15:01:48 #startmeeting Fedora QA 15:01:48 Meeting started Mon Jun 1 15:01:48 2015 UTC. The chair is roshi. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:48 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:56 #topic Roll Call 15:02:01 * pschindl_wfh is here 15:02:04 roshi: thanks 15:02:05 * tflink is here 15:02:16 (I found the SOP now) 15:02:17 * roshi is glad his "make it up as he goes" skills at least have danofsatx fooled :p 15:02:18 * danofsatx is kinda sorta here 15:02:39 oh, I'm not fooled ;) 15:03:43 hehe 15:03:59 #topic Previous Meeting Follow-Up 15:04:40 #link http://meetbot.fedoraproject.org/fedora-meeting/2015-05-04/fedora-qa.2015-05-04-15.01.html 15:04:48 logs from last meeting ^^ 15:05:10 we had three 15:05:26 all of which had to do with adam :) 15:05:38 anyone have any info on FCoE testing? 15:05:40 I think the only one relevant is the council thing 15:05:47 which is a next topic 15:05:56 roshi: we released, so... 15:06:03 no longer needed :) 15:06:11 I was being facetious :p 15:06:16 I'm afraid nobody tested it 15:06:35 #topic Council Report 15:06:40 I *may* be able to secure a resource that can do the FCoE testing for F23+ 15:06:58 he's in the process of moving, so was unavailable for F22 15:07:05 looks like danofsatx is handling the report - you ready for it danofsatx ? 15:07:13 that'd be good for F23 15:07:32 actually, I wanted to address that. Due to recent....events....I am unable to perform the required research for this report. 15:07:43 so adamw volunteered to do the council report, but the last action item says danofsatx would do it, so I'm a bit confused 15:07:57 adamw already wrote up some points, quite a lot of them, into the test list 15:08:00 when is it? 15:08:05 June 8th 15:08:06 Jun 8 15:08:08 when is the report due? 15:08:13 June 8th 15:08:18 so we need to find someone else to do it :) 15:08:19 it will be a teleconference over google hangouts, it seems 15:08:35 I assume it'll be the same as the last 2 or so council meetings 15:09:08 I was willing to step up for adam/danofsatx for an irc conference, I'm not that thrilled about a video conference, I don't enjoy appearing in public :) 15:09:25 but lot of info should be already written up by adam 15:09:32 hangouts is an odd choice to me... 15:09:43 if somebody is willing to go there, I'll be happy to help the person finalize the notes 15:09:55 * tflink can do it if needed 15:10:36 * danofsatx votes for sgallagh - he did fine last week ;) 15:11:08 * satellit late listening 15:11:14 does anybody *want* to do it, or is everyone pretty much "if needed?" 15:11:26 I believe it's the latter :) 15:11:38 danofsatx: Not volunteering, thanks 15:11:39 i suspect that I'll have to get out the windows box for using hangouts, though :-/ 15:11:43 so do I 15:11:52 it should work on linux 15:11:53 it works well enough in chrome 15:12:02 haven't tried firefox with it 15:12:02 there's a plugin for firefox, or plugin-less in chrome 15:12:05 * tflink doesn't have chrome on any of his machines 15:12:14 I wanted to do it, but circumstances have drastically changed 15:12:16 ah, didn't know it worked with firefox now 15:12:26 chrome in a docker container manages to hold it's RAM gobbling in check 15:12:28 danofsatx: no worries, stuff happens :) 15:13:27 I've said that I'd do it - any other reluctant volunteer-ish people? 15:13:47 it looks like I need to fill in some holes in adam's email anyways 15:14:33 well, I was going to say I can do it if needed 15:14:47 but I'm so far back in the queue now I didn't see a reason to :p 15:14:55 we can draw matches or something :) 15:15:02 roshi: I'm not going to fight you for it :-P 15:15:08 nor I :) 15:15:23 I have a face for radio - which is why I like IRC :p 15:15:32 everybody loves meetings and bureaucracy, it seems 15:16:07 I don't think that video is necessary, btw. mainly audio matters, I think 15:16:11 tbh I think it's the video bit :p 15:16:14 yeah 15:16:30 i think the idea is to make the meetings more accessible to folks who don't like text-only 15:16:47 and shorter, probably 15:17:29 * tflink proposes that kparal, roshi and tflink draw straws/fight over who's doing the presentation outside of the meeting 15:17:41 works for me 15:17:51 ok 15:17:56 unless I missed another reluctant person 15:18:08 quake tournament? 15:18:09 hehe 15:18:18 * pingou put back the pop-corn to the kitchen 15:18:31 well, onto the open floor then :) 15:18:35 \รณ/ 15:18:36 #topic Open Floor 15:18:47 anyone have something for open floor? 15:19:04 just a heads up, fedup seems to be deprecated for F23, according to devel@ list 15:19:13 other than the deathmatch over presenting to the council? 