15:10:08 #startmeeting Cockpit public meeting 2014-08-25 15:10:08 Meeting started Mon Aug 25 15:10:08 2014 UTC. The chair is puiterwijk. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:10:08 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:10:10 #chair puiterwijk andreasn mvollmer stefw sgallagh 15:10:10 Current chairs: andreasn mvollmer puiterwijk sgallagh stefw 15:10:26 #topic Welcome 15:11:02 hi, so for the record, this s going to be a first attempt at a meeting for Cockpit to give a short overview and have a time to discuss things that need discussion 15:11:19 who is arround for the meeting? 15:11:45 I'm here 15:12:02 me 15:12:28 #info Stef is currently on PTO 15:12:50 mvollmer: you arround? 15:13:49 okay, guess his delay is too big again, or he's still reading or something. let's just see if he will be there later 15:14:00 puiterwijk, yep 15:14:02 so andreasn sent a short agenda earlier today, so let's get started 15:14:09 #topic Fedora 21 status 15:14:55 so regarding Fedora 21: the freeze was scheduled to start at August 19th, but since we do not yet have a testing compose, this has been delayed until we do 15:15:36 so if we want to get stuff in before Fedora 21 alpha, we are still able to do so, and we should make sure to get any big changes in before that, given that that's the TC we'll be going to use for the test day etc 15:15:49 cockpit status re F21 is captured in two milestoned: Fedora 21 Sprint and Fedora 21. 15:15:57 #info No F21 testing compose yet, freeze starts after we have this 15:16:00 Those should be highest prio now. 15:16:17 mvollmer: okay, are there any things that really need to go in before Alpha, or can they wait until Beta? 15:16:32 maybe the port change? 15:16:40 I personally think that if we are going to merge #870 (port change), it should go in before Alpha 15:16:42 i think we have all agreed on it. 15:16:42 mvollmer: yeah 15:16:52 oh, it's not in yet? 15:17:06 andreasn: no, I think people forgot it. so I'd say let's merge that 15:17:09 no, i wasn't sure of the consensus. 15:17:12 sure 15:17:17 to 9090,right? 15:17:20 mvollmer: could you do that after the meeting? 15:17:22 yinzhang: yup 15:17:30 puiterwijk, ok. 15:17:40 * mvollmer opens another window... 15:17:41 #action mvollmer will merge Pull Request #870, changing our port to 9090 15:17:54 mvollmer: don't worry, it'll be in the meeting minutes :) 15:17:56 we also need a release 15:18:10 yup. do you want me to roll the release after you merged it? 15:18:51 #action puiterwijk will make a new release and submit to Fedora today 15:19:39 mvollmer: are there any other things that we really need to get in before F21 Alpha, so for the test day? 15:19:51 i can't say. 15:20:07 as much as possible, of course, since we don't want to test obsolete stuff. 15:20:22 the networking things. 15:20:45 right, and any things that make it very easy to shoot yourself in the foot 15:21:40 what about #915. 15:21:41 #action puiterwijk to review and merge open PRs 15:22:01 it is probably ok to tell testers to log in as root. 15:22:18 yeah. I think that we can just tell in the test plan to use root 15:22:23 right 15:22:27 but it would be good to get this in before Beta. 15:22:33 yes 15:22:55 so test day testers will be testing the alpha version? 15:23:02 yes 15:23:05 ok 15:23:26 * mvollmer orders brown paper bags for the team. :-) 15:23:48 mvollmer: let's hope we can make it through :-) 15:24:08 andreasn: do you have any things you really want to get in before F21 Alpha? 15:25:20 #info Issue #915 should be fixed before F21 Beta freeze 15:25:20 what's the status of "Docker is not running" issue? 15:25:39 andreasn: hmm, do you remember who was working on that? 15:25:41 that is an entire section of our UI that might not work 15:26:09 andreasn: well, it indicates that Docker is not running currently, just not very nice yet, right? 