19:04:32 <wilcal> #startmeeting
19:04:32 <Inigo_Montoya`> Meeting started Thu Aug 27 19:04:32 2015 UTC.  The chair is wilcal. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:04:32 <Inigo_Montoya`> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:04:42 <wilcal> cool I can start things
19:05:07 <lewyssmith> Only you can be Dave at present.
19:05:30 <wilcal> #chair lewyssmith
19:05:30 <Inigo_Montoya`> Current chairs: lewyssmith wilcal
19:05:47 <wilcal> you can do the dave honors for now lewys
19:05:48 <lewyssmith> #topic Updates
19:06:10 <wilcal> i got really spun up on a couple of the bugs
19:06:27 <lewyssmith> Dont't tell me: VirtualBox.
19:06:49 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=16584
19:06:51 <[mbot> Bug 16584: normal, Normal, qa-bugs, NEW , audit new security issue CVE-2015-5186, audit-2.4.3-1.mga5.src.rpm
19:07:04 <wilcal> there's no consensis on how to test this
19:07:33 <wilcal> I'd like to do a if it installls and updates ok push it
19:08:03 <lewyssmith> Ah.  Advantage of being for M4 & 5. Agree with your verdict.
19:08:40 <wilcal> For now we're waiting for M4 to go obsolete ( 17 Sept  i think )
19:08:59 <lewyssmith> We are going to have to do the same for Bug 16515 and Bug 16249.
19:09:01 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=16515 normal, Normal, qa-bugs, NEW , openafs new security issues CVE-2015-328[2-5] and CVE-2015-3287, openafs-1.6.11-1.mga5.src.rpm
19:09:02 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=16249 normal, Normal, qa-bugs, ASSIGNED , roundcubemail new security issues CVE-2015-218[01] and CVE-2015-5382, roundcubemail-1.0.5-4.mga5.src.rpm
19:09:09 <wilcal> then let Vbox go. I don't understand why I can't get the extensions working
19:09:36 <wilcal> Ya the roundcube thing requires you to make a career out of it to test it properly
19:09:47 <lewyssmith> Just install the originals and make sure the updates do not clag. (Two bugs above).
19:10:17 <wilcal> I think that's a plan
19:10:17 <lewyssmith> So are we agrred to do this for the three bugs noted above?
19:10:38 <wilcal> yes
19:10:38 <lewyssmith> That will give is 5 OK's.
19:10:42 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=16498
19:10:43 <[mbot> Bug 16498: major, Normal, qa-bugs, ASSIGNED , hplip new security issue CVE-2015-0839, hplip-3.14.6-8.mga5.src.rpm
19:11:08 <wilcal> imo that one should be taken out off the list. It's really bad
19:11:36 <wilcal> I think someone said that they would look at it after they came back from vacation
19:11:42 <lewyssmith> I do not understand hplip. Is it relevant if you do not have an HP printer?
19:11:59 <lewyssmith> hviaene: Hello & welcome.
19:12:03 <wilcal> it works for some folks and not at all for others
19:12:07 <hviaene> Hi all
19:12:27 <lewyssmith> wilcal: And my question?
19:12:46 <wilcal> I have a brand new HP 4502 printer and installing 16498 it becomes unusable
19:12:59 <wilcal> my notes are in the bug
19:13:19 <wilcal> M5 with present hplib it works just fine
19:13:31 <lewyssmith> I do not have any HP printer. Can I test hplip? If so, how?
19:13:56 <tmb> nope. hplip are hp-specific
19:14:11 <wilcal> yes
19:14:17 <tmb> and really only needed for multifunction devices
19:14:26 <wilcal> and that's what I have
19:14:31 <lewyssmith> Thanks. I can ignore it with a clean conscience.
19:14:51 <wilcal> and the HP 4502 is the most popular HP printer on the US market right now
19:15:05 <wilcal> and it don't work with that
19:15:21 <lewyssmith> Welcome Dave.
