20:03:33 <DavidWHodgins> #startmeeting
20:03:33 <Inigo_Montoya`> Meeting started Thu Feb 23 20:03:33 2017 UTC.  The chair is DavidWHodgins. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:03:33 <Inigo_Montoya`> Useful Commands: #action #agreed #help #info #idea #link #topic.
20:03:36 <DavidWHodgins> #chair wilcal lewyssmith
20:03:36 <Inigo_Montoya`> Current chairs: DavidWHodgins lewyssmith wilcal
20:03:50 <DavidWHodgins> #topic Who's new? - If you are then come and say hello!
20:04:06 <DavidWHodgins> Anyone here who has not been to a qa team irc meeting before?
20:04:17 <tarazed> Mornin' all
20:04:39 <DavidWHodgins> #topic Mageia 6 - New x86_64-DVD (mageia6-sta2, not mageia6-pretesting)
20:05:04 <lewyssmith> Welcome Thomas.
20:05:05 <tarazed> Just finished dumping it to usb
20:05:11 <wilcal> ya the one I'm using today is the pretesting version
20:05:11 <DavidWHodgins> As per the subject, there is a new x86_64-DVD to test.
20:05:21 <DavidWHodgins> There are two versions on rabbit
20:05:52 <azziam> I'm downloading the new DVD 64 now.  labarhon mentioned a regression of bug 18361.  ??
20:05:53 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=18361 normal, High, kde, REOPENED , In Plasma, the "Add Panel" function makes Plasma to crash.
20:06:11 <ennael> neoclust is on it as I could see
20:06:13 <DavidWHodgins> The pre-testing one is from yesterday, while the mageia6-sta2 is from today
20:06:51 <DavidWHodgins> Also, that won't stop sta2 from being released. Please concentrate on the partitioning testing
20:07:14 <nathan__> ok
20:07:23 <DavidWHodgins> The only thing holding sta2 from public release is the partitioning problems. All other bugs can be fixed after sta2 is released
20:07:34 <neoclust> ennael: i will test but this is not a blocker for sta
20:07:35 <neoclust> 2
20:07:38 <wilcal> including the Nvidia nightmare
20:07:41 <lewyssmith> It should be announced on qa-discuss for wider testing..
20:07:58 <neoclust> ennael: btw i reproduce bug 18361
20:08:17 <lewyssmith> That is not serious.
20:08:26 <DavidWHodgins> ennael: Can you delete the pre-testing version?
20:08:42 <ennael> lewyssmith: wht should be announced ?
20:09:02 <DavidWHodgins> ennael: The presense of two versions on rabbit
20:09:16 <wilcal> as long as you don't install the Nvidia driver your fine
20:09:34 <ennael> sorry I'm lost
20:09:41 <lewyssmith> ennael: Your newest classic ISO. It is in sta2, and way past pre-testing needs.
20:09:46 <DavidWHodgins> wilcal: Make sure that bug is marked as being for errata
20:09:52 <ennael> New DVD64 for classical isos
20:09:55 <ennael> on qa-discuss
20:10:14 <ennael> 13:36
20:10:16 <wilcal> There's like 10 bugs against Nvidia whatever that have been generated this year alone
20:10:23 <DavidWHodgins> ennael: Currently there is a dvd64 in pretesting and one in mageia6-sta2
20:10:40 <DavidWHodgins> The pre-testing one is from yesterday, the other from today
20:10:44 <wilcal> We'll have to settle on one as being the major defect
20:10:46 <azziam> I'll be testing in a while with and without nVidia 340 driver.
20:11:10 <wilcal> Ok I am working with yesterdays pretest. Are they essentially the same?
20:11:34 <ennael> wilcal: some more fixes on stage2
20:11:39 <azziam> BTW, The mentioned bug evidently disappears with a reboot.
20:11:40 <ennael> one I guess
20:12:27 <DavidWHodgins> ennael: Thanks for deleting the pretesting one
20:12:53 <DavidWHodgins> I'm just syncing the latest one now.
20:12:54 <tjandrews> We'd probably be less confusing if we discussed one thing at a time.
