/srv/irclogs.ubuntu.com/2015/06/24/#ubuntu-devel.txt

=== utlemming` is now known as utlemming
=== fginther|lunch is now known as fginther
=== work_alkisg is now known as alkisg
=== Cimi_ is now known as cimi
alkisgarges, good morning, may I ping you to put the SRU'ed x11-utils (LP: #1463663) to precise-updates? I've done the verification-done step. Thank you.04:30
ubottuLaunchpad bug 1463663 in x11-utils (Ubuntu) "[SRU] Enable setting property of type UTF8_STRING" [Medium,In progress] https://launchpad.net/bugs/146366304:30
pittiGood morning04:50
pittismoser: oh, please do ping about these, I need to know what's blocking us04:51
pittidoko: I usually override them after checking, but I didn't see it yet in the queue04:57
pitti(doing in some minutes)04:58
pittidoko: landed now05:57
=== work_alkisg is now known as alkisg
dholbachgood morning06:43
=== dbarth_ is now known as dbarth
jamespageif there are any archive-admins around I have a few binary only main promotions I need to unblock FTBFS in proposed:07:59
jamespagepython-oslo.middleware python3-oslo.middleware python3-oslo.concurrency python-oslo.config python-oslo.vmware07:59
infinityjamespage: On it.08:00
=== davidcalle_ is now known as davidcalle
jamespageinfinity, thankyou08:00
jamespageinfinity, trying to unpick all of the oslo reshuffling :-)08:00
infinityjamespage: Yeah, it's been a pain.08:10
lifelessjamespage: pip it all up :)08:27
infinitylifeless: Not sure how pip helps here.08:29
lifelessinfinity: it doesn't08:30
lifelessI was trolling08:30
infinitylifeless: And you used to be such a nice boy.08:31
lifelessinfinity: your memory is failing in your old age :)08:31
jamespagelifeless, lol08:32
infinitylifeless: I prefer to think of it not as failure, but rather the success of only remembering the good times.08:32
infinitylifeless: Which, for you, is basically just your beard.08:32
lifelessinfinity: :) nice08:32
lifelessinfinity: I see you've joined the grey club08:33
lifelessinfinity: I managed to convince Lynne to stop plucking mine.08:33
lifelessso I'm not bald08:33
infinityIt's the little victories.08:34
infinitylifeless: And yes, I'm greying a tad, though it's pretty slow going, given my age, so I can't complain too much.08:34
infinitylifeless: My brother was about 50/50 by this age, and my mother was white in her early 30s.08:35
lifelesswow08:35
lifelessgenetics huh08:35
lifeless[as in, you beat the odds in your family]08:35
infinitylifeless: My brother's also 6 inches taller, so I think he's still winning. :P08:38
lifelessinfinity: I could make a terrible joke here.08:39
infinitylifeless: Code.  Of.  Conduct.08:39
infinityjamespage: Fixed up python3-tempest-lib too, looks like a bunch of things were waiting on that.08:46
jamespageinfinity, quite likely - we've got alot of py3 enablement coming in from work in Debian08:47
jamespageinfinity, component-mismatches-proposed is just confusing me right now - it looks wrong08:47
infinityjamespage: Needs another couple of publisher cycles to sort that, then keystoneclient should be buildable, which a few other bits are waiting on.08:47
infinityjamespage: And yeah, c-m is a mess.  I need to try to attack it a bit tomorrow and see if I can unmess it.08:47
infinityjamespage: I think it's getting confused by the sheer volume of breakage we brought in this cycle.08:48
jamespageinfinity, well I fixed junit4 two days ago, which was causing the maven explosion - which should have removed alot08:48
jamespagebut that is still showing for some reason08:48
infinityjamespage: Ahh, nice.  Thanks.  Every time maven tries to promote, the world goes to crap.08:48
infinityErk.  And someone made ltsp depend on MATE, which is blowing up the world too.08:49
jamespagelol08:49
infinitystgraber: ltsp depends on MATE now?  Halp.  Is that intentional?  Is it time to drop ltsp out of main?08:50
infinitystgraber: Or do we just need to swap the gnome-session/mate-desktop-env deps?08:51
rbasakinfinity: hey. Can we chat about some dpdk packaging that smb is doing? Involving weird upstream arrangements about sonames, sovers, optimisation levels and how to package it all.09:16
rbasakinfinity: I think I have an idea about how we should be packaging it but I'd like to make sure what I'm suggesting is sane and there isn't a better way.09:16
