16:19:57 <itchka> #startmeeting
16:19:57 <chwido> Woof! Let's start the meeting. It's Wed Mar  9 16:19:57 2016 UTC. The chair is itchka. Information about me at https://wiki.openmandriva.org/en/Chwido.
16:19:57 <chwido> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:19:57 <chwido> You should add extra chair(s) just in case, with #chair nick.
16:19:57 <chwido> Have a good meeting, don't bark too much!
16:21:51 <itchka> who is here crisb fedya HisShadow itchka linukiss live_joel n3npq nicco_ OnlyHuman Pharaoh_Atem plfiorini Son_Goku tapwag Xu_R
16:23:36 <itchka> and ben79
16:23:56 <ben79> yes
16:24:06 <itchka> is here :)
16:24:31 <itchka> #item 1
16:24:53 <itchka> #topic Release Report
16:26:04 <itchka> Since bero nor _TPG are here crisb would would you like to fill in with what you know and I'll fill in any details I have.
16:27:15 <crisb> i have cleared the repoclosure for x86_64 and i586
16:28:00 <n3npq> crisb: w the openssl drown update? good if so ...
16:28:11 <crisb> just finishing off the rebuild of packages for ffmpeg 3.0
16:28:19 <itchka> crisb: Nice does that include restricted and the like.
16:28:41 <crisb> yeah main and restricted and non free
16:29:07 <itchka> crisb: That's really good thatnks for all the effort.
16:29:29 <crisb> n3npq: we have openssl 1.0.2g yeah
16:29:48 <itchka> abf seems to be stabilising now.
16:29:55 <n3npq> nicely done: lots of breakage ripping out SSLv2 everywhere
16:31:13 <itchka> With the main repos up to scratch I think we can give ashledombus the go ahead on mirror synching now.
16:31:59 <crisb> i think so, its sorely needed for 2014
16:32:44 <itchka> _TPG: Has produced a new iso for cooker it kinda works but it has a few foibles.
16:33:05 <itchka> Has anyone here done any testing yet?
16:33:37 <itchka> if so short one line reports please
16:33:40 * blackcrack think, why sync not at all from the abf-downloads to the yandex ? as master mirrorclient or someting for the "customers" where mirroring at all
16:34:07 <OnlyHuman> chwido: bark atestingexperience
16:34:08 <chwido> (bark <target>) -- Barks at someone you tell him to.
16:34:19 <OnlyHuman> chwido: bark testingexperience
16:34:19 <chwido> (bark <target>) -- Barks at someone you tell him to.
16:34:27 <OnlyHuman> chwido: bark testing
16:34:27 * chwido barks at testing while having an interested look at a place somewhere below the belly button.
16:34:58 <tapwag> itchka, I didn't do an installation but tried to work that Mandriva branding gets removed
16:35:16 <ben79> .iso # 97 is installed and working here. Just notice that page up/down aren't working in Konsole.
16:35:57 <itchka> ben79: You had no issues?
16:35:59 <tapwag> I am having some trouble setting up virtual machines
16:36:26 <live_joel> itchka: I tested id, it works in all
16:36:48 <blackcrack> and i had to less time up to now..
16:36:58 <live_joel> but we have always the blockers
16:36:59 <itchka> live_joel: Did you install?
16:37:03 <blackcrack> i want test it later..
16:37:12 <live_joel> live only
16:37:18 <tapwag> I did some translations on a ROSA project called soundKonverter as well.
16:37:37 <live_joel> see bugzilla the last bugs
16:37:52 <tapwag> live_joel, The issue of OpenMandriva not running with NTFS?
16:38:14 <live_joel> no windows here sorry
16:39:01 <ben79> Other than page up/down in Konsole and the issue aroung read speed on SSD no.
16:39:03 <tapwag> live_joel, I just looked it up. itchka you reported it didn't you?
16:39:23 <tapwag> But there were some issues with the applications that we ship as well
16:39:31 <tapwag> gwenview and VLC
16:39:45 <itchka> tapwag: I reported that one there's a simple fix and it will be a small change to a config file when building the iso to make sure that the ntfs module is loaded into the initrd before building the iso.