20:13:09 <DavidWHodgins> :-)
20:13:16 <ennael> would be nice indee
20:13:17 <ennael> d
20:13:48 <DavidWHodgins> Ok. The version issue for the x86-64 dvd has been sorted
20:14:30 <nathan__> perfect
20:14:56 <ennael> as said in mail it's a temporary iso
20:15:05 <DavidWHodgins> So with nvidia, please ensure bug reports specify the workaround (if any), and have the proper keywords. Select the keywords "word" to see the available ones
20:15:10 <ennael> we need tests on partitioning step during install
20:15:46 <wilcal> ok pretest version here installed, and booted to a working desktop, sans nvidia driver here on real hardware
20:16:17 <DavidWHodgins> #info Please concentrate testing with the latest x86-64 dvd on the partitioning problems. Once that's confirmed fixed, the remaining iso images will be generated for further testing
20:16:20 <wilcal> will shift to released version this weekend
20:17:06 <tjandrews> I have never run into the partitioning problem. How would I go about testing for it?
20:17:10 <wilcal> Geting close
20:17:24 <DavidWHodgins> #info Once the testing confirms partitioning fixed, and the remaining iso images generated and tested, we can finally release sta2
20:17:30 <tjandrews> MBR machines only.
20:17:34 <wilcal> What's a worst case David
20:17:50 <DavidWHodgins> worst case is that we find the partitioning problems have not been fixed
20:18:07 <DavidWHodgins> tjandrews: See bug 20074
20:18:08 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=20074 critical, release_blocker, mageiatools, NEW , Partitions with Type: Empty corrupt the partition table, drakxtools
20:18:23 <lewyssmith> -3 Reports have been good so far.
20:19:14 <DavidWHodgins> That's promising. Best if people who have encountered the bug previously retest using the same setup
20:19:41 <wilcal> Probably not something that can be tested in Vbox?
20:19:53 <azziam> TJ, for the partitioning testing, use Custom Partitioning to create a couple new ones.  See if they are numbered correctly.
20:20:23 <DavidWHodgins> It needs testing both in vbox and on real hardware with mbr only, gpt only, and a mix of mbr and gpt partition tables
20:20:31 <wilcal> What if I used something like Gparted to create a couple of fat32 partitions and put data in there
20:20:49 <DavidWHodgins> Can't hurt to try it
20:20:52 <wilcal> then let M6 have its way to install to what's left
20:21:03 <wilcal> That's pretty easy to do
20:21:32 <DavidWHodgins> Everyone clear on what we need to do for x86-64 dvd sta2?
20:21:37 <wilcal> Got it
20:21:45 <tarazed> Reckon so
20:21:59 <tjandrews> I think...
20:22:05 <azziam> The worst of the partitioning problem was when it was wrongly reporting DOS tables as GPT tables and requiring a Boot partitition be made.  Martin fixed that and a lot more.  Cheers for Martin!
20:22:07 <DavidWHodgins> :-)
20:22:22 <DavidWHodgins> Ok, moving on ...
20:22:27 <nathan__> ok
20:22:38 <DavidWHodgins> #topic Testing updates - Any difficulties, problems, issues?
20:23:15 <DavidWHodgins> The kernel updates will likely be closed as wont fix, and new ones created with the latest security patches
20:23:21 <tarazed> util-linux.  I'll try it out on machines with slow and fast cpus.
20:23:40 <wilcal> Lots of OK's in there
20:23:53 <tarazed> quagga needs 32bit test.  vbox todo
20:23:57 <DavidWHodgins> I haven't seen a poc for the kernel problem, but due to the publicity, we'll want to test them quickly
20:24:25 <DavidWHodgins> Given that the latest issue will only require a minor change to the kernel, the prior testing is still valid
20:24:50 <DavidWHodgins> tmb: Are you available?
20:24:51 <tarazed> ming might need a bit more attention.  A bucket-full of CVE's and possible PoCs.
20:25:07 <tmb> DavidWHodgins, yes
20:25:30 <DavidWHodgins> I expect you've seen today's press about the most recent kernel security bug
20:26:02 <tmb> DavidWHodgins, wich one ?
