20:00:30 <DavidWHodgins> #startmeeting 20:00:30 <Inigo_Montoya> Meeting started Thu Feb 14 20:00:30 2019 UTC. The chair is DavidWHodgins. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:30 <Inigo_Montoya> Useful Commands: #action #agreed #help #info #idea #link #topic. 20:00:38 <MageiaTJ> It's OK. Sometimes I don't know where I am, either. 20:00:40 <DavidWHodgins> #chair MageiaTJ wilcal 20:00:40 <Inigo_Montoya> Current chairs: DavidWHodgins MageiaTJ wilcal 20:00:53 <wilcal> hello tmb 20:00:59 <DavidWHodgins> #topic * Who's new? - If you are then come and say hello 20:00:59 <tmb> hi 20:01:10 <DavidWHodgins> HiYa tmb 20:01:29 <DavidWHodgins> EzequielPartida: Have you been to a qa team irc meeting before? 20:02:05 <DavidWHodgins> Shy I guess. :-) 20:02:12 <kekePower> or afk 20:02:21 <DavidWHodgins> * M7 Beta 2, Status & Testing 20:02:54 <tmb> you forgot topic 20:03:16 <wilcal> i did a netinstall to a blank drive on an nvidia platform yesterday and that went well 20:03:19 <DavidWHodgins> I've asked council to decide whether to hold m7 till gnome 3.32 can be included or not. Meant to ask them last week, but forgot till today. 20:03:25 <DavidWHodgins> tmb: Which topic? 20:03:32 <wilcal> Updated my Intel platform this morning and that's fine 20:03:42 <MageiaTJ> The topic command 20:03:45 <tmb> DavidWHodgins, irc command 20:04:00 <wilcal> can i do it 20:04:02 <wilcal> #topic M7 Beta 2, Status & Testing 20:04:06 <wilcal> there you go 20:04:13 <DavidWHodgins> Oops. Thanks. 20:04:37 <bri_an> Sounds like at this point waiting on council to decide on g 3.32 20:04:45 <wilcal> I'm back to testing apps rather then getting it to work at all 20:04:50 <MageiaTJ> Bug 24060 looks to be fixed to me. 20:05:10 <wilcal> Ya the flashing is gone for me too but another issue i need to share here 20:05:22 <DavidWHodgins> My preference is to go ahead and release beta 2 as is for wider testing, and add a beta 3 when the gnome update is available in cauldron 20:05:38 <wilcal> I completely agree with that 20:06:00 <MageiaTJ> Need another round of isos, don't we? 20:06:19 <wilcal> Yes 20:06:24 <bri_an> yes - then I'll resume 20:06:31 <wilcal> before the rc 20:06:49 <tmb> yeah, and we still miss systemd update, and mesa 19 is some weeks out (we are at rc4 for now, but more fixes are needed) 20:06:53 <wilcal> The repo is still very dynamic 20:07:11 <wilcal> 250 updates this morning for me 20:07:14 <DavidWHodgins> I've been debating whether to ask Martin to rebuild the beta 2 iso images now, and release them in a couple of days, or go ahead and release them as is. 20:07:30 <DavidWHodgins> Opinions? 20:07:57 <kekePower> DavidWHodgins: we need to wait for kde apps 18.12.2 to be moved from updates_testing before building new iso's 20:08:19 <tmb> I think it would be nice to sync up to get the big QT bugfix round & glibc updates on it 20:08:21 <DavidWHodgins> kekePower: I thought that had been approved already. 20:08:39 <MageiaTJ> Suppose I do a Plasma install from the iso, get all the updates, and see what happens before I get back to you on that? 20:08:43 <kekePower> last I saw was KF 5.55 approved 20:09:02 <tmb> DavidWHodgins, kde apps only just get fully built in testing 20:09:37 <DavidWHodgins> Yes, but I thought there was a message on the dev mailing list today about going ahead and moving them 20:10:05 <tmb> no, it was only announced ready for tests in testing 4 hours ago 20:10:06 <EzequielPartida> Sorry I was away... Hello Everyone!! 