19:00:09 <wilcal> #startmeeting 19:00:09 <Inigo_Montoya`> Meeting started Thu Apr 17 19:00:09 2014 UTC. The chair is wilcal. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:00:09 <Inigo_Montoya`> Useful Commands: #action #agreed #help #info #idea #link #topic. 19:00:29 <lewyssmith> Good! 19:00:34 <wilcal> Hi ya david 19:00:39 <DavidWHodgins> HiYa 19:00:46 <wilcal> #chair DavidWHodgins 19:00:46 <Inigo_Montoya`> Current chairs: DavidWHodgins wilcal 19:00:50 <tmb> hi all 19:00:51 <wilcal> there you're in charge too 19:00:58 <wilcal> Glad you are here 19:01:07 <wilcal> MrsB does not think she can attend 19:01:33 <lewyssmith> Now we are at least 4. 19:01:48 <DavidWHodgins> Saw the msg a few minutes ago. Standard topics, I guess then. 19:01:53 <DavidWHodgins> #topic Who's new? If you are new then come & meet the team 19:02:04 <wilcal> Yes, well go by the standard format 19:02:23 <DavidWHodgins> So, anyone here who has not been to a qa irc meeting before? 19:03:03 <DavidWHodgins> Doesn't look like it. 19:03:06 <wilcal> Sounds pretty quiet 19:03:08 <DavidWHodgins> #topic Testing updates 19:03:31 <wilcal> I think between us there are a few issues we can "talk" about here 19:03:47 <DavidWHodgins> Go ahead 19:03:58 <wilcal> There are some in the list that have been there a long time 19:04:22 <wilcal> I just gave lightdm a try again and ran into the same thing I saw last time 19:04:30 <lewyssmith> Repeat: they are mostly very obscure. 19:04:59 <wilcal> the NTFS partitions thing has been there way too long 19:05:26 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=12977 19:05:27 <[mbot> Bug 12977: critical, Normal, qa-bugs, ASSIGNED , audit initscript causes ordering cycles at boot leading to dbus+NetworkManager failure incl. KDE plasma applet crashes, auditd 19:05:29 <wilcal> way too long 19:06:11 <DavidWHodgins> As bugfix updates, they've had to take a backseat to all of the security updates. 19:06:42 <wilcal> I've been trying to contribute as best I can to the security stuff 19:06:49 <wilcal> knocked a couple off last weekend 19:07:13 <lewyssmith> The KDE plasma one is wierd because it changes its nature from beginning to end. I looked at it, & was not sure what to actually test. 19:07:27 <wilcal> I'm happy to see you back David as you can help coach me on how to do advisories 19:07:44 <wilcal> A lot of them are that way lewy 19:08:04 <wilcal> we've gotta formulate a plan on how to handle these 19:08:28 <wilcal> Some of these are not bugs their careers 19:08:29 <DavidWHodgins> Which bug is the kde plasma one? 19:08:43 <lewyssmith> 12670 19:09:10 <lewyssmith> Sorry, 12977 19:09:59 <lewyssmith> It is all about 'audit'. 19:10:37 <DavidWHodgins> Ah. I remember that one now. 19:11:05 <lewyssmith> I would do it if I knew what to try. 19:11:28 <DavidWHodgins> Would be good if we can get Albert to help testing that one. 19:12:05 <wilcal> I would accept it if Albert said it was fixed 19:12:11 <wilcal> even though he reported it 19:13:06 <lewyssmith> I think this is a good principle. 19:13:35 <DavidWHodgins> If he confirms it fixes the problem on his system, we still have to test the other arch, to ensure it installs cleanly, and starts ok. 19:14:10 <DavidWHodgins> He reported it for x86_64, so it will still need some testing on i586. 19:14:44 <lewyssmith> Thomas Spuhler has tried it, but his tests had nothing in common with the original complaint. 19:16:07 <DavidWHodgins> systemd startup dropping jobs, really depends on what else you have installed, which does make it harder to duplicate. 19:17:07 <wilcal> Is this a widely used function? Critical to basic operation of M4/KDE? 19:17:27 <DavidWHodgins> When there is a bug we cannot duplicate, we just test that the package installs cleanly, and starts normally. 