19:15:29 <DavidWHodgins> Sorry I'm late
19:15:29 <wilcal> hello david
19:15:35 <wilcal> no prob
19:15:36 <hviaene> hplip is also a choice when you have a plain printer
19:16:07 <hviaene> And I always take that option
19:16:30 <wilcal> anyway it really does not work. I've tested it many times
19:16:36 <lewyssmith> DavidWHodgins: How are you faring with Advisories? We have several single OK'd bugs we can push otherwise.
19:17:00 <lewyssmith> wilcal: We clearly cannot push this yet. Grey it?
19:17:09 <DavidWHodgins> I haven't touched them in a while. tmb has been doing most of them.
19:17:11 <wilcal> grey it is fine
19:17:30 <wilcal> lets push roundcube
19:18:10 <lewyssmith> tmb: Thanks a lot for the advisories. I ma too busy with updates to learn them just now.
19:18:21 <DavidWHodgins> There are no ok's for roundcube yet.
19:18:45 <wilcal> David you ever test audit?
19:18:54 <lewyssmith> wilcal: We agreed this above, but it must be OK'd first. I will do all those 3  bugs.
19:18:59 <DavidWHodgins> Not that I recall
19:19:05 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=16584
19:19:06 <[mbot> Bug 16584: normal, Normal, qa-bugs, NEW , audit new security issue CVE-2015-5186, audit-2.4.3-1.mga5.src.rpm
19:19:51 <wilcal> I can't find a concensis on how to test that and I spent hours Monday trying to test it
19:20:10 <lewyssmith> We are in the same boat with openafs: install, update, OK.
19:20:47 <wilcal> tmb should we work the kernel bugs and continue to wait on the Vbox bug
19:20:48 <DavidWHodgins> I think for both of those, if the update installs cleanly, we should ok and validate them.
19:20:57 <wilcal> agreed
19:21:47 <lewyssmith> DavidWHodgins: We agreed this for  roundcubemail, openafs, audit. I will do.
19:22:14 <wilcal> anyone got this hardware?
19:22:16 <wilcal> https://bugs.mageia.org/show_bug.cgi?id=16390
19:22:18 <[mbot> Bug 16390: normal, Normal, qa-bugs, NEW , broadcom wifi: kernel panic with static IP & with suspend/resume, kmod-broadcom-wl-6.30.223.248-33.mga5.nonfree.src.rpm
19:22:32 <lewyssmith> wilcal: If you want to grey both hplip updates, over to you.
19:22:58 <lewyssmith> 16390: I put out an appeal on the ML; no response.
19:23:03 <wilcal> I've never done the grey out thing but lemme see if i can figure it out
19:23:17 <tmb> comment 5 on bug 16390 confirms its working
19:23:17 <Luigi12_work> for openafs, with dkms-afs if the kernel module builds fine, that's sufficient testing
19:23:18 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=16390 normal, Normal, qa-bugs, NEW , broadcom wifi: kernel panic with static IP & with suspend/resume, kmod-broadcom-wl-6.30.223.248-33.mga5.nonfree.src.rpm
19:23:22 <hviaene> I have a laptop with Broadcom chipset
19:23:24 <lewyssmith> You just put 'feedback' on the Whitboard.
19:23:32 <Luigi12_work> for roundcube, someone said there was an installation/dependencies problem, so that'll need to be addressed first if true
19:23:44 <Luigi12_work> for hplip, it probably needs to be reverted to the previous version and then patched for the security issue
19:23:45 <hviaene> And i use static ip's everywhere
19:23:57 <Luigi12_work> in mga5, the updated hplip version was already rebuilt in backports testing
19:24:20 <Luigi12_work> I hope to test virtualbox on mga4 this weekend, if it looks good there, I'm comfortable pushing it
19:24:39 <hviaene> never had any problem, apart from the installation issues
19:24:47 <Luigi12_work> for audit, it looks like the testing procedure was close, but there was some sort of syntax error in the line that was added to audit.rules, so check the man page for that
19:25:08 <lewyssmith> Luigi12_work: Openafs: Can you note that in the bug, please? With 'howto'.
19:26:27 <lewyssmith> hviaene: Can you try the update 16390  & OK it if it still works?
19:26:59 <hviaene> I'll check that tomorrow
19:27:03 <wilcal> i tinkered with audit for a couple hours on Monday and was never able to find a workable proceedure
19:27:31 <wilcal> the proceedure was close but never worked
19:27:56 <lewyssmith> What are we going to do about Bug 16072?