20:10:23 <kekePower> welcome EzequielPartida 20:10:42 <DavidWHodgins> From Nicolas, Feb 11th at roughly this time of day. 20:10:57 <kekePower> DavidWHodgins: that was KF 20:11:57 <DavidWHodgins> kekePower: Ah. Sorry. Mixing up kde frameworks and the rest of kde 20:12:08 <tmb> yeag, it was daviddavid that announced the kde apps today 20:12:15 <kekePower> with a gazillion packages, that easy! 20:13:09 <DavidWHodgins> The split of kde into kf5 and kde-plasma keeps confusing things for me. :-) 20:13:20 <tmb> but as cauldron is a moving target we just need to decide a cutoff point and go with that 20:13:32 <kekePower> and Plasma 5.15 was recently annouced, so I guess we'll see it updated between b2 and b3 20:13:46 <DavidWHodgins> Anyway backing up the topic to who's new, for a minute. 20:13:56 <MageiaTJ> EzequielPartida: Welcome. If you are new, we'll give you a chance to introduce yourself in a few minutes. 20:14:06 <EzequielPartida> Hello Everyone!! 20:14:18 <DavidWHodgins> EzequielPartida: Welcome to the qa team irc meeting. Please introduce yourself, and everyone else too. 20:14:38 <EzequielPartida> Well as you already know.. my name is Ezequiel Partida (Ezekiel in English) 20:14:58 <EzequielPartida> I am from Tijuana Mexico.. in the border with San Diego California U.S.A 20:15:16 <EzequielPartida> I've been using linux since Mandriva 2006 20:15:24 <DavidWHodgins> I'm David (Dave) Hodgins, leader of the qa team. I'm in London Ontario, Canada. 58 Years old, with a background in ibm mainframe software development 20:15:56 <bri_an> I'm Brian from Illinois, USA (Near St. Louis). 20:16:12 <kekePower> I'm kekePower online, from Norway and fam and friends calls me Stig 20:16:13 <bri_an> I've been using Linux since 2006 or 7 myself 20:16:19 <EzequielPartida> I studied Computer programing on the C=64 them AmigaOS.. and work on the spanish translation of Version 3.5 and 3.9 as well as a well know DTP program calle Pagestream. 20:16:25 <wilcal> Wilcal, Bill, now living just north of Sin City USA ( Las Vegas ) 20:16:47 <EzequielPartida> I do mostly network admin since it' s been a while since I last did programming . 20:16:54 <kekePower> been using Linux since Slackware 3 (ca 1996) 20:17:06 <MageiaTJ> I'm Tom Andrews, aka TJ. I'm 69, and a farmer in Central New York State, near Syracuse. I've been using Linux since Mandrake 8.2, in 2002. 20:17:15 <wilcal> I have been to TJ many times but my favorite is Ensenada 20:17:46 <wilcal> Hussongs He'll know what that is 20:18:07 <EzequielPartida> wilcal ...let me know when you come to TJ and I'll take you to the best Tacos in town. 20:18:16 <wilcal> Will do 20:18:18 <tarazed> Len Lawrence aka tarazed, 80 years old, retired from Royal Observatory in Edinburgh (Scotand) 20:18:21 <tarazed> In QA for maybe five years - Linux since Redhat 0.? and Mandrake7.2. Coding for 55 years. 20:18:24 <tarazed> Welcome aboard. 20:18:33 <EzequielPartida> Thank you tarazed 20:18:42 <EzequielPartida> thank you everyone. 20:19:17 <kekePower> I just love the diversity of this wonderful group :-) 20:19:51 <MageiaTJ> Oh, yes, I've been with Mageia from the beginning, joined QA, what five years ago? Now a deputy leader of QA. 20:19:51 <wilcal> TJ where it's probably raining today 20:19:55 <EzequielPartida> BTW wilcal.. Tijuana is now the Beer Capital of México..So... ;-) 20:20:05 <EzequielPartida> Yes... a lot of rain. 20:20:08 <DavidWHodgins> lol 20:20:43 <wilcal> It's really really good to have someone from Mexico 20:21:05 <kekePower> yes, then I can practice my written spanish :-) 20:21:16 <DavidWHodgins> Let's get back to m7 status now. 20:21:23 <EzequielPartida> I could probably as Miguel de Icaza to join. lol. 20:21:38 <EzequielPartida> thank you again everyone 20:21:45 <MageiaTJ> Oh, dear. I've forgotten about 99.9% of the Spanish I learned in high school. 