19:17:31 <DavidWHodgins> no 19:17:50 <lewyssmith> 13014: setup the mail system using amavisd and clamd for virus 19:17:54 <wilcal> Then we should just make sure it installs and starts normally and push it 19:18:09 <DavidWHodgins> audit doesn't actually have anything to do with kde. 19:18:15 <wilcal> "setup the mail system using amavisd and clamd for virus" that's a career not a test 19:18:34 <lewyssmith> Easier said than done. Why it has waited. 19:18:55 <DavidWHodgins> It's because it's causing systemd to drop various jobs, due to dependency loop, that causes the problems to show up in kde. 19:19:11 <DavidWHodgins> :-) 19:19:24 <wilcal> Sometimes not all the time 19:20:47 <lewyssmith> 13014: Again, I would have a go if I knew what it is about. 19:21:33 <DavidWHodgins> Leave that one to me. I'll try and have a go at it over the next couple of days. 19:22:34 <lewyssmith> Or advise me by e-mail. 19:22:58 <lewyssmith> 12496: Still being debated, it seems to me. Can it be greyed? 19:23:00 <wilcal> Are we in good shape to test: 19:23:01 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=13219 19:23:03 <[mbot> Bug 13219: major, Normal, qa-bugs, NEW , systemd stack-based buffer overflow in systemd-ask-password, systemd-208-14.mga5.src.rpm 19:23:22 <DavidWHodgins> I've looked at amavisd in the past, but it's been a long time, and I'll have to research it a bit. 19:23:54 <DavidWHodgins> Does it have a proof of concept? I haven't looked at that one yet. 19:24:26 <lewyssmith> The fix is really simple, so I think just general stability tests are sufficient (ideally booting in a range of different setups - especially on MGA4 where various other "stable release" patches are included 19:24:33 <wilcal> I see a new systemd rpm in my core_testing repo this morning 19:24:54 <lewyssmith> I cited Clin. 19:24:57 <wilcal> so just update to it, reboot and make sure things are working? 19:24:58 <lewyssmith> Colin. 19:25:11 <lewyssmith> wilcal: Seems so. 19:25:31 <wilcal> I should be able to give that a go tomorrow morning for me 19:25:46 <lewyssmith> Me too, or Saturday. 19:26:00 <wilcal> All the others are bugfixes and I try them as best I can contribute 19:26:27 <wilcal> Lets move on and see if Luigi is here 19:26:53 <wilcal> Luigi12_lappy: ping 19:26:53 <DavidWHodgins> #topic Luigi's Roundup 19:26:55 <tmb> I can try to hammer on the systemd one too as I have both mga3 and mga4 live systems here 19:27:07 <wilcal> thks tmb 19:27:50 <wilcal> The number of security bugs is pretty low 19:28:03 <DavidWHodgins> Luigi12_work: ping 19:28:24 <wilcal> As you probably know I have all the active archs in Vbox and updated 19:28:33 <DavidWHodgins> Finally. Thanks for taking care of them while I've been mostly inactive. 19:28:42 <wilcal> Your the best David 19:28:58 <wilcal> I do as best I can surrounded by geniuses 19:29:18 <wilcal> Hello Luigi 19:29:38 <DavidWHodgins> wilcal: Good. If you have the space, it's also a good idea to have all 4 in real installs, seperate from what you use for normal day to day use. 19:30:11 <wilcal> I'm in love with Vbox on M4 it's really nice 19:30:41 <wilcal> blow up the installs as needed :-)) 19:30:52 <DavidWHodgins> On my system, with 16GB of ram, I've had as many as 7 vb guests running all at the same time. 19:30:55 <wilcal> Luigi is pretty quiet 19:31:13 <DavidWHodgins> Doesn't look like he's around today. 19:31:25 <tmb> speaking ov vbox, there vill be a vbox security update on mga4, and I'll fix up the mga3 one too 19:31:42 <wilcal> thks for the heads up tmb 19:32:01 <wilcal> Vbox in Vbox don't work so good :-(( 19:32:01 <DavidWHodgins> Hopefully, it'll work with m3 i586 guests. 