16:40:20 <tapwag> itchka, Still it would be a matter of getting that into the distribution as well to have a quality release
16:42:23 <itchka> Oh yes I would not call the current iso a beta for the ntfs problem alone.
16:42:26 <live_joel> this iso  is a good starting point
16:45:16 <itchka> I believe There is another issue that mostly laptop users will see the intel915 graphics driver seems to be greedy like grep and always loads even if you main graphics card is something different.
16:46:03 <itchka> I have someone investigating this at the moment.
16:46:40 <OnlyHuman> who is that  columbo?
16:46:51 <ben79> This .iso does seem a good starting point. I'm wondering if it's OK to realease a Beta with spellcheck (KDE apps) and numlock not working? Also no page up/down in Konsole?
16:47:02 <OnlyHuman> or inspector clouseau
16:47:29 <OnlyHuman> :)
16:48:13 <itchka> It's Efrem McCrimon he has loads of hardware as he does QA for a living.
16:49:03 <tapwag> ben79, There is also this issue relating to consoles: https://issues.openmandriva.org/show_bug.cgi?id=1194
16:49:58 <OnlyHuman> h a real inspector
16:51:24 <itchka> Indeed so
16:52:00 <ben79> tapwag: that's something entirely different. I just noticed page up/down broken within the hour. Will file a bug when I can.
16:52:51 <ben79> No spellcheck in KDE apps and numlock are kind of old bugs:
16:52:55 <ben79> https://issues.openmandriva.org/show_bug.cgi?id=1466
16:53:29 <ben79> https://issues.openmandriva.org/show_bug.cgi?id=1314
16:53:57 <itchka> ben79 bugs like this can be fixed once we have issued a QA approved beta.
16:54:17 <ben79> OK, that answers my question
16:54:56 <live_joel> essential is mariadb and updating system
16:56:43 <itchka> Please QA team keep filing bugs.
16:56:50 <live_joel> and Tpg asked help and he says about mariadb : I've lost motivation on this bug.
16:57:15 <live_joel> https://issues.openmandriva.org/show_bug.cgi?id=1520
16:58:01 <tapwag> live_joel, There was also an issue with OMA-welcome you reported - https://issues.openmandriva.org/show_bug.cgi?id=1455
16:58:11 <itchka> live_joel: Ive been looking at this but only in a vm.
16:58:17 <tapwag> I found this fairly major as it is our own software
16:58:40 <live_joel> tapwag: not important
16:59:01 <live_joel> for the moment
16:59:21 <itchka> tapwag: For beta approval we must concentrate on boot and install. Without that people can't test. It's no good just testing the live image.
17:00:02 <itchka> Has anyone tried a uefi install?
17:00:41 <live_joel> with bios ?
17:00:57 <ben79> itchka: My install is UEFI
17:01:54 <itchka> ben79: It worked ok? You got an openmandriva boot entry in you boot menu and it was labelled the default?
17:02:45 <blackcrack> i use only the old methode, i dont like uefi ..
17:02:46 <ben79> itchka: Yes
17:04:10 <ben79> itchka: Yes UEFI install works for OMLx 3 but some other OS's won't boot from our grub2: https://issues.openmandriva.org/show_bug.cgi?id=1533
17:06:46 <itchka> Hmm that's not good news. Secure ffing boot I suppose. Just looking
17:08:33 <itchka> Hmm weird, grub2 was recently updated. Needs looking into.
17:08:49 <ben79> itchka: I disabled secure boot in openSUSE, also currently there is a 0MLX3/2015 partition that won't boot.
17:09:11 <itchka> We won't be able to dual boot Windows 10 of course.
17:09:26 <ben79> In other words the installed yesterday partition boots but my ongoing Cooker partiton won't boot.
17:09:55 <ben79> Odd as both are OMLx/2015.
17:09:59 <Xu_R> ben79: do you have os-prober installed
17:10:00 <Xu_R> ?