20:26:04 <wilcal> I have not
20:26:19 <DavidWHodgins> tmb DCCP
20:26:39 <DavidWHodgins> local user can get root privleges
20:27:04 <DavidWHodgins> CVE-2017-6074
20:27:57 <lewyssmith> tarazed: Len, I have looked briefly at Ming, and will hope to try it for 64-bit unless you get there first.
20:28:15 <tarazed> lewis: OK
20:29:09 <tmb> DavidWHodgins, ah, that one... I guess I should add that on top on current 4.4.50-1 in testing to minimize changes and not invalidate already done 4.4.50 tests
20:29:20 <DavidWHodgins> Please do
20:29:49 <tmb> ok, will respin with only that minimal patch added
20:30:09 <DavidWHodgins> Testing that and getting it released will take priority over everything else, including sta2 testing
20:30:56 <DavidWHodgins> #info New kernel updates being produced. Please give testing of the new kernels priority over everything else, including sta2 testing.
20:31:45 <DavidWHodgins> The other bugs all look pretty standard. Any questions about how to test any the the other updates?
20:32:01 <tjandrews> Sounds like we'll need to nudge qa-discuss once they are ready. It's worked to get them tested fast before.
20:32:18 <DavidWHodgins> tjandrews: Good idea
20:32:21 <tjandrews> kernels, I mean.
20:32:26 <tmb> Yeah, I'll ping qa-discuss as soon as they are ready
20:32:38 <DavidWHodgins> Great. Thanks!
20:33:12 <DavidWHodgins> Any questions or comments about other updates before we move on?
20:33:18 <wilcal> not from me
20:33:25 <nathan__> no
20:33:47 <lewyssmith> Dave: please leave next couple of advisories to me. Nearly there!
20:33:53 <tarazed> webkit2 has a huge list of CVEs, none of which provide much useful information for us
20:34:05 <tarazed> so I guess we ignore them.
20:34:09 <DavidWHodgins> I almost forgot. I included the comment about the kernel in the list of topics, before finding out about the latest kernel cve
20:34:15 <lewyssmith> Yes.
20:34:30 <lewyssmith> (to Len)
20:34:46 <DavidWHodgins> Just wanted to mention, that with kernels we normally want multiple confirmations that they are ok before adding the ok tag
20:35:02 <DavidWHodgins> As wide a variety of hardware as possible
20:35:15 <DavidWHodgins> Don't validate too quickly
20:35:37 <lewyssmith> Yes, individual reports should not cite OK for kernels. That is done by later review.
20:36:13 <DavidWHodgins> With today's kernel update, that's a different matter. It's a high priority security update, so will only require very limited testing before being okd and then validated
20:36:37 <tarazed> But we still leave that to tmb?
20:36:42 <wilcal> Should be released by Monday right
20:37:36 <DavidWHodgins> #info Normally don't add an ok for a kernel update until multiple people have confirmed it's ok on their systems. Kernels need a wide variety of testing on both uefi and non-uefi systems
20:37:36 <lewyssmith> tarazed: Ideally, yes. But one of the chiefs otherwise (= you!).
20:38:05 <DavidWHodgins> tarazed: Actually pushing the kernel updates is normally done by tmb, and only when he's satisfied
20:38:40 <lewyssmith> I agree with that. Just unsure about the formality of it.
20:38:54 <DavidWHodgins> #info The kernel update being built today is a high priority security update. It will be okd and validated with much less testing
20:39:31 <DavidWHodgins> Any other questions about kernel testing or other updates testing before we move on?
20:39:35 <tarazed> I have 4 machines including 1 MBR
20:40:27 <DavidWHodgins> Doesn't look like Luigi12 is available, so skipping the roundup
20:40:46 <DavidWHodgins> That only leaves ...
20:40:49 <DavidWHodgins> #topic Anything else?
20:40:59 <lewyssmith> Yes, here. Quick.