19:27:58 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=16072 normal, Normal, qa-bugs, NEW , Upstream fix (?) at resume screen black while .xsession-error fills w msg about bluedevil, bluedevil, libbluedevil
19:28:25 <lewyssmith> Again a ML appeal went unanswered.
19:28:30 <Luigi12_work> for audit, it looks like the rule is missing the "permission" part, according to the man page
19:29:01 <lewyssmith> Again, David - a note in the bug about that?
19:29:32 <wilcal> I could never get "permission" felt like a 6ht grader for hours
19:30:07 <DavidWHodgins> As per bug 12977, we've never tested more than weather or nor audit installs cleanly.
19:30:09 <[mbot> Bug https://bugs.mageia.org/show_bug.cgi?id=12977 critical, Normal, qa-bugs, RESOLVED FIXED, audit initscript causes ordering cycles at boot leading to dbus+NetworkManager failure incl. KDE plasma applet crashes, auditd
19:30:10 <Luigi12_work> it just tells you that it needs to be -p and then either r, w, x, or a
19:30:14 <Luigi12_work> note posted in the bug
19:30:34 <lewyssmith> Thanks!
19:30:58 <lewyssmith> All your comments really help a lot.
19:31:36 <wilcal> imo always post copious notes
19:31:55 <DavidWHodgins> wilcal: Agreed
19:32:03 <lewyssmith> OK, back to BlueDevil: we do what in the absence of a bluetooth user?
19:32:12 <DavidWHodgins> Just not too copius.
19:32:13 <Luigi12_work> push it if it installs cleanly
19:32:24 <lewyssmith> Will do.
19:32:28 <DavidWHodgins> Yes
19:32:46 <Luigi12_work> we need to get those mga5 bugfix updates pushed
19:33:28 <lewyssmith> That makes an uncomfortable no. of updates to OK blindly, just on not causing obvious grief.
19:33:45 <DavidWHodgins> I'll try to get the advisories uploaded to svn later tonight. That's the only thing holding them back.
19:33:56 <Luigi12_work> if nobody cares to test it or say anything and it's sat for a month, push it
19:34:04 <Luigi12_work> trust that the packager fixed what they intended to fix
19:34:20 <Luigi12_work> anyway most of them do have an OK
19:34:26 <lewyssmith> More a case of not being able to test things, David.
19:34:45 <Luigi12_work> nonsense, *someone* is capable of testing it, if someone out there is using it
19:35:04 <Luigi12_work> if those who use it don't bother to help test it, that's on them if we don't catch an issue with it
19:35:11 <Luigi12_work> if it's that hard to figure out
19:35:19 <DavidWHodgins> Not always. We have a lot of packages that are likely not being used at all.
19:35:24 <lewyssmith> Yes, the real users. We have a bug awaiting such feedback. There are also h/w restraints.
19:35:46 <Luigi12_work> DavidWHodgins: when you suspect that's the case, feel free to send a note to the dev list asking for the package to be removed from Cauldron
19:36:01 <DavidWHodgins> Will do.
19:36:16 <Luigi12_work> again, if nobody bothers to help test who can, trust that the update fixes what it was intended to fix
19:36:31 <lewyssmith> I have often wondered whether we have any way of knowing what software *is* used.
19:36:35 <Luigi12_work> the QA team as it stands does the best they can, but they can't do *everything*
19:36:51 <wilcal> agreed
19:36:57 <Luigi12_work> lewyssmith: no we have no way of knowing really
19:37:06 <Luigi12_work> if we remove a package and someone complains, then we can always put it back
19:37:40 <DavidWHodgins> Luigi12_work: Agreed. If the update of the package installs cleanly, but we don't know how to test it, or don't have the hardware, go ahead and ok/validate it.
19:38:12 <Luigi12_work> yeah.  It's more likely that it does fix what it was intended to fix, so it's not good to just have it stagnate for several weeks.
19:38:33 <lewyssmith> Would it be possible to monitor at download source what packages are specifically installed by users?