15:26:10 can't remember the number right now 15:26:45 yeah, it's not a blocker really, just nice-to-have 15:26:50 how much do we care about docker and geard for F21? 15:27:13 #705, i think 15:27:14 run-time detection of geard is broken. for me at least. 15:27:18 I feel that's f22 15:27:19 mvollmer: not sure about geard, but I think docker is pretty important, given it's one of the things in the release notes 15:27:27 i see. 15:27:41 but docker should be in workable-ish state, not? 15:27:53 yes, docker works. 15:28:04 and installed by default in F21 Server I think 15:28:07 i mean, the cockpit ui for docker works. 15:28:16 there might be issues with resource limits. 15:28:29 right, yeah. but those don't break initial work I guess. 15:28:34 but we find that out on the test day, I guess. 15:28:54 yeah. so this is really just things that need to get in because otherwise testing is hard or whatever 15:29:22 but in general, the more shoot-yourself-in-the-foot issues we can have out of the way, the better 15:29:26 andreasn, in general, i don't like the patternfly notification for the docker status. 15:29:33 andreasn: +1 15:29:49 it's a state of the system, not an event. 15:29:50 so detecting things like too long, or wrongly formatted in dialogs 15:29:50 I will go through PRs today and merge as much as possible before doing a release 15:30:01 mvollmer: yeah, it doesn't look excellent 15:30:16 I'm working through the post-review list 15:30:22 it should go away when docker runs, not when the user dismisses it. 15:30:30 yes 15:30:53 #action andreasn working through post-review, to report to puiterwijk once ETA is available to plan exact release time 15:31:01 sounds good 15:31:10 (but of course, it's better than just showing a broken page.) 15:31:59 okay, so I think that's most F21 specific things, unless someone has more? 15:32:10 one more. 15:32:23 should we merge Fedora 21 Sprint and Fedora 21 milestones? 15:32:23 mvollmer: go ahead 15:32:33 i don't understand the difference. 15:32:48 sure 15:32:52 or is Sprint for alpha? 15:32:57 yeah. I think the difference was that the Sprint was a test day from me, stefw, yinzhang and pdonlon 15:33:02 hmm, that actually makes sense! :-) 15:33:12 so everything we found there was posted in that milestone, if I recall correctly 15:33:19 ok, so Sprint for alpha. 15:33:37 yeah. that makes sense. and we should go through current issues etc to tag the Alpha correctly 15:33:40 let's rename it. 15:34:06 #action mvollmer to rename F21 Sprint to F21 Alpha 15:34:25 #action mvollmer andreasn puiterwijk to review open issues filed under F21 Sprint and F21 milestone to see which are needed before Alpha 15:34:53 we don't have a deadline for Alpha, correct? 15:35:02 mvollmer: no strict one, but that might be any day now 15:35:09 i see. 15:35:14 the day after a working TC is produced, the freeze starts 15:35:42 so, port change, release, and then we see what else we can get in. 15:35:53 yes, that makes sense. 15:36:10 that way we can at least get the port change in before Alpha, and then try to get as much in as possible 15:36:51 sounds good 15:36:57 okay, anything else F21-specific for now? 15:36:59 * mvollmer can be amazingly relaxed about deadlines. 15:37:26 #topic Pull requests that need review or help 15:37:51 okay, so I was looking through some of the PRs and saw at least 3 we probably need to discuss 15:37:57 first up, 1006 15:37:59 #link https://github.com/cockpit-project/cockpit/pull/1006 15:38:28 andreasn: not sure if you saw that, but they wanted to have some screenshots from you :) 15:38:30 whup! Blocking on me! 15:38:36 I haven't seen it, no 15:38:43 I'm happy to create screenshots 15:38:44 okay, then I guess this might be good :) 15:39:04 #action andreasn to create screenshots for #1006 15:39:23 do we need to do something special for starting cockpit from the shell? 