18:01:48 #startmeeting 18:01:48 Let the Jenkins meeting commence! 18:01:58 #chair abayer rtyler hare_brain 18:01:58 Current chairs: abayer hare_brain kohsuke rtyler 18:02:16 #topic recap of actions 18:02:21 #info http://meetings.jenkins-ci.org/jenkins/2012/jenkins.2012-05-30-18.01.html 18:02:48 I think J-on-J is back to action 18:02:56 but I forgot what the root cause was 18:03:24 rtyler: I hope you got the check from SPI 18:03:49 I did 18:03:51 check that off the list 18:03:57 great 18:04:00 I now have money for all kinds of fishes in tokyo 18:04:21 that reminds me if you want me to place a reservation you guys better decide a date soon 18:04:40 rtyler: you up for visiting another city in Japan while at it? 18:04:49 let's figure this out later ;) 18:04:51 There's a conversation of another meet-up 2 hours away 18:04:52 yeah 18:04:54 sorry 18:04:59 let me just record other AIs 18:04:59 #info Meeting Agenda https://wiki.jenkins-ci.org/display/JENKINS/Governance+Meeting+Agenda#GovernanceMeetingAgenda-June13thMeeting 18:05:09 #action kohsuke to write to active users of svn.jenkins-ci.org that they can move to GitHub very easily if they want 18:05:16 #action kohsuke to write a scanner that auto-adds Jenkins jobs for plugins 18:05:22 #action kohsuke to split buildhive code into smaller pieces and install them to jenkins.ci.cloudbees.com 18:05:25 (this work is in progress) 18:05:27 jinkeys 18:05:39 #action rtyler to puppet-ize ircbot on cucumber 18:05:49 #action rtyler and jieryn to KVM up some more slaves somehow 18:06:01 that's not KVMable 18:06:10 I think we'll have to ues jails or something else to isolate the slaves 18:06:13 meaning I should have scratched that? 18:06:17 but that can be further discussed later 18:06:28 OK, moving on to the next topic... 18:06:37 #topic Selenium tests, switch to Cucumber/Capybara (vjuranek) 18:06:40 vjuranek: you awake? :) 18:06:49 rtyler: yup 18:06:58 #info https://github.com/jenkinsci/selenium-tests 18:06:59 great :) 18:07:34 but I have no experience with UI testing, so I don't consider myself to be quailified to make any comments;-) 18:07:49 this repository has a few tests on the "new" capybara/cucumber 18:08:06 and a bigger number of tests in the "old" (rspec?) style 18:08:13 test::unit 18:08:30 and we've made some improvements so that it has abstraction for starting/stopping Jenkins 18:08:37 ... including "/etc/init.d/jenkins start" kind of things 18:09:07 ... as well as vagrant based VM testing 18:09:12 so I thought it'd be nice if some of the LTS tests can be automated in this form 18:09:25 and then that can start running continuously on regular RC branch, etc 18:09:44 where would the jenkins instance run? 18:09:55 we have plenty of free minuts from sauce labs that we're not using, and should be 18:10:01 that's abstracted away inside the driver 18:10:10 the default "local" driver does "java -jar jenkins.war" 18:10:23 * rtyler nods 18:10:29 if we are to run this continuously that's what we'll do 18:10:37 my point was to make some agreement how the tests should look like (cucumber/capybara) 18:10:46 but we can test RC *.deb and *.rpm with a different driver, running in VM 18:11:19 IIUC, this is the "new" format we should be writing this in: https://github.com/jenkinsci/selenium-tests/blob/master/features/freestyle_build.feature 18:11:22 if we want to use the sauce labs stuff, that is best supported through cucumber+capybara, and that's also how we (at Lookout) are doing all of our browser testing 18:11:47 test::unit is just as alien to me, so I'm fine with cucumber+capybara 18:11:54 heh 18:12:15 my primary concern with test::unit, and why I want to switch us away, is that the tests become *utterly* unreadable 18:12:32 at the very least with cucumber at the outside, you have a document in plain english of what the behavior of jenkins should be 18:12:37 (I like that for other reasons too) 18:12:41 maybe we can spend the next office hours going through how you run tests and how to write more 18:12:50 ok, so looks like you agree with cucumber/capybara, so I start learing it and start writing tests in it 18:12:58 kohsuke: that's a good idea 18:13:07 vjuranek: are you around for office hours typically? 