rbasakinfinity: can we mumble with smb later today please?09:17
smbrbasak, infinity, slight caveat today is my EOD at 1500UTC09:17
rbasakI'm free from around 1300 to 1500 UTC09:18
rbasakI have to pop out for an eye test in an hour or so. Not sure exactly how long it'll take or when I'll be back.09:18
rbasak(by 1300 UTC I imagine)09:18
infinityrbasak: Hrm.  1300 UTC is 7am for me.  Not positive I'll be alive, but maaaaybe?09:19
infinityrbasak: "Weird upstream arrangements about sonames" scares me a bit, though. :P09:20
rbasakinfinity, smb: how about I ping when I'm back and we'll see if infinity is awake? Will probably be before 1300 - just not sure exactly when.09:20
smbrbasak, Would work for me though there seems to be a higher chance of awakeness when planning for 1400 utc09:20
infinityjamespage: Aaaand, now we get to python3-requests-mock.  Irksome that c-m is breaking right now, or it would have told me all of this in one shot. :/09:22
smbinfinity, could rather be phrased "weird upstream ideas about ..."09:22
infinityjamespage: Another round, and away we go.09:22
rbasaksmb: infinity's awakeness? I thought that went down from 1300 to 1400, rather than up?09:22
jamespageinfinity, yeah - feel kinda blind without that report :-)09:22
infinityrbasak: I'm more likely to be awake at 8am than 7am.  Though both seem slim, given it's currently 3:22.09:23
rbasakOh, OK.09:23
jamespagerbasak, you would enjoy the use of unversioned so names in liberasurecode :-)09:23
rbasakWell, I'll ping, and you can catch it when you're up, and we'll see?09:23
infinityrbasak: We could chat nowish/soonish instead.09:23
rbasakinfinity, smb: that works too. Mumble now?09:23
smbrbasak, not that I would even speculate about thinking to understand the awake/sleep cycles of infinity09:24
infinityrbasak: For some value of "now" that allows me to grab a bucket of ice water and a smoke first.09:24
smbrbasak, Sure09:24
infinitySo, in like 5m.09:24
rbasakOK. See you on mumble in 5m.09:24
mitya57infinity, stgraber: I think ltsp should still work fine with GNOME (Flashback), but that's also in universe…09:25
infinitymitya57: It's entirely possible that ltsp-in-main is just an artifact of "all flavours used to be in main, and edubuntu seed ltsp", and it never dropped out because it was never a problem before.  Perhaps worth revisiting.09:26
infinitys/seed/seeds/09:26
mitya57Yes09:26
infinityWell, I guess it would be in an Ubuntu seed, not an edubuntu seed, but for edubuntuish reasons.09:26
tseliotinfinity: can you check if mesa-lts-vivid is available in trusty-proposed on ppc64el, please?09:39
wgranttseliot: https://launchpad.net/ubuntu/trusty/ppc64el/libgl1-mesa-dri-lts-vivid?09:58
tseliotwgrant: right, I saw that, but one of my contacts showed me that the specific package couldn't be found09:59
wgranttseliot: Which specific package?09:59
wgrantmesa-lts-vivid isn't a binary package.09:59
tseliotwgrant: sorry, it's libegl1-mesa-drivers-lts-vivid10:00
tseliotor is it a transitional package?10:01
wgranttseliot: That doesn't exist on any architecture.10:02
wgranthttps://launchpadlibrarian.net/206323600/mesa-lts-vivid_10.5.2-0ubuntu1~trusty1_i386.changes10:02
tseliotwgrant: the instructions here (still for lts-utopic though) mention this: apt-get install --install-recommends linux-generic-lts-utopic xserver-xorg-lts-utopic libgl1-mesa-glx-lts-utopic libegl1-mesa-drivers-lts-utopic10:03
tseliotreplace utopic with vivid, and that should work, except libegl1-mesa-drivers-lts-vivid is not available10:04
tseliothttps://wiki.ubuntu.com/Kernel/LTSEnablementStack10:04
wgrantThat sounds to me like the documentation needs to be updated.10:04
tseliottjaalton: ^^10:05
wgrantlibegl1-mesa-drivers is a transitional package.10:05
tseliotright, let's see what Timo says about it10:05
tjaaltondesktop packages on ppc64el?10:07
tseliotah, so there aren't any then10:07
wgrantIs the ppc64elness relevant?10:07
tjaaltoni mean, who'd need to install those :)10:07
tjaaltonno10:07
tjaaltonthe data is old10:07
wgrantI don't think there's anything arch-specific here, apart from the q10:07
tjaaltonthing is that vivid stack isn't in updates yet10:08
tseliot-proposed is ok for now10:08