20:41:09 <wilcal> Ya, Mozilla just released their 2017 schedule
20:41:17 <wilcal> https://www.mozilla.org/en-US/firefox/organizations/faq/
20:41:18 <[mbot> [ Firefox Extended Support Release FAQ — Mozilla ]
20:41:49 <Benmc> good news, when sta2 installed to HDD with empty partition table, bug 20074 appears to be fixed
20:41:51 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=20074 critical, release_blocker, mageiatools, NEW , Partitions with Type: Empty corrupt the partition table, drakxtools
20:41:55 <wilcal> ESR Firefox is transitioning from 45.x -> 52.x right at the release time of M6
20:42:28 <wilcal> also they are really whacking away on the way add-ons work
20:42:42 <wilcal> old ones won't work anymore including Flash
20:42:43 <DavidWHodgins> We won't hold m6 for it. If we can include it great. If not, it'll be an update after release
20:43:10 <tarazed> Ben:  Hooray!
20:43:11 <wilcal> just be aware we may have a major uptick to this after M6 release
20:43:12 <DavidWHodgins> Wow. Feels good to be talking about "after release" :-)
20:43:23 <wilcal> Finaly
20:43:47 <lewyssmith> Re XFCE Lives, are we going to do the 64-bit one? What was the council decision?
20:44:01 <wilcal> still an open issue i think
20:44:09 <lewyssmith> Hell.
20:44:14 <wilcal> but most would like to see an x86_64 version
20:44:27 <DavidWHodgins> Also, I just wanted to publicly thank everyone who stepped out to help out during the times I've been sick over the past several years. Especially lewyssmith and wilcal
20:44:47 <DavidWHodgins> Hopefully my health will continue to improve
20:44:53 <lewyssmith> wilcal: I just want to know for doumentation. I will assume both architectures.
20:44:54 <tarazed> +1
20:45:03 <azziam> "After release" of what started testing in Sept or Oct of 2015!
20:45:11 <wilcal> :-))
20:45:36 <lewyssmith> We are still at STA2...
20:45:39 <wilcal> Major thing with M6 has been Plasma I think
20:45:43 <DavidWHodgins> lewyssmith: It would be nice, but I'm concerned with the increasing workload. With Mageia 5, efi doubled the testing required. gpt then doubled it again.
20:46:13 <DavidWHodgins> The opportunity to reduce it a bit by not having a 64 bit XFCE iso is appealing
20:46:14 <lewyssmith> No: the 2 went together.
20:46:18 <wilcal> yes assume both i586 & x86_64 XFCE for now
20:46:38 <lewyssmith> Thanks.
20:46:39 <DavidWHodgins> Yes, they went together creating 4 times the amount of testing for each iso image
20:46:52 <wilcal> It's definitely a Marketing ploy David and a good one IMO
20:47:06 <wilcal> something to think about
20:47:32 <Pharaoh_Atem> what am I supposed to do if I can confirm the pre-sta2 image fixes a blocker bug I reporteD?
20:47:33 <Pharaoh_Atem> *reported
20:47:45 <DavidWHodgins> Anyway, that's a council decision as it affects multiple teams. Council is considering it, but hasn't actually decided yet
20:47:48 <Benmc> FYI, Gnome nor Plasma Live boot on one of my test units, XFCE does
20:47:59 <lewyssmith> DavidWHodgins: You exaggerate greatky, Dave. EFI+GPT effectively added 1 more platform. A pain, yes.
20:48:11 <DavidWHodgins> Pharaoh_Atem: Comment on the bug report and close it
20:48:13 <Pharaoh_Atem> lewyssmith: you can have BIOS+GPT too
20:48:17 <wilcal> Good'ole XFCE :-))
20:48:31 <wilcal> LIke an old friendly dog
20:48:38 <lewyssmith> Pharaoh_Atem: If you are margianl, you 'can', yes.
20:48:57 <Pharaoh_Atem> lewyssmith: >2TB drive on BIOS system is not marginal these days
20:49:10 <Benmc> wilcal: that will be the next new DE : OFD :)
20:49:25 <lewyssmith> DavidWHodgins: You forget also that we have managed to drop 2 live ISOs.