20:21:46 <DavidWHodgins> I've just sent an email to Martin asking him to rebuild the beta 2 iso images with the latest updates 20:21:51 <wilcal> lemme post my question on M7 performance 20:22:09 <wilcal> So for me on my Intel platform the boot to Plasmadesktop is fine but getting things to work on the desktop sometimes takes many minutes. After things settle down then everything responds normally. Kinda like everything has to go to cache before it really gets going. 20:22:33 <kekePower> wilcal: first boot or every time? 20:22:41 <wilcal> On the nvidia platform and Vbox client I don't see it 20:22:50 <wilcal> Yes, first boot is painful 20:23:03 <DavidWHodgins> EzequielPartida: To get access to the qa testing versions of the iso images, add you info to https://wiki.mageia.org/en/QA_ISO_testers and I'll send you the password etc. 20:23:39 <wilcal> I'm kinda hoping one of these kde/qt/plasma updates will make it all go away 20:23:44 <EzequielPartida> I did my installation from Mageia-7-beta1-x86_64 and it work great 20:23:51 <EzequielPartida> but Beta 2 it fail 20:23:52 <DavidWHodgins> wilcal: Could be the indexing. akonadi and friends 20:23:57 <kekePower> wilcal: I remember I used to wait to log in to Windows when I was using it, to give it time to "settle" 20:24:04 <wilcal> Kinda feels like that 20:24:21 <wilcal> "settle" is a great word for all of this 20:24:33 <wilcal> is it something that needs a bug? 20:24:49 <kekePower> _I_ don't think so 20:24:55 <DavidWHodgins> I'm guessing that is on a spinning rust drive. On an ssd drive, it's barely noticeable. 20:25:45 <wilcal> The drive I'm using ks a 2TB Seagate drive so it's pretty quick 20:25:53 <DavidWHodgins> Not much point in a bug. It's the way akonadi is designed to work. I always disable it on my main system, and ensure the test installs do not have read access to most of my system. 20:26:04 <wilcal> But, I can try an ssd on the intel platform 20:26:24 <wilcal> How do you disable it 20:26:39 <tarazed> Good question 20:26:55 <kekePower> but this is the first impression a user gets when installing it, so I shouldn't be neglected 20:27:04 <wilcal> Is it only on Plasma 20:27:05 <kekePower> s/I/it 20:27:26 <wilcal> Thanks all, so this is an issue 20:27:30 <MageiaTJ> EzequielPartida: The first round of beta2 test isos were messy. Round 2 was better, and round 3 should be better yet. 20:27:36 <DavidWHodgins> This is old, but the akonadi parts work. http://www.linux-databook.info?page_id=3728 20:27:37 <[mbot> [ DataBook® for Linux – Linux Technical Data ] 20:27:38 <[mbot`> [ DataBook® for Linux – Linux Technical Data ] 20:27:44 <kekePower> the process could prolly get a better "nice" value 20:28:06 <wilcal> After first boot and first setup it gets quicker 20:28:17 <DavidWHodgins> It's hard drive contention that it causes problems with. That's why an ssd drive speeds things up so much. 20:28:39 <kekePower> tmb: is this something that can be helped with a kernel setting? 20:28:46 <wilcal> I do see a LOT of HD activity during the process 20:28:59 <DavidWHodgins> On a low ram or slow cpu, those would also be factors, but it's mostly hard drive access. 20:29:35 <DavidWHodgins> kekePower: Not really. It's reading every file the user has read access to, for indexing. 20:29:42 <EzequielPartida> I have that same Hi HD activity on my Intel i5 4gb Ram. 20:30:07 <DavidWHodgins> It even includes files available over sshfs, which can be painfully slow. 20:30:07 <wilcal> So we're seeing the same thing i think 20:30:11 <kekePower> I'm using the joeghi kernel and it feels quicker than the default 20:31:09 <tmb> the service starting akonadi indexing should run it under "nice control" so it wont hog all resources 20:31:16 <DavidWHodgins> Easiest way to deal with it is to log in to kde the first time, and just leave the system running for a few hours, or how ever long it takes for the hard drive activity to settle down. 20:31:25 <kekePower> can run a long compilation of something and watch a YT video without it stuttering 20:31:39 <wilcal> gracious 20:31:55 <DavidWHodgins> tmb: That doesn't matter. On a spinning rust drive the constant head movement slows down all other file access 20:32:34 <kekePower> so even at nice 20 it will still hog all resources? 