[03:42] Hello. Can I ask questions about package development here? [03:50] Good morning [03:51] morning Pici [03:51] pitti: [03:52] does this file tell ubuntu what the dependencies of a package are: https://bazaar.launchpad.net/~ubuntu-core-dev/apt-clone/trunk/view/head:/debian/control ? [03:53] yes [03:54] see e.g. http://people.canonical.com/~cjwatson/ubuntu-policy/policy.html/ch-controlfields.html for details [03:57] dax: This is probably off-topic, but nobody in #ubuntu is answering me. for the apt-clone software, does it clone the archived packages also? [03:57] no idea, never used apt-clone :\ [03:58] okay thank you, that is all I wanted to ask [08:05] Hi! I write app on java for ubuntu unity 16.04. Standard class SystemTray provide terrible result. Is there a native library for java to work with Tray? [08:28] A fixed apt should be ready in about 30 minutes, it's currently running its test suite on travis. Is that still early enough for the beta 1? If not, I'd delay it and sync the next RC instead [08:29] ~rc2ubuntu3 only has the emergency fixes of ~rc3 basically [08:30] Or well, we can fix it anyway, then I can delay ~rc3 a bit to get the portability fixes in [08:33] * juliank is porting APT to FreeBSD. This involves including a few more headers that are currently missing, some library magic (only use -lresolv and -ldl if the exist, add -lintl), some header magic (endian.h vs sys/endian.h), and a few test suite fixes. Really boring nitpicking kind of work [08:34] (The freebsd port also fixes a bug if you make your /bin/sh bash - stupid dashism in the test suite) [09:01] One thing I noticed today: dpkg-genchanges only seems to register "LP: #number", not "LP: number" [09:01] One APT team member has the tendency to write bug numbers without a # [09:07] right, LP: NNNN is not valid (unlike Closes: nnn for debbugs) [09:08] Yeah, maybe gbp should be teached to fix that up if it sees no # [09:12] So, large file support seems restored in apt 1.3~rc2ubuntu3, but testing can't hurt, obviously [09:13] It's 150 lines of awesome CMake code doing the detection of LFS flags. [09:15] For the platforms I test with, 2 lines would have been fine as well... [09:16] This looks a bit scary: dpkg-shlibdeps: warning: symbol __aeabi_atexit@CXXABI_ARM_1.3.3 used by debian/libapt-inst2.0/usr/lib/arm-linux-gnueabihf/libapt-inst.so.2.0.0 found in none of the libraries [09:42] uh huh [09:43] sbuild -d sid has stopped working for me [09:43] dpkg-deb: building package 'sbuild-build-depends-core-dummy' in '/<>/resolver-f_CDlo/apt_archive/sbuild-build-depends-core-dummy.deb'. [09:43] gpg: /<>/resolver-f_CDlo/gpg/trustdb.gpg: trustdb created [09:43] gpg: Warning: not using 'Sbuild Signer' as default key: No secret key [09:44] oh it's the gnupg2 thing [09:44] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=827315 [09:44] Debian bug 827315 in src:sbuild "sbuild: Does not work with gnupg 2.x installed in the chroot" [Important,Fixed] [09:46] but I merged sbuild mwhudson [09:46] are you using the one from xenial? [09:46] yeah i'm on xenial [09:46] mmm ok :) [09:46] you might want to try a backport of the yakkety version then ;) [09:47] yeah, or apparently remove /var/lib/sbuild/apt-keys ? [09:48] that appears to work, will it bite me in the bum when i try to build for trusty or something? [09:48] not sure... [09:48] I don't even remember the fix [09:48] * mwhudson tries science [09:49] https://anonscm.debian.org/cgit/buildd-tools/sbuild.git/commit/?h=debian/unstable&id=8bc84e14f0bc5fda075925ba4abbc644c1c4f2ec [09:50] + $self->log_error("To fix this problem, either (if you don't need squeeze):\n"); [09:50] there is a log error now, explaining how to fix [09:50] sbuild -d trusty seems to be working with no /var/lib/sbuild/apt-keys [09:50] haha [09:51] everything shoudl work without that directory... [09:51] just the new sbuild throws an error in case the directory is here but gnupg is not working [09:51] saying something like "hey, delete that dir" [10:11] infinity, I'm doing some work, I think I found some sort of "solution" for fpc === hikiko is now known as hikiko|ln [10:40] LocutusOfBorg: The solution is for me to bootstrap it... [10:41] LocutusOfBorg: Please don't do something hackish. [10:42] juliank: That eabi warning is harmless. There's already a filter for those symbols, it might just need updating. [10:42] infinity, the problem is that the fix seems to be not working [10:43] LocutusOfBorg: Oh. [10:43] LocutusOfBorg: Well, that's less ideal. [10:43] LocutusOfBorg: Will talk more after I've slept a bit. [10:43] infinity: Oh, OK [10:43] I copy-pasted fpcsrc/utils/fpcmkcfg/fpcmkcfg.pp from fpc 2.6.4, rebuilt fpc 3 with it and everything works [10:43] so, I'm debugging that file right now === hikiko|ln is now known as hikiko [11:43] can I somewhere see the queue of things waiting to get their autopkgtest ran for proposed migration to get a feeling for the "ETA" of an upload to be processed in that regard? [11:44] * cpaelzer searching around at http://autopkgtest.ubuntu.com/ if there is a queue somewhere [11:44] ah, sorry for the noise - found http://autopkgtest.ubuntu.com/running.shtml which kind of is what I wanted [12:04] pitti, hi, do you know any reason for which a crash file might not contain the CoreDump field? [12:11] abeato: the reason should be in /var/log/apport.log; the most common one is that it was too big to handle [12:12] pitti, hm, interesting... how is decided that it is too big? avaialble space? [12:12] abeato: > ¾ of available RAM [12:13] pitti, I see, thanks. I guess that can happen easily in touch [12:21] tumbleweed, mind syncing rails? [12:22] I think ruby-swd needs it === _salem is now known as salem_ [12:48] Hello, has anyone here worked with compiling Ubuntu ISO images with support UEFI? [12:48] they already do [12:56] pitti: i not understand your reply [12:57] JHOSMAN: ubuntu isos have supported UEFI for a long time already [12:57] yes i know [12:58] I tell you, I'm making an ISO Unattended Ubuntu, but I can not make it do the UEFI boot [12:58] A normal iso says this: http://paste.ubuntu.com/23084984/ [12:59] My iso make unnatended says: http://paste.ubuntu.com/23084986/ [13:00] in mode EFI active in bios, the iso says BosyBox and the install not running, in mode legacy run OK. [13:09] pitti: You have a guide who can share me to perform an unattended installation with Ubuntu 14.04? Perhaps I may be doing something wrong. [13:21] I don't know the intrinsics of how the official images are built, sorry [13:23] A channel or person that could refer? [13:34] coreycb: did openstack packages in xenial get accidentally uploaded again? [13:35] arges, hopefully not. I think you rejected everything in the queue yesterday so I uploaded again. [13:35] coreycb: i did, then i accepted teh packages yesterday [13:35] then this morning there are new uploads [13:35] * coreycb looks [13:36] arges, ok everything's in proposed that should be so can you reject those from the queue? sorry about that. [13:37] coreycb: ok will do [13:37] arges, thanks :) [13:38] coreycb: done. [15:26] juliank: do we need today's apt update in Beta 1 or is https://launchpad.net/ubuntu/+source/ubiquity/16.10.8 enough? [15:29] jbicha: I'd recommend it. There are issues with apt if you use --purge, it also has no support for large files which is like a huge regression from xenial. [15:29] Possibly even without --purge, I'm not entirely sure [15:29] the configures were ordered wrongly with regards to the removals in some way [16:03] hum [16:04] wgrant, pitti, dpm, it looks like we didn't have any langpack update in yakkety yet, do you know what needs to be done there? [16:04] would have been nice before beta1 [16:04] but I doubt we are going to get those now, even less of a respin [16:07] seb128, let me see if langpacks are being generated in LP for Yakketty [16:07] dpm, thanks [16:09] dpm, https://translations.launchpad.net/ubuntu/yakkety/+language-packs seems to suggest that not? [16:11] seb128, hm, looking at it from a private browser window, seems like yakkety's translations haven't been open in LP [16:11] shrug [16:11] dpm, is that a launchpad team thing? [16:12] seb128, for the opening itself, they generally do the copying of templates from the previous series before opening [16:12] https://translations.launchpad.net/ubuntu/yakkety/+translations-admin [16:14] so IIRC 1) LP team copies over templates 2) "Defer translation imports" is unchecked 3) Review if the templates look fine 4) Uncheck "Hide translations for this release" [16:14] cjwatson, wgrant, could you help us initiating the LP translations opening process for yakkety? === davmor2_ is now known as davmor2 [18:11] mdeslaur: hi, do you have plans to merge sudo? I got pinged about bug 1607666 which is fixed in 1.8.17 [18:11] bug 1607666 in sudo (Ubuntu) "sudo fails with host netgroup returned from freeipa" [Undecided,New] https://launchpad.net/bugs/1607666 [18:17] tjaalton: I wasn't planning on merging it [18:18] tjaalton: if you get an ffe exception approved, I'll do it though [18:19] mdeslaur: ok [19:18] Hi. I need a help. On ubuntu my traditional permission is not working, specially chgrp and chown [19:22] bipul: what do you mean by "is not working". Is an error message shown [19:22] chgrp: changing group of 'A/': Operation not permitted [19:22] sladen, http://paste.ubuntu.net/23085951/ [19:23] bipul: does that account have permission to change the group [19:23] I guess yes, since bipul is the owner of the file. [19:26] bipul: is the account a member of the destination group? [19:27] bipul: try running 'groups' [19:27] bipul: you ran 'id bipul' -- about a bare 'id'? if there are two bipul accounts on the system, that could account for these confusing results [19:28] sladen, strange i don't found that name in groups [19:28] bipul: how about on other filesystems? do you have other filesystems mounted that you could check? [19:28] bipul@bipul:~$ cat /etc/group | grep controller [19:28] controller:x:1005:bipul [19:29] but when i typing groups, i don't found controller there. [19:29] sarnold, No i don't have any other file system mounted [19:29] bipul: in which case either do sudo chgrp controller A/ or sudo addser bipul controller ; chgrp controller A/ [19:36] sladen, Why would i do "sudo" when i am the owner of the directory. And i have controller as a supplementary group. [19:36] I already run sudo usermod -aG controller bipul [19:37] did you use newgrp or sg to get that group added to that specific shell instance? or login afresh? [19:40] sarnold, Yes, that's what i am doing, i am restarting my machine. [19:41] between i have created controller 2 days ago and i booted my system 9 times [19:41] let me restart. [19:52] Yes, it works. === sakrecoer is now known as sakrecoer_ [21:56] in an autopkgtest is there any way for the test to get hold of the debs that were installed? [21:59] i guess the test can just build them, given appropriate depends/restrictions === salem_ is now known as _salem