19:02:42 <DavidWHodgins> #startmeeting
19:02:42 <Inigo_Montoya> Meeting started Thu May 10 19:02:42 2018 UTC.  The chair is DavidWHodgins. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:02:42 <Inigo_Montoya> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:02:58 <DavidWHodgins> #chair wilcal lewyssmith
19:02:58 <Inigo_Montoya> Current chairs: DavidWHodgins lewyssmith wilcal
19:03:04 <DavidWHodgins> #topic * Who's new? - If you are then come and say hello
19:03:20 <DavidWHodgins> Anyone here who hasn't attended an qa team irc meeting before?
19:03:42 <DavidWHodgins> #topic * The Grand ( Plasma/kf5/kde/qt ) Update is upon us.
19:04:11 <DavidWHodgins> The sysadmins have been working hard! Thanks
19:04:24 <wilcal> 19:00 UTC time tomorrow is the official release time I think
19:04:24 <tjandrews> I see every time one glitch is cleared up, another rears its head.
19:04:33 <DavidWHodgins> And thanks to all of the testers who have worked so hard on these updages for so long!
19:04:44 <wilcal> Scary thing it is
19:04:54 <wilcal> to see all those apps go rolling by
19:04:56 <DavidWHodgins> I see from the qa discuss list that, yes problems are still being found
19:05:26 <wilcal> Martin I think
19:05:35 <lewyssmith> tmb said "If someone with local mirror want to do an extra test if we got all rpms
19:05:35 <lewyssmith> > listed you can regenerate hdlists / repodata for Core Updates"
19:05:40 <DavidWHodgins> Don't be surprised if we end up with some very high priority updates after these updates do go public
19:05:42 <lewyssmith> How do we do this?
19:06:35 <lewyssmith> James used genhdlist2 ; Martin gendistrib
19:06:36 <DavidWHodgins> If you have a local mirror, as root cd to the updates directory and run genhdlist2
19:06:58 <wilcal> hello tmb
19:07:00 <lewyssmith> Welcome Thomas.
19:07:06 <tmb> hi all
19:07:16 <DavidWHodgins> HiYa tmb. Thanks for all the great work!
19:07:20 <wilcal> were talking the  big one now
19:07:26 <lewyssmith> DavidWHodgins, What is a 'local mirror'?
19:07:56 <lewyssmith> On your box?
19:07:58 <DavidWHodgins> That's where you use rsync or similar, to copy all of the contents of another mirror to your hard drive
19:08:31 <lewyssmith> Not for me. When can we try areal mirrors?
19:08:32 <DavidWHodgins> I used to have one, but that computer died along with 3 of it's 4 hard drives
19:09:04 <wilcal> I run a local repo that's mirrored to kernel.org
19:09:04 <DavidWHodgins> lewyssmith: That will have to wait until tomorrow after the hdlists are rebuilt and made public
19:09:34 <wilcal> how long will it take to trickle down through the mirrors
19:10:16 <DavidWHodgins> Most of the mirrors seem to have most, if not all of the updates synced already. The hdlists will only take a few minutes when a mirror syncs, to download them.
19:10:21 <brian__> it started "trickling" yesterday correct?
19:10:29 <DavidWHodgins> Yes
19:10:32 <lewyssmith> I have a base system all ready to update for real.
19:10:55 <brian__> me 2 - I have one
19:10:58 <wilcal> I'll first try with a victim Vbox client
19:11:00 <DavidWHodgins> It's the mirrors that sync once a day that tend to cause problems for their users
19:11:20 <DavidWHodgins> Most sync more often than that
19:11:36 <lewyssmith> What are hdlists?
19:12:13 <tjandrews> I have one last system to update - my brother's - that I will get once it's released.
19:12:20 <DavidWHodgins> Those are the files in each repository that urpmi uses to figure out what packages are available, and what versions
19:12:47 <lewyssmith> Which is why pkgs can be fed in advance?
19:12:55 <DavidWHodgins> Yes
19:13:19 <lewyssmith> Why did Martine hit so many problems, before just 1?