15:19:14 might be of interested to people around here 15:19:15 * tflink ducks 15:19:39 how do we get to f23 for testing? no fedup yet? 15:19:43 i need to catch up on that thread, are there any concrete plans for fedup's replacement? 15:19:54 tflink: systemd's offline updates 15:20:32 wwoods was short in the description, but he claims it will be much simpler than the current solution 15:20:33 at the very least, I'd like to see that transition work better from a testing POV than the preupgrade->fedup transition 15:20:34 I hate offline updates 15:20:49 but that wouldn't take much 15:21:05 tflink: there is already a dnf plugin 15:21:28 might be worth testing, but if it's also used in gnome-applications then that's another path to test 15:21:42 then we're already in better shape - IIRC, fedup wasn't even functioning correctly until after F18 beta freeze 15:22:00 and another interesting news, liveusb-creator gets a new facelift. I already commented on devel@ that it also needs to change the plumbing (dd mode by default, instead of cp mode) it they really want to fix the most problematic issues 15:22:06 and the fedup DNF plugin already works: https://github.com/wgwoods/dnf-plugin-fedup 15:23:21 the conversation with the DNF maintainers makes it sound like they'd prefer to implement the Offline Updates spec in DNF (and/or dnf-plugins-core) itself 15:23:22 sounds like a good thing to start poking at :) 15:23:23 wwoods: welcome. does that mean that we'll be able now to easily do something like distro-sync? 15:23:35 kparal: yes. already supported by fedup-dnf-plugin 15:24:15 wwoods: great. and will it be performed by dnf, instead of core rpm, right? so that information like user-installed leaf packages will not be lost on upgrade 15:24:20 kparal: correct 15:24:24 awesome news 15:24:28 what's the catch? 15:24:36 kparal: it works ;-) 15:24:40 there's always a catch... 15:25:20 the catch? nothing really important - we're throwing out the upgrade.img 15:26:00 that seems like another benefit. it made testing more difficult 15:26:00 so things that were theoretically possible to do during the upgrade (like, say, filesystem migrations) aren't possible *before* the upgrade 15:26:08 but we never did them that way 15:26:50 the only thing we did do like that was /usrmove, and that got handled by a dracut scriptlet in the post-upgrade system 15:27:23 ok, sounds great, then 15:27:36 agreed, sound good to me 15:27:37 so yeah, we've learned a bunch of lessons about how to do upgrade - fedup was designed to work like anaconda used to, and it turns out we inherited a lot of its problems 15:27:58 so fedup2 (or whatever you want to call it) throws away the unnecessary complexity 15:28:20 do we want to get someone started on looking at the upgrade testcases? 15:28:42 rather, is someone interested in starting to look at them? 15:29:04 * danofsatx isn't taking on any action items at this time 15:29:30 wwoods: btw, is the dnf plugin packaged already? 15:29:34 * tflink figures that the earlier we look at it, the earlier we find any potential problems and we aren't dealing with last-minute crap during F23 beta 15:29:56 pingou: no, because I'm not certain that plugin will be the final form 15:30:10 it might instead be in upstream dnf-plugins-core, or upstream DNF itself 15:30:10 +1 for testing early 15:30:13 wwoods: could be merged into dnf-* later with a Provides: 15:30:50 but it will definitely work the same way 15:33:29 pingou: nah. "git clone ... ; sudo make install ; sudo systemctl enable fedup-system-upgrade.service" is fine for a pre-alpha prototype 15:34:17 wwoods: might do that :) 15:34:25 #info fedup will be depreciated in F23 15:34:34 pingou: also the plymouth display stuff doesn't work without patching DNF, so I'm not going to package something that can't work correctly 15:34:47 #info Upgrades will be handled by a plugin in DNF 15:34:51 ^^ that sound right? 15:34:53 wwoods: good point 15:35:13 roshi: sure, DNF plugin or extension 15:35:39 #undo 15:35:39 Removing item from minutes: INFO by roshi at 15:34:47 : Upgrades will be handled by a plugin in DNF 15:35:46 #info Upgrades will be handled by a plugin in DNF or an extension 15:36:22 really it's more like: DNF is probably going to get support for handling Offline Updates, and the new fedup will use that to do system upgrades 15:36:56 well, we can test it when there's something ready for us to test 15:37:00 indeed 15:37:07 :) 15:37:22 thanks for the info wwoods 15:37:35 sure! just ask if you have other questions 15:38:00 anyone have anything else on F23 upgrades or for Open Floor? 15:38:06 or should I set the fuse? 15:38:36 nothing else here 15:39:03 * roshi sets the fuse 15:39:04 3... 15:41:55 2... 15:42:00 1... 15:42:04 thanks for coming folks! 15:42:07 #endmeeting