19:32:22 <DavidWHodgins> It isn't expected to. :-) 19:32:22 <wilcal> Last Vbox update I built new systems 19:32:24 <tmb> and then there will land full sets of kernel updates on both mga3 and mga4... I already have them built, but I wnat to verify some stuff before handing them over to qa 19:32:40 <wilcal> Removable HD systems make it eas but time consuming 19:32:47 <wilcal> easy 19:33:19 <wilcal> the kernel stuff can be tested in Vbox for me 19:34:05 <DavidWHodgins> I have 4 installs on real hardware, I use for testing things like kernels, vbox, etc. And use a script to backup/restore each, as needed, similar to vbox snapshots. 19:34:17 <tmb> you can test the kernels right now if you want... (otoh I havent rebuilt the kmods yet) 19:34:27 <DavidWHodgins> wilcal: The kernels should be tested both in vb and on real hardware. 19:34:51 <wilcal> I agree David just need an open weekend with lots of Baseball games on TV :-)) 19:35:22 <DavidWHodgins> Yep. Testing all of the kernels is tedious. 19:35:48 <wilcal> On to anything else??? 19:35:55 <DavidWHodgins> tmb: What's the difference between the tmb kernels and the regular kernels? 19:36:26 <wilcal> Specail tmb kernels are not for the feint of heart :-)) 19:36:37 <DavidWHodgins> #topic Anything else? 19:36:40 <lewyssmith> QA are perilously few at present. 19:36:52 <wilcal> I've one thing to present here 19:37:01 <tmb> DavidWHodgins, reiser4 support, BFS cpu scheduler, TuxOnIce suspend/resume support, other stuff tested before landing in core kernel 19:37:33 <DavidWHodgins> tmb: Ok. Thanks. 19:37:38 <DavidWHodgins> wilcal: Go ahead 19:37:40 <wilcal> openssl-1.0.1e-8.2.mga4.i586.rpm dated 7 April 2014 19:37:41 <wilcal> openssl-1.0.1e-1.5.mga3.i586.rpm dated 7 April 2014 19:37:43 <wilcal> All included a heartbleed fix backported from openssl-1.0.1g 19:37:44 <wilcal> true or false? 19:37:52 <DavidWHodgins> True 19:38:12 <wilcal> Kool then I think just this simple thing should be in our Blog 19:38:20 <wikigazer> are there plans to make 1.0.1g (or later)? 19:38:27 <wilcal> tmb? 19:38:44 <tmb> not unless we have to 19:38:49 <wikigazer> k 19:39:01 <DavidWHodgins> wilcal: We don't upgrade versions on stable release, unless the fix is too difficult to backport. 19:39:25 <wilcal> Thks but I don't think that needs to be in the blog just the simple statement 19:39:31 <tmb> libs used in a lot of places we dont want to upgrade as they might trigger a lot of needed rebuilds 19:39:45 <wilcal> that for doktor500 to point to on the forum 19:40:15 <tmb> they are supposed to be api/abi stable, but sometimes there are issues 19:40:26 <wikigazer> stable is good 19:40:44 <wikigazer> stick with stable 19:41:06 <wilcal> The heartbleed issue(s) are still screaming through the press over here 19:41:32 <wilcal> David the Canadian Tax department shut down over it??? 19:41:37 <wikigazer> yes, it's really huge. It could have effects for years with fallout from certificate issues 19:41:41 <tmb> yeah, and we haven't seen the full fallout of it yet 19:42:00 <wilcal> We just need a simple statement easy to understand to point to 19:42:03 <wikigazer> top marks to mageia for getting the fix out quick 19:42:28 <wilcal> Mageia reacted in about 4-days 19:42:29 <DavidWHodgins> Yes, but they are back up now, and caught the guy who was trying out the exploit on the system. 19:42:34 <tmb> and what many people dont realize is that you can abuse on both server and client side 19:42:44 <wikigazer> yes 19:43:03 <DavidWHodgins> Yes, a malicious server can exploit it. 19:43:47 <wikigazer> https://twitter.com/lennyzeltser/status/455405328260399104 19:43:48 <[mbot> [ Twitter / lennyzeltser: 3 excellent explanations of ... ] 19:44:58 <DavidWHodgins> Ok. Anything else before we close the meeting? 19:45:09 <lewyssmith> Repeat: we are too few. 19:45:14 <wilcal> I'm done thank you all for a very productive meeting 19:45:27 <DavidWHodgins> :-) 19:45:36 <wilcal> waiting bugs list is slowly coming down 19:45:41 <wilcal> few security bugs 19:45:56 <wilcal> I opened it you get to close it David 19:46:06 <DavidWHodgins> #endmeeting