tjaaltonso just drop the packages that don't exist from the apt-get line10:08
tseliotok, thanks10:09
tjaaltonthat page won't get updated before .3 is released I think10:09
tseliotyes, and that's reasonable10:10
=== alkisg is now known as work_alkisg
=== MacSlow is now known as MacSlow|lunch
=== bduncan_ is now known as bduncan
=== MacSlow|lunch is now known as MacSlow
micahghi Laney, I saw you working on libproj transition, here's the story with that: postgis seems to be a false positive on the ben tracker (migration seems happy enough with it), zygrib has issues with qwt 6.1.1 even with 7.0, mapnik needs its own transition and I think that we should just copy back the older version since the transition seems involved (discussed this one with infinity, to be done only once the rest of the transition is ready), and12:44
micahg pyspatialite has its own mini transition as well, but that one might be easily solvable12:44
micahgI saw the libwps got entangled and tried to dig a little into zygrib last night without success12:45
Laneyhi micahg12:48
micahghi, I'm only here for a few minutes12:49
LaneyI saw that postgis had a dep on libproj012:50
Laneydidn't get much further than seeing that zygrib doesn't build12:50
micahgorly?  that's very weird, I rebuilt it twice12:50
Laneyya, look in the build log12:50
micahgand confirmed in my local build logs (and I thought on LP as well) that it was libproj912:51
Laneydo you have a path out of the jungle?12:52
LaneyI mainly started caring when it got involved with poppler and wps12:52
micahgat this point, I was planning to fix spatialite, do the switcharoo with mapnik and probably remove the binaries for zygrib  unless I could find a fix soonish12:54
micahgthat should release everything12:54
Laneymicahg: I don't see the spatialite thing, seems happy enough to go in from what I can see12:59
micahgtilelite?12:59
micahghrm13:00
micahgoh, haha, that's part of mapnik transition13:00
micahgyay, so maybe just remove binaries for zygrib + copy back the rebuild I did with the old mapnik13:01
ari-tczewcyphermox: hi, have you got any plans for merge of network-manager in wily cycle?13:03
=== _salem is now known as salem_
ari-tczewmvo: I just needed to stole your merge of ust. It's needed by another packages in universe. bug 146834513:18
ubottubug 1468345 in ust (Ubuntu) "Merge ust 2.6.2-1 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/146834513:18
mvoari-tczew: thanks!13:25
ari-tczewYou're welcome.13:28
cyphermoxari-tczew: I'd like to, but I'm busy with other stuff... and I think awe and others might want us to wait and test that phone things aren't broken by merging 1.0 or later.13:53
ogra_well, wily is already broken on the phone since the last upload it seems (not sure if NM is at fault, but i think that was the only non siloed upload yet)13:54
awecyphermox, I have a standup in a few minutes, but at some point we should discuss plans for moving to 1.x13:54
aweogra_, indeed... because we're not doing *any* testing on wily13:54
cyphermoxawe: let's do that tomorrow or something13:54
awecyphermox, ack13:54
ogra_awe, well, uploads should be siloed and self-tested ...13:55
ogra_but yeah, there is no further QA13:55
=== ValicekB_ is now known as ValicekB
=== work_alkisg is now known as alkisg
jibelogra_, awe it's true that there is very little manual QA during the release cycle of the devel release but there are integration test executed automatically when the package is uploaded. In the case of network-manager tests have been failing for nearly 2 weeks. Maybe it's the place to start https://jenkins.qa.ubuntu.com/view/Wily/view/AutoPkgTest/job/wily-adt-network-manager/16:53
ogra_jibel, well, the last two NM uploads were simple dputs to the archive16:54
aweso far... none of my NM changes for the phone have landed in wily16:55
ogra_no idea if either of them have been tested at all ( i assume the first one hasnt because that was the initial sync fromm vivid)16:55
awecyphermox and I will work this out16:55
awebut I've done *zero* testing on wily16:55
aweto-date16:55
awethe ofono changes land in both places from a single tree16:55
aweNM, not so much16:55
ogra_awe, well, and you shouldnt invest to much into it16:56
aweI haven't16:56