19:38:41 <DavidWHodgins> Especially for security bugs.
19:38:49 <Luigi12_work> lewyssmith: not without driving away most of our user base
19:38:53 <wilcal> there's 30,000 rpms in the M5 i586 repo
19:38:55 <DavidWHodgins> lewyssmith: No
19:39:13 <Luigi12_work> in the Linux world, you can't get away with spying on your users, they won't tolerate it
19:39:21 <Luigi12_work> they're not Windows-using sheep
19:39:21 <lewyssmith> I mean simply a boolean per package, that *someone* has installed it.
19:39:29 <Luigi12_work> again, no
19:39:46 <wilcal> If you want that move to WinBlows 10
19:39:48 <DavidWHodgins> We can't monitor what's installed from a mirror.
19:39:52 <Luigi12_work> that'd require one of two things, monitoring the mirrors which we don't have access to, or having Mageia systems "phone home" which will piss everyone off
19:40:28 <Luigi12_work> would it be nice to have package usage statistics?  Yes.  Is there a good easy way to do it?  Not really.
19:40:37 <DavidWHodgins> I'd vote against that idea if it were ever proposed.
19:40:43 <Guest53403> Brian__
19:40:47 <Luigi12_work> Debian has an opt-in thing that reports your installed packages, so they do have *somewhat* of an idea of what's used and how widely
19:40:54 <Brian__> hellow
19:40:55 <Luigi12_work> so something like that *could* be implemented I suppose
19:41:00 <wilcal> hello bria
19:41:38 <DavidWHodgins> Luigi12_work: I'd be against it.
19:42:05 <Luigi12_work> if it was opt-in, it needn't be objectionable, but yeah I'm still not a huge fan of things that phone home
19:42:27 <Luigi12_work> I've done a lot of work in one of the packages I maintain to disable such things
19:42:33 <lewyssmith> I have just installed the desktop kernel update on a Radeon box, & it took out fglrx! Does not seem right.
19:43:09 <Luigi12_work> yeah that shouldn't have happened, since fglrx was rebuilt for this kernel update
19:43:13 <tmb> lewyssmith, did you have nonfree updates_testing enabled ?
19:43:19 <lewyssmith> Yes.
19:43:21 <Luigi12_work> as opposed to nvidia, broadcom-wl, and virtualbox
19:43:22 <DavidWHodgins> lewyssmith: i586 or x86_64?  Mageia 4 or 5?
19:43:33 <lewyssmith> 5/64
19:44:04 <DavidWHodgins> Perhaps the mirror wasn't fully synced at the time.
19:44:14 <lewyssmith> I only use the best.
19:44:48 <lewyssmith> I will ask on the ML after poking around.
19:45:06 <DavidWHodgins> Any big updates, like kernels, kde, gnome, etc., I always wait a few hours before trying to update.
19:45:46 <wilcal> rsync to my local mirror and let a day go by and do it again then test
19:46:02 <lewyssmith> Hello Ben.
19:46:21 <Benmc> Good morning Lewis, QA
19:46:53 <DavidWHodgins> Hiya Benmc
19:47:39 <tmb> lewyssmith, you caught a bug / there is a wrong conflict in the kernel, so I need another rebuild of the kernel or rebuild the fglrx stack to match
19:47:44 <DavidWHodgins> I rsync my local repo from mirrors.kernel.org/mageia once every hour.
19:47:53 <tmb> sorry about that
19:48:14 <wilcal> i do the same once a day
19:48:36 <wilcal> but for cauldron I rsync then go back and do it again
19:48:59 <lewyssmith> tmb: I have said nothing yet on the bug. Yet to re-boot. But Updates Testing has been wanting to take out fglrx for ages before doing the kernel update.
19:49:03 <tmb> I thought I had it fixed, but I only fixed the kmod package not the main fglrx one
19:49:17 <DavidWHodgins> Got an email yesterday from my isp. I'm now over 50% of my bandwith limit.
19:50:00 <DavidWHodgins> With only 5 days left, I'm not worried about it.
19:50:02 <Brian__> what do they give you per month?
19:50:10 <DavidWHodgins> 400 GB
19:50:12 <lewyssmith> tmb: I removed orphaned fglrx; do I need to re-install it?