15:39:23 next on my list was #870 (port change), but that has already been discussed 15:39:50 like, making sure it isn't exposed on the network but still runs locally? 15:39:54 mvollmer: I don't think so, given it's just a shortcut to the web browser 15:40:12 so, installing cockpit will open the firewall? 15:40:35 I think it's allowed by default on local. not sure about other zones though 15:40:38 that's cool, of course. 15:40:56 +Exec=xdg-open http://localhost:12345/admin 15:41:08 yeah, so that needs to change to 9090 once poort change goes in 15:41:22 will localhost always work? 15:41:24 andreasn: note there's also a FIXME in the appdata.xml.in where he asks if that should be you for the updatecontact 15:41:30 yep 15:41:43 andreasn: it should... if not, you did something weird to your system :) 15:41:56 puiterwijk: cool, just wanted to make sure 15:42:37 andreasn: could you also update that issue with a comment regarding the port number once #870 is merged? 15:42:47 yep, will do 15:42:57 #action andreasn to add a comment to #1006 once #870 is merged 15:43:17 okay, I also wanted to discuss shortly on #1047/#1048 15:43:21 #link https://github.com/cockpit-project/cockpit/pull/1048 15:43:23 #link https://github.com/cockpit-project/cockpit/pull/1047 15:43:38 mvollmer: that's the server avatar polish, where 1048 says it might replace 1047 15:44:14 haven 15:44:25 t tried #1047, but sounds good 15:44:36 andreasn: am I correct in assuming that, given it has the andreasn-post-review flag set, you will be comparing them and seeing which to merge? 15:44:39 I mean 1048 15:44:43 i think 1048 is better. 15:44:52 :-) 15:45:06 I don't really have an opinion, just wanted to make sure it didn't fall off the radar :) 15:45:17 I'll try them both, but 1048 sounds best as far as I can see 15:46:05 okay. that was the list of PRs that I have seen that needed discussion, did anyone else have any others in mind? 15:46:54 not that I can think of right now 15:46:55 the network changes are pretty big, but I think we don't need to be too careful about them if they go in the right direction. 15:46:56 oh, and we have #1060 off course 15:47:18 mvollmer: what do you mean "the right direction"? 15:47:25 UI wise. 15:47:30 ah, right. 15:47:41 andreasn: could you UI-review the networking changes? 15:47:49 they implement what andreasn and I have cooked up in Brno 15:48:14 and the guideline needs discussion : https://github.com/cockpit-project/cockpit/wiki/HIG:-Dialogs . The reason why we write this is most of our dialogs need subtle polishment(validate input, disable buttons properly, etc), a guideline may help us make them unified 15:48:14 #action andreasn to review the networking PRs UI-wise 15:48:31 #action puiterwijk to review the networking PRs code-wise 15:48:35 is 1060 the right number? it's about RAID 15:49:06 andreasn: yes, but it seems mvollmer had some remarks on it. wasn't sure if he wanted to add more. (althoguh we don't have Philip here unfortunately) 15:49:12 yinzhang: excellent start on the page! 15:49:32 yinzhang, the dialog polish is very important indeed. 15:49:52 I'll try out 1060 15:50:00 what issue was the network stuff? 15:50:18 andreasn: a lot of the PRs that are currently open :-) 15:50:27 all right 15:50:32 andreasn, #1059 15:50:52 #1049, #1040 15:50:54 ok, I'll go through it asap 15:51:25 #info networking changes are #1059, #1049 and #1040 15:51:26 puiterwijk, we should add something about the dialogs to the test day guidelines. 15:51:47 mvollmer: yeah, makes sense 15:52:16 like "most dialogs don't do good input validation. We know, please don't waste time reporting 'obvious' things." 