ogra_a simple self test and langind through a silo should be enough16:56
ogra_i suspect neither happened for the two versions wily has16:56
aweyet now there's an open bug about mobile-data being broken on wily, and some very bizarre results.  I suspect the device tarballs for mako aren't in very good shape16:56
ogra_(actually the first one wasnt the sync i thought, it was a security fix in the ofono plugin)16:57
awealso, it only looks like there's been one update16:57
awe4ubuntu1616:57
ogra_0.9.10.0-4ubuntu15.116:58
ogra_thats the first one i see on -changes16:58
ogra_synced from vivid-security16:58
aweok16:58
cyphermoxTomorrow we can make sure your nm code got landed in the packaging branch and do the testing in a silo, it's not complicated16:59
awe16 has the changes for routing which pmcgowan had reported broken?16:59
ogra_bug 143633016:59
ubottubug 1436330 in network-manager (Ubuntu Vivid) "Network Manager doesn't set metric for local networks any more, causing connection issues" [Critical,Fix committed] https://launchpad.net/bugs/143633016:59
ogra_thats the bug mentioned in the changelog16:59
cyphermoxRouting changes in wily don't break the phone, but they don't completely fix that bug either16:59
aweno pmcgowan and mdeslaur both reported that the change didn't fix the problem for them on desktop17:00
LaneyIs that meant to be in wily?17:00
aweso this probably shouldn't have landed in wily17:00
cyphermoxRight, that's what I'm saying17:00
Laneybecause it is definitely not fixed if so :)17:00
awe+117:00
cyphermoxThe fix isn't complete, but it does improve a bit17:01
LaneyI have to fettle when I want to go to/from wired17:01
cyphermoxMetric for network routes is correct now, but it's missing it for the default route17:01
awecould you update the bug?17:02
Laneyawe: are you doing something with https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1445134 btw?17:02
ubottuUbuntu bug 1445134 in network-manager (Ubuntu) "Network manager never scanning for new access points" [High,Confirmed]17:02
LaneyI see the assignee ;)17:02
cyphermoxI'm pretty sure it already was updated17:02
aweit was marked FixReleased, as an upload was done and it's in vivid-proposed, and then a bunch of folks reported it didn't fix the problem17:03
aweso maybe you could add a note about what remains to be fixed?17:03
cyphermoxWell I won't be able to do that now, I'm not home, it's a national holiday so I'm away and using my phone17:04
aweLaney, I will be looking at the scanning bug, but I have one more phone bug stacked up ahead of it.   That said, I may try and do some testing this week17:04
cyphermoxAwe: could you do the update?17:04
aweLaney, what device are you seeing this on?17:04
Laneymy laptop, assuming it is the same bug and not another one with same symptoms17:04
LaneyI posted what happend in comment #417:05
Laneygot to go now, sorry17:05
aweLaney, ok.  I'm not sure if it's the same or not.  I'll try and take a look this week17:05
aweLaney, np17:05
awecyphermox, I don't fully understand the bug and/or how it was patched.  So you're saying what remain is to backport additional logic from 1.x that sets the route metric for the default routes?17:06
Laneyif you can create an AP e.g with an android phone then it should be easy to recreate what happened to me17:06
Laneysee you17:06
aweyea, I think I should be able to handle that Laney!  ;)-17:06
cyphermoxAwe: yes. Backport being an oversimplification of the work that needs to be done17:06
awesure, but the gist of it is that the metrics aren't being set for the default routes17:07
aweI'll update it17:07
aweare you still on vacation?17:07
cyphermoxNo, but off today17:08
awek17:08
aweI have to head out to grab some lunch now, but will update it when I get back17:08
cyphermoxIf I was on vacation I wouldn't touch irc at all :-)17:08
awegood17:08
awenow enjoy the rest of your day off, and I'll catch up with you tomorrow17:08
cyphermoxSo let's take care of this tomorrow we should be able to completely fix this all17:09
infinitystgraber: *nudge*18:03
infinitystgraber: Did you see my whining about ltsp vs main vs mate last night?18:04
stgraberinfinity: yeah, ltsp shouldn't depend on mate, let me take a look18:08
highvoltagethat's really odd18:09
ogra_yeah18:09
