=== maclin1 is now known as maclin [04:08] Could a MOTU please take a look at bug 1703307 and upload the package to zesty-proposed to start the SRU process? [04:08] bug 1703307 in lxterminal (Ubuntu Zesty) "[SRU] Unable to rename tabs" [Undecided,In progress] https://launchpad.net/bugs/1703307 [04:25] Could someone also nominate bug 1690416 for Xenial and Trusty please? [04:25] bug 1690416 in lxterminal (Ubuntu Artful) "[CVE] socket can be blocked by another user" [Undecided,Fix released] https://launchpad.net/bugs/1690416 [06:03] juliank: Can you gimme a poke when you're around? [06:04] juliank: Also, is it an intentional change that repositories without Release files are now considered invalid? That's a pretty major behaviour change. [06:08] infinity: As the news said, apt-get now also requires explictly setting allow-insecure-repositories [06:08] Then they work just fine [06:09] (with warnings) [06:10] juliank: Ahh. Yeah, looks like [trusted=yes] works (I think suggesting people set the apt.conf option is irresponsible, as that allows ALL sources to be insecure, not just the one I know I trust) [06:10] All other apt frontends already worked that way for some time, it's just the switch now being toggled around for apt-get to [06:10] s/to/too/ [06:15] juliank: To be fair, applying that policy to file:/ and copy:/ URIs (which is where I just had to add a trusted=yes to fix d-i) seems a bit busted. [06:15] juliank: I mean, sure, someone could hack my local filesystem and all, but I feel like I have bigger problems if I can't trust a local repo. [06:15] Well, maybe you use some partial mirroring tool, or it's on an NFS share, or whatever [06:16] Yeah, I suppose. [06:32] infinity: (re poke) Is there something else you wanted to discuss, or just that? I'm a bit slow right now, as I basically just woke up (brain is about to get carbs any minute now) :) [06:34] * juliank woke up figuring out if he actually wanted to get up or not, and then saw the messages and decided to actually get up. [06:48] juliank: That was it. Just being grumpy because you made me do work. ;) [06:49] Ah, OK; the 'also' made me think it might be two separate things :) [06:50] infinity: You are not the first one, we also broke siduction's ISO building, or whatever [07:45] hello Adri2000 seems that _rene_ wants to team upload libfilezilla (see #debian-release channel) [07:47] LocutusOfBorg: thanks for the notice, that's ok if he has an upload ready, I don't have much time to do it myself now :/ [07:47] I will do it, and upgrade to the latest release [07:49] thanks [08:06] hello, are there plans to pick up boost 1.64 for artful? [08:56] ricotz, maybe you need to talk with xnox :) [08:57] ricotz, is there a need for 1.64? [08:57] i'd rather not go ahead of debian. [08:57] 62 is currently default, with 63 available [08:58] xnox, not yet, I guess, currently libreoffice internally defaults to 1.63, but with 1.64 already available seems better to push for that [08:59] so avoiding to do two transitions [08:59] of course debian should update to 1.64 too [08:59] xnox, I guess you are the right person to ask for a MIR of libboost-locale-dev [09:00] ricotz, it doesn't need an mir. [09:00] ricotz, you should simply poke archive admin to promote that binary if something needs it. E.g. ask infinity. [09:00] ah, than just an component switch of those binaries [09:00] yes, boost itself is in main. [09:00] unfortunately. [09:00] I see, libreoffice 5.4.0 will require it [09:01] so when that is uploaded, it will be highlighted in components missmatches, and then we promote things. [09:01] I see [09:01] boost binaries move between main-universe a lot, depending on what we end up shipping in the distro. [09:02] ricotz, also if libboost-locale-dev is headers only, it needs not any promotion. can't remember if locale-dev is or isn't header/template only. [09:02] however things should be declaring built-using, but whatever. [09:02] it has a library too [09:03] ack, so it will be a normal promotion of that library package to main. [09:03] debian might go with with 1.65 transition and just skip 1.64 to be honest. [09:03] cyphermox_, xnox, slangasek, hey, any news about the nplan/n-m issue? [09:03] xnox, ok, I guess this mean artful will stay with 1.62 [09:16] Trevinho, regarding the mouse click event: i am