17:10:30 <ben79> It must be or grub2 wouldn't be listing 5 operating systems
17:11:12 <itchka> ben79: That kinda ties up with what I have found over the last couple of hours when tring to install. I think Calamares may have an issue.
17:11:14 <ben79> Wait, maybe it's 4 OS's
17:11:56 <ben79> itchka: No it's more as running grub2-mkconfig in installed system leads to same problem.
17:12:57 <ben79> Xu_R: os-prober-1.65-2-omv2015.0.x86_64 is installed
17:13:11 <itchka> ben79: Post your grub.cfg against the bug.
17:13:13 <Xu_R> mm. weird.
17:16:26 <itchka> Ok I think it's clear that QA need to look hard around the booting and install area and report bugs and potential solutions whereever possile.
17:17:40 <itchka> Are all QA team agreed that the current iso is not suitable for beta due to install isuues?
17:18:30 <itchka> issues
17:20:36 <ben79> Yes, but it is getting close.
17:21:14 <Pharaoh_Atem> itchka: here
17:21:15 <ben79> itchka: added grub.cfg to https://issues.openmandriva.org/show_bug.cgi?id=1533
17:22:58 <itchka> Thanks and Yes I think you are right. lets hope we can find out what's wrong so the devteam can fix it.
17:25:13 <itchka> Ok lets move onto the next item which apparently #share
17:26:20 <itchka> I assume klebedeff put this in to remind me to add any news.
17:27:05 <itchka> #share Main cooler repos are now in good shape
17:28:02 <ben79> Yeah, this is where the GRAND POOHBAH (you!) get to briefly write what we need to share with community from above discussion.
17:28:42 <ben79> Actually who writes it should be chair or whomever was leading/controlling conversation.
17:29:47 <itchka> #share A new iso has been issued for QA approval for beta some boot and issues were found that need fixing so this candidate was not accepted.
17:31:27 <itchka> #share As the repos are now in good shape the process of connecting the new ABF to the mirror network will now be started.
17:32:45 <itchka> Does that cover it?
17:34:10 <n3npq> itchka: is mirroring from Yandex a separate issue?
17:34:41 <itchka> n3npq: Yes that is infras ballpark.
17:41:04 <itchka> Ok lests move onto the next item
17:41:14 <itchka> #item 2
17:41:58 <itchka> #topic Beta ETA
17:44:50 <n3npq> what is the current date? 2+ weeks from last week was the overly optimistic date from last week
17:45:22 <n3npq> w 8 weeks between beta -> release
17:46:33 <n3npq> presumably with the rejects the past week this (at least) moves to now+2weeks == beta, +8weeks == release
17:47:01 <ben79> 8 wks tween beta and release seems OK but we seem to need at least 2 more weeks for a BETA? Or so I'm thinking?
17:47:49 <itchka> I really don't know what to say about this one. We can'y really go ahead without mariadb being sorted and the boot/grub issues. I'm hoping that none of there are too serious and can be fixed in the next build. If we are lucky we may have a fixed build by Friday; basic test over the weekend and hopefully QA approval then official release.
17:48:42 <itchka> I'd say beta release should be possible by next TC
17:48:53 <n3npq> itchka: good: itemized impediments rather than imp[licit date ETA focuses on what needs doing
17:50:49 <itchka> crisb: Your opinion please.
17:51:48 <live_joel> I repeat TPG needs help with mariadb, maybe other dev can look into ?
17:52:12 <itchka> live_joel: That has been noted :)
17:54:10 <itchka> live_joel: I'm pretty sure I already know what the problem is It's almost certainly due to a syntax error in a patch.
17:54:17 <itchka> Hi Kvok
17:54:44 <n3npq> itchka: as in will not build? do you have a ptr to the build log?
17:55:55 <itchka> n3npq: yes but it doesn't give much away. Hang on
17:57:48 <itchka> n3npq: You'll find the logs here.