19:14:27 <DavidWHodgins> By using a local mirror with updated hdlist files, and not using the testing repos, he found that some other updates (appstream) are needed too.
19:14:49 <tmb> no
19:15:01 <wilcal> there was a very big update on my local repo this morning
19:15:02 <lewyssmith> I noted that in a ML post; but forgot its name. That forms *part of the update*.
19:15:18 <tmb> the real problem is rpm ordering not picking up task-obsolete fast enough
19:15:30 <tjandrews> I ran into something similar, early in testing. In my case, a package was being updated before one of its dependencies was downloaded.
19:15:37 <DavidWHodgins> Ah. I stand corrected.
19:15:43 <lewyssmith> And me.
19:16:01 <tmb> Maybe a dependency loop somewhere wich makes rpm breaking transaction ordering to get out of the loop
19:16:07 <brian__> how many megabytes is the download.  Should we put that in the blog?
19:16:19 <lewyssmith> Hundreds.
19:16:37 <DavidWHodgins> It depends on what packages you have installed as to how many updates you'll get and how big.
19:16:42 <tjandrews> People may be scared enough as it is... ;-)
19:16:46 <brian__> yeah, I think in my initial it was around 400, but I probably missed some repos
19:16:51 <wilcal> I think it's big. Took about 80 minutes here
19:17:02 <wilcal> to download repo updates
19:17:08 <lewyssmith> I would expect that sort of thing.
19:17:19 <brian__> ok - so let's not state the size as it will vary
19:17:34 <lewyssmith> But warn it will be huge & take ages.
19:17:39 <DavidWHodgins> Right. Same with number of packages.
19:17:44 <benmc> sorry, got to go. may be back later. bye all
19:17:47 <wilcal> but I'm mirroring the whole repo locally
19:18:03 <brian__> benmc - hopefully enjoy an omellette
19:18:18 <DavidWHodgins> Also, keep in mind the 400+ number is the number of srpm packages. Any given srpm package may generate a number of rpm packages.
19:18:33 <DavidWHodgins> benmc: Ok. Thanks for coming
19:18:41 <lewyssmith> I have done it >once from Updates Testing, & posted figures.
19:18:44 <brian__> true - often one of my tests is one source rpm and man objects
19:19:01 <wilcal> How do you feel about how the individual apps will fare through all of this
19:19:23 <rindolf> hi all
19:19:30 <wilcal> Basic things like OpenOffice
19:19:34 <DavidWHodgins> I'm confident that with these updates, kde users will find it's much better than before the updates
19:19:35 <brian__> rindolf: hi!
19:19:36 <wilcal> hi rin
19:19:41 <tmb> Ah, martinw just found a missing obsoletes for *-eo-* packs wich explains why task-obsolete did not get pulled
19:19:47 <lewyssmith> I found almost no problems on the Plasma front. In the end, very few on the LXQt side.
19:20:25 * tmb thikns this stagen rollout saved our but now...
19:20:49 * tmb cant type anymore...
19:20:50 <DavidWHodgins> Yes. Glad you did it. Thanks!
19:20:59 <rindolf> tmb: hi, should we update to the latest 4.14.x kernel?
19:21:43 <tmb> rindolf, if you want you could start rolling up cauldron to 4.16 :)
19:22:13 <rindolf> tmb: ah
19:22:18 <lewyssmith> wilcal, At one point, ages ago, I tried *all* the applications I had installed; not to do often...
19:22:37 <tmb> rindolf, otherwise, feel free to update cauldron to latest 4.14.x ... but please pick up the fixes from -stable queue too then.
19:22:51 <rindolf> tmb: ah
19:22:54 <wilcal> last weekend I did a full new install then did the big one on it and all ran fine
19:23:13 <DavidWHodgins> Any other questions or comments on the big update?
19:23:41 <DavidWHodgins> #topic * Testing updates - Any questions or problems
19:23:53 <DavidWHodgins> tmb: madb.mageia.org is not responding
19:24:04 <lewyssmith> Should we announce on QA-discuss before elsewhere? To give us a chance to spot bad things?
19:24:12 <tmb> DavidWHodgins, thats a server managed by Stormi
19:24:42 <DavidWHodgins> lewyssmith: Announce what? The hdlists being regenerated?