currently testing the system with a different mouse. no issues so far. === gurmble is now known as grumble [09:47] Laney, is transition tracker ok? e.g. from http://people.canonical.com/~ubuntu-archive/transitions/html/html/ocaml.html ocaml-melt and ocamlviz are not installable on amd64, but they seem like they are to me. [09:47] or maybe i am missing something. [09:48] xnox, they are both green here [09:48] cache? [09:49] LocutusOfBorg, ah, that! thanks [09:49] problem between the keyboard and chair [09:49] symptoms to be fixed with coffee [09:49] ._. [09:50] xnox, I'm doing a photo shot of my current t-shirt [09:56] http://imgur.com/e8GUJer [09:56] xnox, ^^ :) [10:28] Hi. I'm having trouble building a package from source, perhaps someone could help, please? I'm doing the build inside a docker container so my system isn't soiled with all sorts of weird dependencies. Here's a self-contained Dockerfile: https://pastebin.com/wutSLrWq And this is how to test it, ie. make it build (and fail): docker build -t aseprite . Not that this builds fine when invoken the plain cmake/make commands (also specified in Dockerfile) [10:30] 2 minutes and I'll post the error message =) .. [10:38] https://pastebin.com/zatpVKhN [10:46] Zta77: You should really regard the packaging as (at least mostly) source code, rather than something you generate with dh_make; dh_make is usually at best only useful as a guideline. [10:46] Zta77: Anyway, I think the actual error is above what you cut off in that paste. [10:49] I'm rebuilding now; I'll paste a complete error code if possible. Perhaps there's more info further up in the scroll buffer. [10:50] I basically want to git clone some source and build a .deb that I can install later. I will start from scratch when I want to build an updated version (i.e. start a new docker container). [10:52] cjwatson: There's more here: https://pastebin.com/VBicjBHa [10:53] Zta77: Looks like missing Build-Depends: pkg-config [10:53] Possibly others [10:54] Thanks, I'll try adding them and rebuild. [10:54] (Also slightly odd that you're installing sbuild in your container but not actually using it, so with what you have at the moment you'd also need to specify additional build-dependencies in the Dockerfile) [10:54] Oh, and that error about -DCMAKE_BUILD_TYPE=None looks suspicious [10:54] Hm, no, that's debhelper's default [10:55] You might need to add -DCMAKE_BUILD_TYPE=Release or something like that [10:55] Which would go in override_dh_auto_configure [10:56] (I think sbuild is a left over from another experiment where I was building locally on my own machine) [10:56] Right. What was the first Build-Depends you mentioned? I lost my history.. [10:56] pkg-config [10:57] That's not the actual fatal error in this case (the fatal error appears to be about CMAKE_BUILD_TYPE), but I expect you'll run into it later anyway [11:07] Isn't it weird that the build succeeds if run with plain cmake/make but fails when invoked though dpkg-buildpackage ? [11:08] Zta77: Not particularly, because debhelper (used via debian/rules) supplies a bunch of cmake defaults [11:09] Zta77: One of those is -DCMAKE_BUILD_TYPE=None, which apparently the embedded libarchive doesn't like [11:09] Zta77: See /usr/share/perl5/Debian/Debhelper/Buildsystem/cmake.pm [11:10] Zta77: Like I say you can override that by adding parameters after "dh_auto_configure --" [11:13] (I think -DCMAKE_BUILD_TYPE=RelWithDebInfo is actually a better override) [11:13] See https://bugs.debian.org/701233 for the history here [11:13] Debian bug 701233 in debhelper "debhelper: Add -DCMAKE_BUILD_TYPE=RelWithDebInfo to standard set of cmake flags" [Important,Fixed] [11:18] Now I'm getting somewhere! It's building. [11:19] Using RelWithDebInfo [11:53] cjwatson: So the build actually completes .. and then it fails =) Any idea what could be going on here? https://pastebin.com/Ngs7FABP [11:54] Zta77: Packages shouldn't install stuff into /usr/local [11:55] Zta77: You should be using /usr as the prefix [11:56] Zta77: If you reeeeeeeeeeally don't want to comply with that, you should (a) reconsider (b) if you still don't want to, override dh_usrlocal to do nothing [11:56] Oh, I was just about to ask if this could be related to my custom prefix [11:57] how do I override dh_usrlocal ? [11:58] I'd like to be able