsonstandups: sync-up w/meta.dave regarding jenkins deployments, etc.
22:12jwhitlockstandups: working on MDN in AWS blog post
22:12jwhitlockgoodnight #mdndev!
23:45sheppyCan&#39;t believe tomorrow is the big day.
23:47rjohnsonsheppy: i know, especially given that bug 1110799 is over 3 years old!
23:47firebothttps://bugzil.la/1110799 ASSIGNED, jwhitlock@mozilla.com [tracker][infra] Move MDN from SCL3 to AWS
23:48sheppyrjohnson: Yeah! It&#39;s been a long road. The funny thing is how much of the time IT was in near-panic mode when other things had to take priority. :)
23:50rjohnsonsheppy: i know from the user&#39;s perspective, nothing will look different, but we&#39;re entering a whole new world in terms of what we can do on the backend
23:50sheppyYes. I&#39;m very happy about that. I&#39;m also admittedly happy about the amount of developer time this will free up once it&#39;s all done and we&#39;re sure it&#39;s stable etc.
23:51sheppyI&#39;m hoping that we can do a nice round of user-facing improvements after this.
23:56rjohnsonsheppy: i was about to provide an analogy that describes this exciting new backend world, thinking you&#39;d be excited to hear it, but now i see the reality!
23:56sheppyhaha
23:56sheppy:)
23:57sheppyThe thing is, of course, that a lot of those user-facing improvements we&#39;ve wanted for so long should be easier and safer to build going forward. Hopefully. :)
23:59rjohnsonyes, i hope so! i&#39;m calling it for today, have a good night sheppy!
23:59rjohnsongood night #mdndev!
23:59rjohnsoninto AWS tomorrow!
23:59sheppyNight rjohnson -- here&#39;s to a happy tomorrow! :)
10 Oct 2017
No messages
   
Last message: 7 days and 23 hours ago