17:57:54 <itchka> https://abf.openmandriva.org/build_lists/28635
18:00:31 <itchka> Ok folks I'm going to stick my neck out here and say that we will release a good beta on or before next weeks TC. Probably get hung by _TPG for it so it's up to you lot to get me out of the poop:)
18:01:04 <n3npq> itchka: this line
18:01:06 <n3npq> install -p -m 755 scripts/mysql-prepare-db-dir /builddir/build/BUILDROOT/mariadb-10.1.12-2-omv2015.0.x86_64-buildroot/usr/libexec/mysql-prepare-db-dir
18:01:07 <n3npq> install: cannot stat 'scripts/mysql-prepare-db-dir': No such file or directory
18:01:15 <n3npq> needs appended .sh suffix
18:01:29 <n3npq> should be a spec file patch
18:02:44 <n3npq> the file is created here (note .sh suffix)+ cp /builddir/build/SOURCES/mysql.tmpfiles.d.in /builddir/build/SOURCES/mysql.service.in /builddir/build/SOURCES/mysql-prepare-db-dir.sh /builddir/build/SOURCES/mysql-wait-ready.sh /builddir/build/SOURCES/mysql-check-socket.sh /builddir/build/SOURCES/mysql-scripts-common.sh /builddir/build/SOURCES/mysql-check-upgrade.sh /builddir/build/SOURCES/mysql-wait-stop.sh
18:02:45 <n3npq> /builddir/build/SOURCES/mysql@.service.in scripts
18:02:51 <itchka> That's the problem yes. _TPG thinks its this patch https://github.com/OpenMandrivaAssociation/mariadb/blob/master/mariadb-scripts.patch
18:03:51 <n3npq> so append .sh in this line in the patch
18:03:53 <n3npq> +    mysql-prepare-db-dir
18:04:28 <itchka> I'm pretty sure there's a syntax error in it. Look at lines 25-26 if I'm not mistaken there's an unclosed ( there.
18:04:47 <crisb> itchka: sure, i can look into mariadb
18:05:17 <itchka> crisb: Good man :) Thanks
18:05:43 <itchka> I just haven't had time.
18:06:38 <itchka> ok I guess it's the share bit again.
18:07:45 <itchka> #share ETA for QA approved beta release is on or before 16th March
18:08:16 <itchka> Ok I'm just going to feed the cats and then we will do Any Other Business
18:10:51 * n3npq would like a plan for switching away from urpmi … ask the cats what to do ;-)
18:12:48 <crisb> fedya: abf seems stuck regenerating main metadata?
18:13:17 <itchka> Just wehen everything was going so well....
18:13:50 <itchka> #item 3
18:14:00 <itchka> #topic AOB
18:14:01 <n3npq> itchka: its a multiline expansion with the closing paren on the next line. the issue seems to be purely a missing .sh extension with an additional "Make It Work!" hack in the spec file. jmho, a syntax error would not have done the copy afaict
18:14:49 <n3npq> disclaimer I cannot tell whether systemdb wants .sh or not because of lack of context in examining logs
18:14:57 <n3npq> systemd
18:15:57 <itchka> n3npq: there's no backslash for the line continuation...was what I thought but I wasn't sure what he was patching. I only had a cursory look.
18:16:04 <n3npq> re AOB: what plan for switching away from urpmi exists? if none, one is needed, lest we all wander blindly backwatds into the future.
18:16:04 <crisb> itchka: have you access to abf-downloads directly?
18:16:35 <itchka> I have access to the server hosting it.
18:17:04 <crisb> itchka: is it accessible through abf.openmandriva.org?
18:17:29 <n3npq> itchka: escaping backslash may not be needed. the cp "works" and puts mysql-prepare-db-dir.sh into the scripts/ dir where the specfile cannot find
18:18:44 <itchka> crisb: It appears not at least not in chrome. Let me try firefox
18:18:58 <n3npq> my guess is that " mv scripts/mysql-prepare-db-dir.sh scripots/mysql-prepare-db-dir" just before the install likely fixes the build
18:19:40 <itchka> crisb: I can access the index through firefox.