18:13:15 (meaning next week same time) 18:14:00 #action rtyler to make sure kohsuke and vjuranek have the sauce labs credentials 18:14:02 rtyler: not always, but next week it could be probably possible 18:14:33 I'm also curious what can be improved in the core to help tests 18:14:57 I suspect verifying the intended side effects can be done more easily with groovy script execution 18:14:59 or even jruby 18:15:23 #action kohsuke to schedule next office hours for selenium-tests 18:15:30 should we move on? 18:15:48 #topic New home for wiki.jenkins-ci.org 18:15:50 kohsuke: every element having unique id? IIRC I have some troublems to find some elements in xpath when looked on it last time 18:15:58 oops 18:16:00 heh 18:16:10 np, just quick idea, we can go on 18:16:21 let's come back to that topic after the meeting 18:16:25 I want to pick kohsuke's brain too ;) 18:16:46 #info I attempted JIRA and Wiki updates Monday evening 18:16:58 #info JIRA updated smoothly but Confluence didn't 18:17:19 In the end it came up all right, but it renders the "Plugins" page incredibly slow 18:17:32 and it quickly chokes up when I exposed it to public 18:17:39 in all fairness ,that page is enormous 18:17:44 was this an upgrade to confluence 4? 18:17:51 Yes, 18:17:54 4.2.4 to be exact 18:18:07 So in the end, like in 3am , I had to cut loss and roll back 18:18:32 but there's a security advisory out there, so I do need to upgrade 18:18:33 À 18:18:51 there's not a fix in 3.xx? 18:18:56 No 18:19:05 well shit 18:19:10 has anyone run 1.470 with the new LDAP Plugin, here? 18:19:11 do we have any suppor contacts at atlassian? 18:19:11 #info I checked with rtyler and I'm going to try setting up a new instance with our current database dump on cabbage 18:19:20 stephend: in the middle of a meeting, hold up 18:19:25 stephend: oh no, don't tell me it's broken again 18:19:30 I tried 1.469 last night and it didn't work 18:19:35 no, I just haven't yet tried it 18:19:36 MEETING! 18:19:47 k, sry 18:20:00 #info Confluence is already acting strange even in the current version 18:20:03 kohsuke: we can work together and try to bring up a new confluence on another host and see what we see 18:20:16 it's constantly using 30% CPU, and always rather slow 18:20:31 yeah, needs to move off that vm anyways 18:20:40 Yeah, and by moving it to another machine we can give it more memory 18:20:52 Both JIRA and Confluence could really use more memory 18:21:02 so that's the plan. 18:21:03 will you have time this weekend? I can make room in my schedule for jenkins infra this upcoming weekend 18:21:21 rtyler: I need to check with my wife 18:21:28 but I'd like to take that offer 18:21:41 I could really use your Puppet-fu 18:21:42 (argh, sorry, in a meeting) 18:21:52 #action rtyler and kohsuke to look at moving confluence over to a new host this coming weekend (pending Mrs. Kawaguchi approval) 18:22:10 so that's it for this. mostly FYI 18:22:12 kohsuke: I have a cookout saturday afternoon, so in the morning or on Sunday 18:22:25 "cookout"? another lookout thing? 18:22:52 moving on to the next topic? 18:23:01 #topic maven.jenkins-ci.org to repo.jenkins-ci.org switch over 18:23:05 kohsuke: grilling, burgers, sausages, etc ;) 18:23:15 I already like this topic 18:23:17 BBQ outing, got it. 18:23:18 death to nexus 18:23:45 #info I switched over the update center index generation entirely to Artifactory a few days ago 18:24:00 and I verified that it's serving the same things 18:24:22 so on all technical fronts, we are ready to retire maven.jenkins-ci.org now 18:24:32 jieryn: are you awake? 18:25:01 when we do the switch over, we'll set up Apache to reverse proxy maven.jenkins-ci.org:8081 (deploy endpoint) to repo.jenkins-ci.org 18:25:12 and HTTP redirect for maven.jenkins-ci.org to repo.jenkins-ci.org 18:25:29 there'll be one last rsync to move bits from maven.j.o to repo.j.o 18:25:43 and putting JFrog logo into the sponsor section (which has been long overdue) 18:26:17 so I guess I'm seeking the blessing of this meeting to this, which is technically still a "proof of concept" 18:26:18 I'm okay with all of those words 18:26:28 http://i.imgur.com/OabsH.jpg 18:26:44 I figure abayer and jieryn are better blessing giving candidates 18:26:54 I take that image as your +1 18:27:15 abayer: ^^ 18:27:40 anyone else cares to send in their +1/-1s? 18:28:04 :-( 18:28:19 +1 18:28:20 where's dean, this meeting is disappointing 18:28:40 kohsuke: I say just go for it then 18:28:41 he's somewhere in a beach of Hawaii 18:28:45 heh 18:28:58 Yeah, I think I'll go ahead and do it anyway 18:29:11 So that's it for today 18:29:19 send it home 18:29:27 #topic next meeting time 18:29:46 I guess June 27th, the same time 18:29:59 works for me 18:30:03 #info the next meeting is in 2 weeks (Jun 27), the same time 18:30:06 regular cadence is good IMO 18:30:11 Aye 18:30:20 #endmeeting