to install my package side-by-side with the official package. [11:58] I don't really want to teach you the details because I think you are on a mistaken route [11:58] But if you really want to you can read "man dh" (which gives general advice on how to override debhelper commands) [11:59] There's a "Here is a way to prevent dh from running several commands that you don't want it to run" section [11:59] It's worth learning how overrides work anyway: it's easy and regular [11:59] This whole project is a mistaken route. [12:00] :) [12:03] But I'm learning docker, learning a but about .deb, learning how to populate a repository, and getting the latest build of aseprite. So it's not all bad. [13:37] apw: hello, if you have some cycles, could you review a few openstack SRU uploads for me? In the queue I have libvirt for zesty/xenial, and python-cinderclient/python-openstackclient for xenial. [13:45] coreycb, are these two identicle in the zesty queue ? [13:46] coreycb, and the same question for xenial [13:47] apw: for libvirt there should be 2 uploads for zesty and 2 for xenial, of which the older ones can be rejected [13:51] coreycb, there is already a libvirt in zesty-proposed [13:51] are we replacing that ? [13:58] coreycb: is that the update to have both rules? [13:59] cpaelzer: yes [13:59] coreycb: did you change the changelog - since I prefeched that into my merge and want history to be correct [13:59] apw: checking [13:59] cpaelzer: here's the zesty upload - http://launchpadlibrarian.net/327345846/libvirt_2.5.0-3ubuntu5.2_2.5.0-3ubuntu5.3.diff.gz [14:00] ok thats the odl message - I'm good then [14:01] cpaelzer: you probably want to release 5.2 rather than having the 5.3 version replace it, right? [14:01] cpaelzer: in zesty-proposed [14:01] coreycb: yeah 5.2 is in proposed quite a while now [14:01] I'd prefer to not restart its amturing cycle [14:03] apw: it looks like 5.2 is verified, perhaps we can release that to updates and accept 5.3 into proposed? [14:03] cpaelzer: ^ [14:04] coreycb, ok its not marked verified according to the reports, i think they may have used the wrong tag form [14:04] verification-zesty-done since 22. June [14:05] there was the global change by sil2100 - maybe affected by that in some way [14:05] apw: ^^ [14:05] i think the form is verification-done-zesty ... [14:05] Yeah, it's verification-done-zesty as apw says ;) [14:06] well I added the wrong way before you added it :-) [14:06] cpaelzer, if you could flip that (too dangerous to overlap with you) [14:06] then we can indeed release this one [14:07] apw: cpaelzer: thanks [14:07] apw: I corrected the tag [14:11] cpaelzer, coreycb, ok that one released [14:11] thanks apw [14:11] apw: thanks [14:11] i'll come back to the new one once the world notices [14:45] mbiebl_, hi. I copied (more or less) your patch for disabling bluetooth audio in gdm for Ubuntu, I wondering if it's likely to land in Debian soon, or did you decide that the a11y impact was too great? [14:45] rdiff in question: https://paste.debian.net/975780/ === JanC is now known as Guest50617 === JanC_ is now known as JanC [15:57] cyphermox_, slangasek, xnox, hey again, sorry to nag but is annoying working on unblock that n-m update? [15:57] yes, annoying is working on unblocking that [15:58] fwiw, I told jbicha two weeks ago that it was a NM regression [15:58] I think I can work around it in nplan, working on that right now [16:40] Given the recent discovery of skylake/kabylake hyperthreading problems, could we get the latest intel-microcode backported to xenial? That should fix the issue on a sizeable percentage of systems. [16:40] dmj_s76: LP: #1700373 [16:40] Launchpad bug 1700373 in intel-microcode (Ubuntu Yakkety) "intel-microcode is out of date, version 20170511 fixes errata on 6th and 7th generation platforms" [Undecided,Confirmed] https://launchpad.net/bugs/1700373 [16:48] nacc: Thanks [16:48] dmj_s76: yw [17:59] Trevinho, after one day use, i can say that the unwanted click events were caused by broken hardware === zyga_ is now known as zyga [19:55] Seems like seviper's smtp does not speak TLS? gmail shows annoying red crossed-out locks for ubuntu-devel mails :( === santa is now known as Guest89941 === Guest89941 is now known as santa_ [21:39] bdrung_work: sad to hear for you, but also good for us :-D