18:19:50 <n3npq> or at least continues the build to next problem
18:20:19 <Pharaoh_Atem> n3npq: well, I've been kind of stumped in trying to fix test failures in hawkey
18:20:47 <n3npq> Pharaoh_Atem: show me
18:20:59 <Pharaoh_Atem> hold on a moment
18:21:31 <n3npq> meanwhile hawkey is way different than deltarpm causing createrepo_c build failure as on list
18:22:34 <crisb> itchka: need someone with sudo permission on abf.openmandriva.org
18:22:49 <ashledombos> itchka: did you receive the connection info to yandex and ibilio?
18:23:10 <crisb> ashledombos: do you have sudo on abf.openmandriva.org?
18:23:30 <ashledombos> crisb: mmh i'm not sure even i have an account tthere
18:23:33 <itchka> ashledombos: I have not see it but last checked my mail early this am.
18:24:22 <ashledombos> ok all info is written about how to rsync to both these mirrors and make them T1
18:25:04 <n3npq> ashledombos: is there a ptr to the info? if so, I will try to set up a mirror @jbj.org and perhaps elsewhere
18:25:05 <ashledombos> yandex is urgent as rosa mirror will sync with it until all other mirrors change to yandex or ibiblio
18:25:49 <ashledombos> a pointer?
18:25:55 * n3npq suggests that rsync url's are deserving of # s h a r e
18:26:19 <n3npq> yes a a link to the details of the url's to be used
18:26:28 <Son_Goku> n3npq: I’m running the build for hawkey right now
18:26:29 <Son_Goku> so that I can give you the output
18:27:07 <n3npq> Son_Goku: I'm gonna need whatever changes you have made to really bore in
18:27:34 <Son_Goku> n3npq: I’ll also prepare you the SRPMs of everything if I haven’t already uploaded them to abf
18:27:37 <ashledombos> ftp://distro.ibiblio.org/openmandriva/
18:27:39 <Son_Goku> n3npq: here’s the log: http://fpaste.org/336345/45754802/
18:27:46 <n3npq> but I can likely guess whether trivial or serious issue if I have symptom details
18:27:48 <ashledombos> for ibiblio but it still empty
18:28:03 <ashledombos> it needs to be synced from abf-dl
18:28:33 <ashledombos> ftp://mirror.yandex.ru/openmandriva/ for openmandriva but old
18:28:42 <n3npq> ashledombos: thanks. sounds like "Not quite yet." if still empty
18:28:42 <ashledombos> for yandex i mean
18:28:55 * n3npq will loook at yandex
18:29:07 <Son_Goku> n3npq: looks like I’ve pushed everything up to abf: https://abf.io/ngompa
18:29:20 <Son_Goku> you’ll need my build of libsolv, too
18:29:26 <ashledombos> yes in fact we need a script that sends data from abf-dl to these mirrors
18:30:08 <ashledombos> with filtering of generally useless data such as src.rpm, debug etc
18:30:21 <ashledombos> also private folders
18:31:20 <Son_Goku> n3npq: in order for the test failures to be visible, you just need to download the hawkey sources from abf and do “rpmbuild -ba hawkey.spec —with tests”
18:31:21 <n3npq> Son_Goku: do you have ptrs to the necessary SRPM's?
18:32:00 <n3npq> Son_Goku: my builds look very very different that yours
18:32:05 <n3npq> than
18:32:31 <Son_Goku> n3npq: annoyingly enough, it seems abf didn’t generate the hawkey srpm
18:32:44 <Son_Goku> it did generate a libsolv one, though: http://abf-downloads.abf.io/ngompa_personal/repository/cooker/SRPMS/main/release/libsolv-0.6.14-1.src.rpm
18:33:52 <n3npq> Son_Goku: meanwhile (guessing) the tests seem to be relying on rpm-python in order to access rpm libraries. that is a risky assumption (and Fedora has boogered up the rpm-python with at least 3 different attempts at sub-classing last I looked)
18:34:09 <Son_Goku> the python tests pass
18:34:15 <Son_Goku> the C and valgrind tests fail
18:35:04 <Son_Goku> in the build log I pasted earlier, you’ll see the tests are divided into thirds
18:35:17 * n3npq downloads and sees
18:35:18 <n3npq> error: Bad package specification: %package -n %mklibname solv 0
18:35:27 <Son_Goku> eh?!