20:49:26 <wilcal> I use 2TB drives on my "production" machines
20:49:32 <DavidWHodgins> lewyssmith: We went from testing i586 and x86-64 installs (for the classical iso images) to testing each with bios+mbr, bios+gpt, uefi+mbr, uefi+gpt, both with i586 and x86_64
20:49:42 <Pharaoh_Atem> anyway, Martin has been a godsend on fixing installer bugs :)
20:49:47 <wilcal> he has
20:49:51 <DavidWHodgins> lewyssmith: Dropped 2, and added one
20:49:58 <lewyssmith> Not BIOOS + GPT, that was for extremists.
20:50:18 <DavidWHodgins> bios +gpt is needed for anyone with a hard drive over 2TB
20:50:30 <DavidWHodgins> That's not extremist anymore
20:50:54 <Pharaoh_Atem> my desktop computer came with a 2TB hard drive standard
20:51:01 <Pharaoh_Atem> you can't format that MBR, it has to be GPT
20:51:02 <azziam> My main machine has a 3TB.  I'll keep it GPT only.
20:51:19 <lewyssmith> With BIOS?
20:51:25 <DavidWHodgins> Pharaoh_Atem: 2TB is the limit for mbr
20:51:31 <wilcal> How much of that 3TB drive can you get to?
20:51:44 <DavidWHodgins> lewyssmith: I can easily add a 3TB drive in my existing bios system
20:51:48 <Pharaoh_Atem> well, 2 TiB, yes, but depending on the drive maker, it might actually cross that line while being 2TB :/
20:51:54 <azziam> All of it or as much isn't formatting space used up.
20:52:06 <DavidWHodgins> I wont as I'm sticking with ssd drives for new hardware
20:52:09 <wilcal> What's the limit
20:52:09 <Pharaoh_Atem> yay for base 2 vs base 10 data measurements :(
20:52:19 <Pharaoh_Atem> 2 tebibytes
20:52:25 <DavidWHodgins> :-)
20:52:29 <azziam> My boot drive is SSD.
20:52:33 <Pharaoh_Atem> which is base 2 notation for 1.38 TB
20:52:41 <Pharaoh_Atem> err, 1.98 TB
20:52:43 <wilcal> don't you have to go to EFI to get the full 3TB
20:52:48 <Pharaoh_Atem> no, just GPT
20:52:55 <DavidWHodgins> wilcal: No
20:53:24 <wilcal> If you put ma 3TB drive on a non GPT system how much do you get
20:53:32 <Pharaoh_Atem> the BIOS boot partition is a ~1MB partition at the beginning of the disk that has a stub to boot a GPT disk on BIOS
20:53:35 <azziam> 2TB
20:53:45 <DavidWHodgins> The /boot filesystem likely has to be in the first 2 TB, but once booted all of the drive is available
20:53:51 <Pharaoh_Atem> yep
20:54:27 <DavidWHodgins> The bios only has to read the initrd. Once that's done all of the drive is available
20:54:29 <Pharaoh_Atem> I think /boot needs to be way earlier than that, (I think it's the first 4MB? don't remember exactly)
20:54:31 <wilcal> So I'm gonna transition to GPT on two of my systems for M7
20:55:10 <DavidWHodgins> I have 2 ssd drives and one hd. The second ssd drive I use for my gpt testing
20:55:13 <wilcal> You can mount an ext4 10TB drive as long as it's not the boot drive
20:55:17 <Pharaoh_Atem> as long as the /boot partition is early enough on the disk and you have a BIOS boot part at the beginning of the disk, everything's gravy
20:56:12 <DavidWHodgins> Anything else?
20:56:21 <lewyssmith> Not here.
20:56:32 <DavidWHodgins> Countdown time then
20:56:37 <DavidWHodgins> T - 5
20:56:40 <DavidWHodgins> 4
20:56:43 <DavidWHodgins> 3
20:56:47 <DavidWHodgins> 2
20:56:49 <wilcal> bye all
20:56:50 <DavidWHodgins> 1
20:56:51 <tjandrews> I guess I'll just sit here with a 500GB and a 350GB drive on my machine, feeling inadequate...
20:56:52 <lewyssmith> Thanks everyone for coming. Goodbye.
20:57:01 <DavidWHodgins> Thanks for coming everyone
20:57:02 <Pharaoh_Atem> thanks for doing the good work, everyone!
20:57:10 <DavidWHodgins> #endmeeting