15:52:19 I'm not sure we have a begin for the guidelines yet, though I seem to remember there is one 15:53:02 mvollmer: right. would you be willing to do that, or shall we leave that just remarked for when we know what exact release we are going to have tested? 15:53:17 (given we don't know yet how much of the polish we can get into Alpha) 15:53:23 I can write the text. Where should it go? 15:53:53 mvollmer: our test day page is here: https://fedoraproject.org/wiki/Test_Day:2014-09-16_Cockpit?rd=Test_Day:2014-09-11_Cockpit 15:54:20 #action mvollmer to update the test day page to include warning about error checking in dialogs 15:54:24 #link https://fedoraproject.org/wiki/Test_Day:2014-09-16_Cockpit?rd=Test_Day:2014-09-11_Cockpit 15:54:34 #undo 15:54:34 Removing item from minutes: 15:54:35 thanks, bookmarked. 15:54:36 #link https://fedoraproject.org/wiki/Test_Day:2014-09-16_Cockpit 15:55:03 note to everyone: remember that all the meeting minutes are logged and will be published, so you don't need to write everything down :) 15:55:16 (that's why I'm doing all the stupid link, action, info, etc things :) ) 15:55:51 okay, anyone else has PRs we need to tend to right now? 15:56:35 okay, let's continue for now. 15:56:38 #topic Important issues that need attending 15:56:59 I know we discussed most of the issues during the F21 stuff already, but anyone has any more important issues that need attending to? 15:57:13 doesn't need to be F21, just issues that they hit and want to discuss with the rest? 15:57:44 mvollmer, andreasn, yinzhang: anything? 15:57:45 so one thing with F21 is that our login page says "This is pre-release software. Only use it on a system that you are willing to lose. " 15:57:55 does that recommendation stand? 15:58:08 i say yes. 15:58:17 andreasn: right. I personally think we should keep that in for the Alpha, but should be removed when we go to Beta 15:58:30 agree 15:58:32 although, yeah, it is still very easy to shoot yourself in the foot 15:58:39 yeah 15:58:43 so at least for now, it should stay in, but we should revisit before Beta 15:58:47 sure 15:59:07 #info pre-release warning on login page to be left for Alpha, revisit for Beta 15:59:24 andreasn: did you file an issue for that yet in github? 15:59:31 I'll do that now 15:59:31 the criterion is mostly bugs in Cockpit, not how fool-proof the UI is. 16:00:03 and we just don't know much about the bugs yet. 16:00:15 mvollmer: okay, that makes sense. so that means we should at least wait for the test day 16:00:46 yeah 16:00:53 okay 16:01:00 filed as https://github.com/cockpit-project/cockpit/issues/1075 16:01:10 #link https://github.com/cockpit-project/cockpit/issues/1075 16:01:14 andreasn: thanks :) 16:01:30 #undo 16:01:30 Removing item from minutes: 16:01:44 #link https://github.com/cockpit-project/cockpit/issues/1075 Issue to track pre-release warning on login page 16:02:04 okay, so any other major issues that need attending to right now? 16:02:41 #topic Navigation redesign 16:02:55 andreasn: any updates on this? 16:03:32 not at the moment, I just showed it to LHinson and julim earlier today 16:03:38 I think you had some mockups, right? are those in the wiki or cockpit-design repos? 16:03:47 they are in the design repos 16:04:00 andreasn: got a link? 16:04:17 https://raw.githubusercontent.com/cockpit-project/cockpit-design/master/navigation/navigation.png 16:04:33 #link https://github.com/cockpit-project/cockpit-design/blob/master/navigation/navigation.png new navigation mockups 16:04:51 we need to collect feedback somewhere. We talked about it a bit when we met in person a couple of weeks ago 16:04:57 but I don't remember it all 16:05:15 maybe a feedback session on the wiki page is best? 16:05:29 andreasn: okay, I think this would need discussion o nthe mailing list? 16:05:46 I can post it there, sure 16:05:50 #action andreasn to send an email to the mailing list to solicit feedback on the new navigation 16:06:22 andreasn: wiki page editing is still only for people with write access ,that's why I think this would be better in the mailing list :) 16:06:33 oh, I see 16:07:08 andreasn: how about we discuss the mailing list thread during the next meeting? 