20:32:40 <DavidWHodgins> Since ssd drives don't have heads to move, they don't show the problem. 20:32:44 <DavidWHodgins> kekePower: Yes 20:33:09 <kekePower> daym 20:33:31 <DavidWHodgins> It isn't hogging the cpu or ram. It's hogging drive access. Once it starts a read i/o, the system has to wait for that i/o to fininsh, no matter what the nice value is. 20:33:32 <wilcal> So when I did my M7 test install yesterday that was to an ssd drive. Didn't see the problem 20:33:49 <tmb> well there are several ways of restricting resources, but basically akonadi is badly designed... 20:34:02 <DavidWHodgins> Correct. Or at least it's much less of an impact 20:34:03 <kekePower> perhaps there is a way to make it start at an odd hour instead of immediately? 20:34:06 <wilcal> If I boot to Mate I should not see it, correct? 20:34:46 <DavidWHodgins> If akonadi is installed, it will probably start even if you log in to another desktop environment 20:34:55 <wilcal> Good discussion on this 20:35:29 <MageiaTJ> And several KDE apps require it. 20:35:57 <MageiaTJ> Or at least use it. 20:36:19 <kekePower> this must be mentioned in the release notes or something 20:36:30 <wilcal> I have noticed that as I use the install the wait times get shorter 20:36:31 <DavidWHodgins> Anyway, Martin or I will update the qa discuss list when the updated beta 2 iso images are available. I'll keep the list updated when council decides whether to hold release to have gnome 3.32 or not. 20:36:47 <wilcal> sounds like a plan 20:36:57 <DavidWHodgins> kekePower: I think it was, though for Mageia 5 iirc 20:37:33 <kekePower> okay 20:38:05 <DavidWHodgins> Regarding bug 24060, is it fixed? 20:38:07 <[mbot> Bug: ['Qt 5.12 & Plasma 5.14.4 & KF 5.53 update causes Plasma desktop screen to flicker when OpenGL compositing background is used (when using intel or nouveau video drivers)', 'NEW', 'KDE maintainers'] https://bugs.mageia.org/show_bug.cgi?id=24060 20:38:07 <[mbot`> Bug: ['Qt 5.12 & Plasma 5.14.4 & KF 5.53 update causes Plasma desktop screen to flicker when OpenGL compositing background is used (when using intel or nouveau video drivers)', 'NEW', 'KDE maintainers'] https://bugs.mageia.org/show_bug.cgi?id=24060 20:38:09 <[mbot`> [ 24060 – Qt 5.12 & Plasma 5.14.4 & KF 5.53 update causes Plasma desktop screen to flicker when OpenGL compositing background is used (when using intel or nouveau video drivers) ] 20:38:09 <[mbot> [ 24060 – Qt 5.12 & Plasma 5.14.4 & KF 5.53 update causes Plasma desktop screen to flicker when OpenGL compositing background is used (when using intel or nouveau video drivers) ] 20:38:22 <DavidWHodgins> Two mbots running. :-) 20:38:38 <wilcal> The flicker is definitely gone 20:38:54 <DavidWHodgins> Good 20:38:59 <wilcal> but gets masked by the Akonadi thingy 20:39:07 <wilcal> :-( 20:39:29 <DavidWHodgins> I have amd graphics here, which wasn't affected. 