[00:00] <mwhudson> that  1.7.3-1ubuntu2 was published
[00:00] <mwhudson> https://launchpad.net/ubuntu/+source/golang-context/1.1-1ubuntu5/+build/11088558
[05:37] <cpaelzer> thanks for the RC nacc
[05:37] <cpaelzer> nacc: only pending qemu SRU for Trusty, and incoming libvirt for Z/Y/X but takes while
[05:37] <cpaelzer> nacc: so qemu in X is free for you if you got something
[05:54] <cpaelzer> good morning
[07:01] <pitti> Good morning
[07:31] <smb> nacc, I got nothing pending (neither qemu nor libvirt) right now
[10:34] <doko> LocutusOfBorg: please stop trying to discard needed llvm changes in Ubuntu
[10:37] <LocutusOfBorg> doko, context please?
[10:38] <doko> [Bug 1632357] Re: Sync llvm-toolchain-3.6 1:3.6.2-4 (main) from Debian unstable (main)
[10:38] <LocutusOfBorg> doko, 1) llvm-toolchain-3.6 is removed in unstable, and -4 contains all the delta AFAIK
[10:39] <LocutusOfBorg> https://packages.qa.debian.org/l/llvm-toolchain-3.6/news/20161007T184934Z.html
[10:39] <LocutusOfBorg> 2) I would like to demote/remove it
[10:40] <LocutusOfBorg> pretty sure I didn't drop stuff
[12:44] <pitti> hey
[12:45] <pitti> sil2100: it's simple, ppc64el is still squeaking under the perl load; amd64/i386 just caught up
[12:45] <pitti> sil2100: it's in http://autopkgtest.ubuntu.com/running#queue-ubuntu-zesty-ppc64el
[12:45] <pitti> i. e. queued
[12:47] <LocutusOfBorg> the queue looks too green for perl
[13:01] <sil2100> pitti: ah, thanks! :)
[13:18] <doko> hmm, how do I make a haskell build verbose?
[13:18] <doko> slangasek: ^^^
[13:40] <LocutusOfBorg> doko, "-v"
[13:41] <LocutusOfBorg> -v[0-4] gives you the right verbosity
[13:53] <doko> LocutusOfBorg: passed to?
[13:56] <LocutusOfBorg> ghc build
[13:57] <LocutusOfBorg> "haskell build"  has no meaning, you want to build ghc in verbose mode or an haskell library?
[13:57] <LocutusOfBorg> I did an hello world and used -v as flag
[13:57] <LocutusOfBorg> it worked
[14:29] <xnox> has arm64 switched to GLES by default for qt5?
[14:30] <mitya57> xnox, yes, since 5.6.0+dfsg-2ubuntu1
[14:30] <xnox> mitya57, ok, and is there non-GLES available for arm64? E.g. i know we do compile two types of qt stacks on some arches in ubuntu.
[14:31] <mitya57> I think it works only in reverse direction...
[14:31] <xnox> ok.
[14:43] <nacc> smb: thanks!
[14:55] <nacc> smoser: fyi, apt should import now (fixed the usd-import-team tree)
[15:15] <nacc> LocutusOfBorg: did you see that -4 was removed from debian altogether though?
[15:16] <LocutusOfBorg> nacc, yes, this is why the bug is now "please remove llvm-3.6 from zesty"
[15:16] <LocutusOfBorg> btw I'm the one who is maintaining llvm-* in debian :)
[15:17] <LocutusOfBorg> at least I'm partially working on it
[15:17] <nacc> LocutusOfBorg: ack, i gathered as much :)
[15:19] <LocutusOfBorg> and removing llvm-3.6 means, somebody have to merge qtcreator and clamav
[15:43] <nacc> LocutusOfBorg: yep :)
[15:44] <nacc> caribou: --^ i assume you are planning on that, or I can put it on my plate for z (for clamav)
[15:44] <LocutusOfBorg> qtcreator needs somebody working on it, I don't feel confident enough in a merge
[15:46] <nacc> LocutusOfBorg: delta seems pretty minimal, but yeah, probably deserves testing/experience by someone who knows :)
[15:46] <LocutusOfBorg> nacc, if you have a patch I'm happy to test it
[15:46] <LocutusOfBorg> a new release is needed for the compatibility
[15:46] <nacc> LocutusOfBorg: ah
[15:47] <nacc> LocutusOfBorg: i can look at it later today, maybe
[15:47] <LocutusOfBorg> thanks
[15:47] <LocutusOfBorg> I have permissions to work on it from the current maintainer
[16:09] <ioria> hello ! i made a simple textual editor for .xpm icons... where i can upload it to share ?