16:07:15 sure 16:07:32 #info Navigation redesign to be re-discussed during next week's meeting 16:07:44 When dialog pops up, when we press back key, it close the dialog, and navigates back to the previous page, but i think the dialog should be modal, and should not be affected by other factors unless user close the dialog explicitly 16:08:12 yinzhang, the user is king. 16:08:37 i think we should close the dialog cleanly. 16:08:41 which we might not do. 16:09:46 yinzhang: well, I think that we can't really stop people from hitting the back button. Technically we could disable, but I don't think we should, or that we should still keep the dialog open 16:10:10 yeah, sure 16:10:50 yinzhang: but yeah, closing dialogs neatly might be a good thing to do, as mvollmer said 16:11:06 But I noticed the current implementation behaves differently on chrome and firefox when pressing back key 16:11:19 ah, so that is wrong, I think.. 16:11:34 yeah, the current behavior is mostly accidental. 16:12:20 okay, so that might be worth looking in to. yinzhang, do you have an issue filed on this already? 16:13:23 I think so, one sec 16:15:02 #info yinzhang noticed a bug in the handling of dialog boxes 16:15:49 I modified that, might need work https://github.com/cockpit-project/cockpit/issues/822 16:16:23 yinzhang: okay, could you look into that? 16:16:30 yeah, sure 16:16:31 #link https://github.com/cockpit-project/cockpit/issues/822 16:16:43 #action yinzhang to look further into what's needed for the dialog closing handling 16:17:02 anything else on the Navigation redesign? 16:17:51 andreasn: anything from you? 16:17:52 nope 16:18:01 okay, let's move on :) 16:18:02 #topic Atomic features 16:18:18 so, did we have an overview of features we wanted to implement already? 16:18:30 mvollmer, andreasn: I think you had a meeting with Colin on this, right? 16:18:30 yes, we have https://github.com/cockpit-project/cockpit/wiki/Atomic:-OSTree-Update 16:18:45 I haven't met with colin yet, no 16:18:49 puiterwijk, I didn't. :) 16:18:52 https://github.com/cockpit-project/cockpit/wiki/Atomic:-OSTree-Update 16:19:01 https://github.com/cockpit-project/cockpit/wiki/Atomic:-OSTree-Update 16:19:10 #link https://github.com/cockpit-project/cockpit/wiki/Atomic:-OSTree-Update 16:19:36 andreasn: are those three different links? or all the same? 16:19:36 ups, same link 3 times 16:19:38 https://github.com/cockpit-project/cockpit/wiki/Atomic:-Docker-Storage 16:19:41 okay, thanks :) 16:19:48 #link https://github.com/cockpit-project/cockpit/wiki/Atomic:-Docker-Storage 16:19:50 https://github.com/cockpit-project/cockpit/wiki/Atomic:-Resource-Controls 16:19:59 #link https://github.com/cockpit-project/cockpit/wiki/Atomic:-Resource-Controls 16:20:02 those are the ones 16:20:11 andreasn: thanks 16:20:24 OSTree updates have the most meat on them so far, with use cases and stuff 16:20:33 and I'll write up the two others ASAP 16:20:36 okay. 16:20:45 #action andreasn to fill the other Atomic usecase wiki pages 16:21:03 andreasn: do you think it'd be a good idea to discuss this now, or on the mailing list? 16:21:22 lets wait until we have all 3 for discussion on the list 16:21:36 but please read up on OSTree updates for now if you find the time 16:21:41 okay. can you send a mail to the list once you have them filled? 