18:35:36 <n3npq> so I have to unwire all the OMA package macros to look
18:35:52 <Son_Goku> ah
18:35:54 <Son_Goku> sorry
18:35:57 <n3npq> Son_Goku: hint: I don't run openmadriva here.
18:36:03 <n3npq> no worries
18:36:29 * Son_Goku doesn’t like the oma macros either
18:36:41 <n3npq> I will take a look. meanwhile rpm-python not being an exact replacement is likely the problem
18:37:36 <n3npq> %mklibname is definitely spoor of mageia/mandriva packaging
18:42:38 <Son_Goku> n3npq: I’d like to kill it with fire someday
18:43:06 <Son_Goku> as well as the quirky naming convention for packages that’s used in mga/omdv
18:43:54 <Son_Goku> my own personal style for packages (which I use in Mageia) lends me to use the fewest number of distro specific macros as possible
18:44:01 <itchka> Ah the charm of history...
18:44:20 <Son_Goku> unfortunately, %mklibname is not one of those I can get away with not using :(
18:44:28 <n3npq> Son_Goku: talk to mageia/openmandriva then … meanwile %{?dist} everywhere is similarly bonkers, a very simple patch into rpmbuild would add the magic fedora urine to every package. instead, all n00bs have to endure the rigors of a mindless pkg review
18:44:49 <Son_Goku> yeah, I’d like to see that go away too
18:45:01 <Son_Goku> %mkrel and %{?dist} should both be unnecessary
18:45:24 <n3npq> au contraire: this fixed me up mostly
18:45:25 <n3npq> %define libname libsolv0
18:45:26 <n3npq> %define extlibname libsolvext0
18:45:26 <n3npq> %define devname libsolv0-devel
18:45:44 <Son_Goku> because I’m not a complete fool :)
18:45:55 <Son_Goku> I don’t use the %mklibname macro directly in the spec
18:46:16 <n3npq> and you can leave most all of the goop behind by macro expanding a spec file with the appropriate macro dialect befor trying to use a spec file
18:46:40 <Son_Goku> mmhmm
18:46:44 <n3npq> not using directly is the emprors clothing with recursions
18:47:01 <Son_Goku> well, yeah, but it makes my brain hurt less
18:47:21 <Son_Goku> and I can quickly test elsewhere by just swapping the definitions
18:47:54 <Son_Goku> at least when I do Mageia package development, I test on both Fedora and Mageia
18:48:57 <Son_Goku> I’ve been asked by no less than six people to develop some kind of “sane” scheme for packaging software
18:49:00 <itchka> Folks I have to feed the troops so if there is no structured AOB I'll leave you all to chat and close the meeting. Hope that's ok with everyone. Meeting will close in 5 minutes.
18:49:02 <Son_Goku> I’m not sure that’s even possible
18:49:30 <Son_Goku> itchka: ’tis cool
18:49:54 <Son_Goku> I view the Fedora guidelines as the least nutty, but some of the stuff it there needs to be fixed :/
18:51:32 <Kvok> Son_Goku: strange, your apostrophs are'nt displayed ok in the chwido log
18:51:45 <Son_Goku> eh?!
18:51:51 <Son_Goku> I write in UTF-8
18:52:01 <Kvok> https://chwido.openmandriva.org/meetings/openmandriva-cooker/2016-03-09/openmandriva-cooker.2016-03-09-16.19.log.txt
18:52:46 <Son_Goku> ah
18:52:57 <Son_Goku> that’s because Colloquy is changing the apostrophes
18:53:41 <Son_Goku> see ’ vs '
18:53:55 <Son_Goku> n3npq: anyway, heading back to be Pharaoh_Atem
18:54:35 <itchka> Doesn't the schizophenia get you doen:)
18:54:41 <itchka> down
18:55:25 <itchka> #endmeeting