[16:31] <nacc> smoser: i think i can remove at least one call to dpkg-parsechangelog with a bit of reorg
[16:32] <smoser> memoize makes magic faster.
[16:32] <smoser> and removes 1/2 in my experience. i think for each branch we end up re-calling
[16:35] <nacc> smoser: do you have the current list of failures?
[16:36] <smoser> i can try to find
[16:53] <smoser> nacc, at this point i think everything is working that i've gotten to.
[16:54] <nacc> smoser: so the backoff fixed the download errors you were getting before?
[16:55] <smoser> well, no. but i'm ignoring those for now.
[16:55] <nacc> smoser: ok
[16:55] <smoser> they still occur fairly regularly.
[16:55] <nacc> smoser: note that i can't reproduce them
[16:56] <smoser> but are you trying hundreds of packages ?
[16:56] <nacc> smoser: are you running these in parallel?
[16:56] <smoser> yeah
[16:56] <nacc> ah
[16:56] <smoser> but that really shouldnt matter.
[16:56] <nacc> that's probably launchpad getting mad at you?
[16:57] <nacc> :)
[16:57] <nacc> i hope you talked to the launchpad folks first
[16:57] <smoser> i'mi doing 16 at a time on that vm, load sits about 10
[16:57] <smoser> well, i'm tyring more hard now to save the files i download (that was what i said i'd like as a feature)
[16:58] <smoser> to be able to save .dsc and referenced files off to another directory.
[16:58] <nacc> smoser: but why do you need them (generally)?
[16:58] <smoser> i am a load, for sure, but i'm 16 connections... not like hundreds.
[16:58] <smoser> just ro r-start it, but your'e right i dont really.
[16:58] <nacc> ah ok
[16:59] <nacc> yeah, to continue, but if you reuse the directory, it should just pick up from where it stopped (which would be where it failed to download before)
[16:59] <smoser> this is true.
[16:59] <nacc> that assumes it works :) which it does in my testing
[16:59] <smoser> i am currently intrigued by the idea of pristine-tarball
[16:59] <nacc> yeah, i think that's the direction we need to look at next
[17:00] <smoser> i belive it can be added
[17:00] <smoser> and given a tree with all the .dsc and .orig.tarball would be really easy.
[17:00] <smoser> and even added in a one by one time.
[17:01] <smoser> if we dont worry about a merge commit between upstream/ branches and ubuntu branches (or debian branches)
[17:01] <nacc> right
[17:01] <smoser> then i think we can just insert the upstream/tag files whenever
[17:02] <smoser> and even re-order that branch and force-overwrite
[17:02] <nacc> right, because nothing really depends on it, beyond for building purposes (afaict)
[17:02] <nacc> smoser: i wonder if we can reuse gbp for this
[17:02] <smoser> ie, no one should be depending on the state of that branch other than that it can recreate tarballs.  i think generally it would be nice to have the correct order if you logged it
[17:02] <nacc> smoser: via import-orig or import-dsc maybe
[17:02] <smoser> yeah.
[17:03] <nacc> smoser: i will spend some time researching that today
[17:04] <smoser> it says 'The sources are placed on  the  upstream  branch  (default:  upstream) tagged and merged onto the debian branch (default: master).'
[17:04] <smoser> we wan the first 2 pieces of that import-orig but not the third
[17:05] <nacc> yeah, there's also a --pristine-tar falg
[17:05] <nacc> *flag
[17:05] <nacc> "If you enable pristine-tar, delta files are committed to a pristine-tar branch if you call git-import-dsc or git-import-orig. When you build the package using gbp buildpackage, the exact tarball is regenerated using pristine-tar. "
[17:05] <nacc> from https://wiki.debian.org/PackagingWithGit
[17:05] <smoser> right. and clearly we want that :)
[17:05] <nacc> yep :)
[17:05] <nacc> i'll play with the options to see if we can get this working
[17:10] <smoser> nacc, this one is awesome...