19:24:52 <lewyssmith> Yes
19:25:20 <DavidWHodgins> If it works, the updates will become available to everyone. If it doesn't, they won't see the updates so nothing will happen.
19:25:56 <wilcal> same here David
19:26:17 <lewyssmith> Re updates: Problem: a huge number to do!
19:26:52 <DavidWHodgins> Without madb, we can't currently see the list. Suffice it to say there are a lot, so when it does become available, please work on them asap
19:27:08 <brian__> will do when bandwidth allows
19:27:19 <brian__> mine, not isp
19:27:30 <DavidWHodgins> :-)
19:27:33 <wrw105> have the build problems for chromium ever been resolved?
19:27:35 <tmb> https://bugs.mageia.org/buglist.cgi?cmdtype=runnamed&list_id=91374&namedcmd=QA-Pending
19:28:44 <tjandrews> Sigh. I shall have to set aside my intriguing laptop restoration project and use my limited free time on updates.
19:29:00 <tmb> (thats my pre-madb updates tracker)
19:29:08 <brian__> :tj I have a Zenith 286 you can work on next.  ;-)
19:29:13 <lewyssmith> tjandrews, Thanks TJ.
19:30:46 <DavidWHodgins> tmb: That link doesn't work here, but noticed the "waiting for QA test" link at the bottom under "My Bugs". Not sure if everyone sees that or if that's something I added
19:31:31 <DavidWHodgins> Hovering over that it shows "Shared by Manuel Hiebel"
19:32:08 <wrw105> Yeah, not recognizing the qapending here
19:32:59 <lewyssmith> No matter. There are pages of bugs.
19:33:14 <DavidWHodgins> Yeah
19:33:31 <DavidWHodgins> #topic * Anything else?
19:33:32 <tmb> DavidWHodgins, try again
19:33:58 <lewyssmith> tmb, Are you a wizard of sorts?
19:34:11 <tmb> lewyssmith, why ?
19:34:16 <DavidWHodgins> Still says "The search named QA-Pending does not exist."
19:34:22 <lewyssmith> Doing magical things.
19:34:25 <brian__> I just received a gt730 card.  Any tips beyond installing nvidia 390 drivers?
19:34:26 <wilcal> back
19:35:04 <tmb> ah, you need to use: https://bugs.mageia.org/buglist.cgi?cmdtype=dorem&list_id=91387&namedcmd=QA-Pending&remaction=run&sharer_id=62
19:35:24 <brian__> that worked
19:35:30 <wilcal> When exactly will the hdlist file be turned loose? 19:00 UTC tomorrow?
19:35:52 <tmb> I will regenerate them around that time, yes
19:36:13 <tmb> lewyssmith, well I'm a Mageian, so ... :)
19:36:18 <DavidWHodgins> Thanks! Bookmarking that link
19:36:25 <DavidWHodgins> lol
19:36:32 <wilcal> So by Sat morning California time it should all be trickled through the repos
19:37:04 <wilcal> so my sat morning repo rsync should be the one
19:37:07 <lewyssmith> I can try tomorrow Europe evening if I can work out the local time!
19:37:23 <DavidWHodgins> wilcal: Should be roughly noon on Friday for you
19:38:07 <brian__> so I'll boot up my plasma disk tomorrow and if it pops up, the big downlaod has arrived
19:39:17 <DavidWHodgins> Looks like countdown time
19:39:19 <tjandrews> 19:00 UTC is 3pm here. I figure I may have to wait another hour for kernel.org to sync.
19:39:30 <DavidWHodgins> t - 5
19:39:31 <wilcal> Thanks to David for running the show. Super job
19:39:34 <DavidWHodgins> 4
19:39:37 <DavidWHodgins> 3
19:39:41 <DavidWHodgins> 2
19:39:44 <DavidWHodgins> 1
19:39:49 <lewyssmith> Thanks to all who are here. Goodbye.
19:40:03 <DavidWHodgins> Thanks for coming everyone, and for all the hard work on these updates.
19:40:09 <DavidWHodgins> #endmeeting