19:50:51 <tmb> lewyssmith, yeah, the reason I conflict old drivers is to force new ones that is known to work with the kernel... but this one I screwed up... but the probuilt fglrx mods should still work afaik
19:51:50 <tmb> so you should be able to urpmi fglrx-kernel-desktop-latest
19:52:17 <lewyssmith> From Testing? After you have fixed it?
19:52:47 <tmb> from testing... and it should work already...
19:53:13 <tmb> since I only conflict the dkms package
19:53:19 <DavidWHodgins> Run the urpmi command twice, and it will never be marked as an orphan again.
19:53:58 <Luigi12_work> once will suffice as long as you explicitly list any package names you don't want orphaned
19:54:17 <lewyssmith> The update took forever. I wonder whether I am hitting the 10m wait grub problem I had with M5 testing.
19:55:04 <DavidWHodgins> lewyssmith: IIRC, that seems like a hardware problem.
19:55:22 <lewyssmith> OS-prober was the culprit.
19:55:35 <Luigi12_work> ahh grub2, yuck
19:56:06 <lewyssmith> Enough of this. We have plenty to get on with.
19:56:29 <wilcal> Ya keep plug'n away on the updates
19:56:47 <wilcal> i touch one a day but this week has been frustrating
19:56:55 <DavidWHodgins> I have seven linux installs, and haven't had a problem. 2 on an ssd drive,  the rest on a spinning hard drive.
19:57:17 <wilcal> M5 is really quite solid right now
19:57:28 <wilcal> the polkit thing still plagues me
19:57:36 <wilcal> whatever it is
19:57:44 <Luigi12_work> it shouldn't
19:57:44 <lewyssmith> DavidWHodgins: About the same here, all on a hard disc.
19:57:52 <DavidWHodgins> It was solid when it was (delayed) released.
19:58:23 <lewyssmith> Change topic?
19:58:29 <wilcal> yes please
19:58:38 <lewyssmith> #topic Luigi's Roundup
19:58:47 <wilcal> it's luigi time
19:58:48 <DavidWHodgins> wilcal: With all updates installed, polkit shouldn't still be causing problems.
19:59:33 <wilcal> i only see it on my machine that's running 7/24/360
19:59:35 <Luigi12_work> indeed, it was just a one-time problem when the update was first installed.  There was no persistent problem associated with it.
19:59:37 <lewyssmith> [Roundup capitalised looks like that well-known poison].
19:59:40 <Luigi12_work> you see "what" exactly?
20:00:01 <wilcal> MCC won't launch from icon. launches from su terminal
20:00:15 <wilcal> drakconf
20:00:15 <Luigi12_work> have you ever restarted the polkit service since you installed that update?
20:00:20 <wilcal> yes
20:00:36 <lewyssmith> That is known to be intermittent on some desktops, down to two known bugs.
20:00:42 <wilcal> reboot the system works every time
20:00:54 <wilcal> restart x11 sometimes fixes it
20:01:10 <DavidWHodgins> Hibernate/sleep problem?
20:01:11 <papoteur> Hello, I don't agree with wilcal
20:01:15 <lewyssmith> Related to shutting down with applications still running.
20:01:31 <lewyssmith> Bonsoir Yves.
20:01:39 <papoteur> hello lewyssmith
20:01:46 <wilcal> Yes my 7/24/360 does go into hibernate mode
20:01:46 <Luigi12_work> well if you can come up with a reproducible test case for showing it still has a problem, there may be a different issue with it
20:01:54 <Luigi12_work> anyway, roundup
20:02:05 <wilcal> go for it luigi
20:02:08 <lewyssmith> [with a small 'r']
20:02:21 <Luigi12_work> I've still been insanely busy at work, so I still need packager help if we're going to get any security updates done.  I post things to the dev list and hear crickets.
20:02:44 <Luigi12_work> in the past week, new bugs filed for php-ZendFramework, php-ZendFramework2, jasper, util-linux, and uglify-js
20:03:03 <Luigi12_work> php-ZendFramework update is ready.  php-ZendFramework2 is fixed in Cauldron, but the packager failed to respond to my question of what to do about Mageia 5.