infinitystgraber: It's in your merge.18:12
infinity-  gnome-session | x-session-manager | x-window-manager,18:12
infinity+  mate-desktop-environment | gnome-session | x-session-manager | x-window-manager,18:12
infinity(in two spots)18:12
infinitystgraber: Wasn't sure if you wanted to drop that, or change the order, or just move ltsp out of main and stay closer in sync with Debian.18:13
stgraberinfinity: we have some magic in debian/control in Debian to spit out different recommends/depends on Ubuntu than on Debian, I'll bug vagrantc to get this resolved18:23
stgraberinfinity: that's just spamming c-m right, nothing more urgent than that? (otherwise I can just cowboy the fix in Ubuntu and then merge again whenever vagrantc fixes it in Debian)18:26
infinitystgraber: Yeah, just a lot of spam.18:27
=== work_alkisg is now known as alkisg
=== salem_ is now known as _salem
infinitystgraber: Do you have any insight into the lxc autopkgtest regression?20:17
infinitystgraber: Is that just a "we need to fix network access" thing, or something else?20:18
stgraberinfinity: my guess is that it's network related, yes20:18
infinitycyphermox: How about you?  Any ideas on the network-manager adt regression?20:20
cyphermoxinfinity: I think there has been a change in something on the system running the tests, if it's still that wpa error20:26
infinitycyphermox: It's that, yes.20:26
infinitycyphermox: So, if you're sure it's a testbed problem, I'll let the world skip it for now, and we can sort it with pitti.20:27
cyphermoxYeah, feel free to ignore it I'd say.20:27
cyphermoxNm hasn't exploded here, in any case20:27
infinitycyphermox: Well, I assume you also don't run proposed.20:27
cyphermoxNo, but I'm not sure what would break things that badly.20:28
cyphermoxIs there a new wpa in proposed?20:28
infinitycyphermox: New wpa for security issues, but that came a day after the test started failing.20:28
infinitycyphermox: The only obvious change I can see around when it started failing was a dnsmasq rebuild against a newer libnettle.20:29
infinitycyphermox: If dnsmasq exploding could cause those sorts of errors, that could be it.  If not, then it's probably the testbed.20:29
cyphermoxThat wouldn't break wifi20:29
* infinity nods.20:29
cyphermoxIt seemed more like some hostapd issue with the hardware on the system, so maybe the kernel / hwsim module changed enough to break things, or there really is a wifi device on the test system now and things are confused20:31
cyphermoxinfinity: I already have other things for NM in to-do for tomorrow, so I will debug this too20:32
infinitystgraber: Okay, from looking at your tests and where this appears to time out, I concur it's probably network sadness.  Would be nice if your tests failed more verbosely, mind you. :)20:32
infinitystgraber: When pitti moves all of this, I think I can get the right holes punched to let these tests work instead of neutering your testsuite, FWIW.20:33
stgraberinfinity: typically it's gpg getting stuck on no-network and it never seems to timeout for some reason20:33
infinitystgraber: Might be worth a skeleton "test-host-port" function that you can reuse to bang every host/port combo you intend to abuse and die early if they're broken with a useful error.20:34
stgraberindeed20:34
infinitystgraber: Alright, I'm going to badtest lxc for now.  It's a lie, but it's the best lie I can come up with.20:38
infinitycyphermox: Ditto for n-m.20:39
cyphermoxOK20:40
=== athairus is now known as afkthairus
=== alkisg is now known as work_alkisg
=== _salem is now known as salem_
Unit193cjwatson: Sorry to bother again, but got any estimate on the newer openssh entering Debian/Ubuntu?22:14
cjwatsonUnit193: you mean 6.8p1?22:18
Unit193Yes sir.22:19
cjwatsonUnit193: I'm in the middle of the very very complicated rebase of the gssapi patch22:20
Unit193Oi, sorry about that. :322:20
cjwatsonUnit193: once that's done it should be straightforward, but I don't yet have a time estimate; I'm on holiday this weekend so it won't be then22:21
Unit193cjwatson: Sure that's fine, and understandable on both accounts.22:22
Unit193Thanks for the information!22:23
=== afkthairus is now known as athairus

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!