16:21:56 sure 16:22:34 also, I assume people can just ping you if they have feedback before you finished them all? 16:23:01 yes 16:23:31 #info in case of feedback to the Atomic files, please talk to people in the #cockpit channel 16:23:57 andreasn: I think it might be good to discuss this with Colin at some point, agreed? 16:24:02 yep 16:24:14 will try and set something up with him soon 16:24:22 okay. so do you want me to schedule a meeting with at least the three of us on it? 16:24:39 or do you want to arrange that yourself, so you don't have to bother with me? :) 16:25:16 I can set it up 16:25:22 okay, great 16:25:34 #action andreasn to talk with cwalters on the Atomic stuff in Cockpit 16:25:59 andreasn: do you think you can have anything over this week, or would that be too quick with all the other stuff as well? 16:26:17 depends on when stefw is back too 16:26:33 ideally 16:26:50 but regardless I should sync up with Colin in general, so this week probably 16:26:57 okay. so let's just put it on the agenda for next week, and we can always delay then, agreed? 16:27:20 #info Atomic features revisited next week, but also dependent on return time of Stef 16:27:36 sure 16:27:54 okay, anyone other stuff regarding the Atomic features? 16:28:42 I guess not. let's just move on then to the last item on the agenda 16:28:48 #topic Open floor 16:29:06 okay, so anyone got any last remarks or things they want to discuss? 16:29:25 Two weeks ago, when I was attending intern expo and was asked by other people this issue: https://github.com/cockpit-project/cockpit/issues/1069 16:30:24 yinzhang: hum, that seems very weird yeah. 16:30:42 what's the CLI command for memory usage? 16:30:48 andreasn: that's free -g 16:30:56 (well, -g for gb, -m for mb) 16:31:18 mvollmer: do you have any idea about this? 16:32:21 total used free shared buffers cached 16:32:21 Mem: 15 8 6 1 0 5 16:32:21 -/+ buffers/cache: 2 12 16:32:21 Swap: 14 0 14 16:32:21 and cockpit on my machine is 14.9GB 16:32:32 is cockpit combining used+cached? 16:32:41 andreasn: yeah, it looks like that. 16:33:11 monitor says 2.8 of 15.4gb 16:33:11 although I start to think that cockpit might also just show Total or something 16:33:15 no idea, sorry. this is ooooold code. 16:33:30 hum, okay. then I guess someone needs to revisit that code, as it seems not right at all 16:33:41 https://github.com/cockpit-project/cockpit/issues/1069 16:33:47 puiterwijk, ^ 16:34:03 #link https://github.com/cockpit-project/cockpit/issues/1069 Memory display issue 16:34:18 * mvollmer has to log out now. 16:34:27 mvollmer: okay, thanks for attending to the meeting. 16:34:31 thanks! 16:34:35 later! 16:34:52 yinzhang: would you be able to look at where it goes wrong? 16:34:52 puiterwijk, I can finish https://github.com/cockpit-project/cockpit/issues/1069 later 16:34:59 mvollmer: okay, thanks. 16:35:06 #action mvollmer to look into #1069 16:35:19 mvollmer: just one last question ,if you have one more minute? 16:35:27 #action puiterwijk check review comment on #1069 16:35:38 ok, let's see. 16:35:39 mvollmer: thanks :) 16:35:52 question to all: some quick feedback on this meeting? :) 16:35:53 i get some 'help' with typing here.... :-) 16:36:18 (the little one wants to press the button with the green light. :-) 16:36:48 mvollmer: hah. then I'd say have a good evening, and thanks again for attending. the minutes will be sent to the list 16:36:51 good meeting! 16:36:55 thanks for chairing it! 16:37:06 thanks 16:37:08 i am not sure if I can make time for two per week, though. 16:37:29 right, we might discuss that during the other meeting then. maybe we should just only have this one? 16:37:46 andreasn: any last remarks from you about this meeting? 16:39:22 #info Seems to have been a useful meeting for getting a quick overview on things, let's try it at least one more week and then discuss whether or not to continue 16:39:30 thanks all for coming! 16:39:50 #endmeeting