20:03:15 <Luigi12_work> jasper I'm waiting on usable patches, which aren't currently available AFAIK
20:03:26 <Luigi12_work> util-linux I had a question about that I posted to the dev list and received no replies
20:03:28 <DavidWHodgins> :-)
20:03:40 <Luigi12_work> uglify-js is assigned to Joseph Wang
20:03:56 <Luigi12_work> there are several older ones waiting too
20:04:05 <DavidWHodgins> Sorry, I just had my cat jump onto my lap. Making it hard to type.
20:04:24 <lewyssmith> You managed that sentance fine!
20:04:28 <Luigi12_work> I'll still be very busy through the end of the year
20:04:36 <Luigi12_work> that's all, unless there's questions
20:04:49 <DavidWHodgins> wilcal: You or lewyssmith will have to do the #info messages as I'm not a chair.
20:05:01 <wilcal> k
20:05:07 <lewyssmith> I leave that to Bill.
20:05:12 <wilcal> #topic Anything Else?
20:05:20 <wilcal> it's working for me today
20:05:30 <DavidWHodgins> Thanks Luigi12_work
20:05:48 <DavidWHodgins> Nothing else from me.
20:05:55 <wilcal> any tips tmb why i can get the extensions working in my vbox5 test?
20:06:22 <DavidWHodgins> wilcal: can, or cannot?
20:06:42 <wilcal> I can't get the Vbox5 extensions working
20:07:11 <wilcal> won't see a usb drive with extensions installed
20:07:23 <tmb> wilcal, what are you using to download the extensions ? I dont have any problem downloading them... firefox even suggests to hand it over to vbox directly and it installs the extension pack
20:07:58 <wilcal> i didn't know it would self install the extensions
20:08:15 <tmb> wilcal, I mean if you open http://download.virtualbox.org/virtualbox/5.0.2/Oracle_VM_VirtualBox_Extension_Pack-5.0.2-102096.vbox-extpack
20:08:33 <wilcal> anyway no bigs i'll look at it again
20:08:41 <DavidWHodgins> It's a bug in vbox. Either run vb as root, to install the extensions, or (also as root) use VBoxManage to uninstall the old version and install the new versions.
20:08:49 <wilcal> let M4 go eol and push Vbox 5
20:08:59 <tmb> in firefox you get the popup with "Open with" and it suggests virtualbox
20:09:15 <Luigi12_work> wilcal: no reason we *need* to wait for mga4 EOL if the mga4 update works fine too
20:09:40 <wilcal> only problem is reallly M4 sound and that's not a big problem
20:09:43 <Luigi12_work> I'd like to push it next week if it works fine
20:09:50 <Luigi12_work> wilcal: didn't we solve the sound issue?
20:10:05 <DavidWHodgins> Note that if the file ends with .gz, it must be renamed to end with vbox-extpack
20:10:14 <wilcal> for M5 not for M4
20:10:32 <Luigi12_work> hmm well I'll hopefully try it this weekend
20:10:40 <DavidWHodgins> Sound is working ok here.
20:10:41 <tmb> DavidWHodgins, well as you see the above link it is properly named...
20:12:28 <DavidWHodgins> tmb: Yes, that link seems to be named ok, but when I first installed the update, and downloaded the extension, it had a name ending with .gz, instead of vbox-extpack, so vb wouldn't see it.
20:12:42 <DavidWHodgins> Likely they've fixed it.
20:13:01 <tmb> DavidWHodgins, yeah, maybe the vbox upstream messed up at some point...
20:13:14 <DavidWHodgins> I think so.
20:13:14 <lewyssmith> Countdown?
20:13:25 <DavidWHodgins> lewyssmith: Fine with me.
20:13:29 <wilcal> yes count it down
20:13:35 <lewyssmith> -5
20:13:41 <lewyssmith> -4
20:13:45 <lewyssmith> -3
20:13:48 <lewyssmith> Thanks to all who came.
20:13:52 <lewyssmith> -2
20:13:56 <lewyssmith> -1
20:14:00 <lewyssmith> 0
20:14:03 <wilcal> #endmeeting