[17:11] <smoser> http://paste.ubuntu.com/23393687/
[17:12] <smoser> i'm pretty sure... unless i'm just being dumb that there are 2 different orig tarballs of the same upstream version there.
[17:12] <smoser> and so downloader is saying that is imposible
[17:13] <nacc> of differnt sizes
[17:13] <smoser> yeah.
[17:13] <nacc> does the changelog indicate if they broke or changed something?
[17:14] <smoser> idont know. didnt really look
[17:15] <nacc> smoser: the current changelog indicates 0.3.1-1 does not exist :)
[17:15] <nacc> smoser: so probably 0.3.1-1 was a mistaken publish
[17:18] <jgrimm> smoser, how's that cloud-init SRU coming along?
[17:24] <smoser> :)
[17:24] <smoser> working on it.
[18:01] <jgrimm> :)
[19:31] <slangasek> doko: haskell build verbosity> nooooo ideaaaa
[22:11] <nacc> smoser: hrm, git-buildpackage does not like xgitified repositories
[22:18] <nacc> barry: feel free to tell me this is just a bad idea, but i'd like to use git-buildpackage's python code to basically do what `gbp import-orig` does; however gbp is python2 and the importer is python3. Is my only option to call gbp directly (which doesn't work with our repository format, apparently)
[22:20] <nacc> smoser: actually, to be specific, gbp doesn't like not having a .git unless it's a bare repository, rather than using an alternative GIT_DIR
[22:21] <barry> nacc: first, it depends on whether the gbp code is even python3 compatible, which i don't know.  if it is (either by design or accidentally), then it's mostly a matter of getting it installed on a path that python3 can find it.  i'd try a virtualenv for that just to see if you can make it work
[22:22] <nacc> barry: ack, thanks!
[22:22] <barry> nacc: ack! :)
[22:29] <nacc> https://github.com/agx/git-buildpackage/pull/16 looks to be the same issue i'm running into
[22:46] <eu__> Hi all. I have installed Ubuntu Mate 16.10, AppArmor 2.10.95-4ubuntu5.1, QEMU emulator version 2.6.1, libvirtd (libvirt) 2.1.0. I'm create shared directory with virt-manager and successfully mount it inside guest. But i cannot nor write nor `ls' inside it. How to fix it?
[22:47] <nacc> eu__: you probably want #ubuntu
[22:47] <eu__> @nacc Sorry. Thank you.
[22:47] <udevbot> Error: "nacc" is not a valid command.
[22:47] <nacc> smoser: is there a reason you didn't convert all of the subprocess.run to run?
[22:49] <smoser> nacc, i just missed it
[22:49] <smoser> i noticed that today too
[22:49] <nacc> smoser: ack, not a big deal, just wnated to check
[22:49] <nacc> smoser: i'm working on switching us away from xgit-ified dependencies, and see if we can just use gbp as-is
[22:54] <smoser> nacc, 558 total. 16 working. 524 pass. 18 fail.
[22:54] <smoser> golang-golang-x-net-dev                  Unknown: AssertionError: source pkg version: 1:0.0+git20150817.66f0418-1 != changelog version: 0.0+git20150817.66f0418-1
[22:54] <smoser> that one is interesting, then tehre is the gexiv2 error (different orig tarballs)
[22:54] <smoser> and the rest are download errors
[22:55] <smoser> http://paste.ubuntu.com/23395022/
[22:56] <smoser> http://paste.ubuntu.com/23395027/ <-- i s gnutls28
[22:56] <smoser> and /me goes
[22:57] <smoser> wow
[22:57] <smoser> $ du -hs firefox/git/
[22:57] <smoser> 1.2G    firefox/git/
[22:57] <smoser> and its not done
[22:59] <slangasek> heh, golang-golang-x-net-dev has both 0.0+git20150817.66f0418-1 and 1:0.0+git20150817.66f0418-1 in its changelog
[23:17] <nacc> urgh, and the 1: version's .dsc file says it's not an epoch bump, even though changelog does
[23:17] <nacc> https://launchpad.net/debian/+source/golang-golang-x-net-dev/1:0.0+git20150817.66f0418-1
[23:54] <rcj> cyphermox, When you have a chance, the livecd-rootfs change for cloud images is ready for review @ https://code.launchpad.net/~rcj/livecd-rootfs/zesty_grub/+merge/309517