20:40:05 <DavidWHodgins> #topic * Testing updates - Any questions or problems 20:40:44 <wilcal> where are we at with 20:40:45 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=24331 20:40:47 <[mbot`> [ 24331 – Update request: kernel-4.14.98-1.mga6 ] 20:40:47 <[mbot> [ 24331 – Update request: kernel-4.14.98-1.mga6 ] 20:40:59 <DavidWHodgins> The reason lxc and opencontainers-runc were such a high priority, is described in https://blog.dragonsector.pl/2019/02/cve-2019-5736-escape-from-docker-and.html 20:40:59 <MageiaTJ> Working well here. 20:41:00 <[mbot> [ Dragon Sector: CVE-2019-5736: Escape from Docker and Kubernetes containers to root on host ] 20:41:00 <[mbot`> [ Dragon Sector: CVE-2019-5736: Escape from Docker and Kubernetes containers to root on host ] 20:41:14 <wilcal> tmb mentioned a change 20:41:42 <DavidWHodgins> Gives guest installs, where the user has root access, root access on the host. Worst possible type of bug for a cloud provider 20:42:09 <tmb> wilcal, yes, new kernels should land tomorrow... they are going through upstream review/tests currently 20:42:26 <wilcal> ready by this weekend 20:42:39 <wilcal> I have some time then 20:42:41 <bri_an> I should have some bandwidth to help on kernel 20:42:56 <DavidWHodgins> I'm guessing the new update is a kernel level fix for the container breakout bugs 20:44:45 <bri_an> ok 20:44:53 <DavidWHodgins> Regarding bug 22570, that one shows why we should always use the qarepo tool, to ensure updates that are not included in the current test, are not also installed. 20:44:55 <[mbot> Bug: ['zziplib new security issue CVE-2018-6381, CVE-2018-6484, CVE-2018-654[0-2], CVE-2018-6869, CVE-2018-772[5-7], CVE-2018-16548, CVE-2018-17828', 'NEW', 'QA Team'] https://bugs.mageia.org/show_bug.cgi?id=22570 20:44:55 <[mbot`> Bug: ['zziplib new security issue CVE-2018-6381, CVE-2018-6484, CVE-2018-654[0-2], CVE-2018-6869, CVE-2018-772[5-7], CVE-2018-16548, CVE-2018-17828', 'NEW', 'QA Team'] https://bugs.mageia.org/show_bug.cgi?id=22570 20:44:57 <[mbot> [ 22570 – zziplib new security issue CVE-2018-6381, CVE-2018-6484, CVE-2018-654[0-2], CVE-2018-6869, CVE-2018-772[5-7], CVE-2018-16548, CVE-2018-17828 ] 20:44:57 <[mbot`> [ 22570 – zziplib new security issue CVE-2018-6381, CVE-2018-6484, CVE-2018-654[0-2], CVE-2018-6869, CVE-2018-772[5-7], CVE-2018-16548, CVE-2018-17828 ] 20:46:13 <DavidWHodgins> Any questions, comments, or problems with the other updates ready for testing at https://madb.mageia.org/tools/updates ? 20:46:15 <[mbot> [ Mageia App Db - Current Update candidates ] 20:46:15 <[mbot`> [ Mageia App Db - Current Update candidates ] 20:47:00 <wilcal> not from me 20:47:13 <MageiaTJ> The mythtv update seems to have a host of problems. 20:48:00 <DavidWHodgins> There are not many people who use it as few have the hardware for it. Those who do use it, need it though. 20:48:11 <DavidWHodgins> I don't have a tv card here. 20:48:28 <wilcal> I tried it some time ago and the tv part of it did not work at all 20:48:37 <wilcal> About a year ago 20:48:46 <wilcal> and it was kinda klunky 20:49:22 <wilcal> I found Kaffine to work and be much easier to use 20:50:09 <DavidWHodgins> The last comment from Alan Richter indicates it's ok, and as it's a config file, that shouldn't be overwritten by the update, so should be ok for existing users. 20:50:12 <wilcal> And that with the most popular USB tv card 20:51:22 <DavidWHodgins> I'll post a question to the users in that bug report later today, and then decide whether to push it as is or not. 20:51:44 <DavidWHodgins> Last topic then ... 20:51:47 <DavidWHodgins> #topic * Anything else? 20:51:52 <wilcal> not from me 20:52:15 <DavidWHodgins> Looks like countdown time then. 20:52:18 <DavidWHodgins> t - 5 20:52:18 <MageiaTJ> The sun is out and it's getting warmer! 20:52:22 <DavidWHodgins> 4 20:52:25 <DavidWHodgins> 3 20:52:27 <DavidWHodgins> 2 20:52:28 <wilcal> bye all 20:52:29 <DavidWHodgins> 1 20:52:35 <DavidWHodgins> Thanks for coming everyone 20:52:39 <DavidWHodgins> #endmeeting