[00:44] <xnox> barry: actual dictionary hash ordering build-failure: https://launchpad.net/~ubuntu-rebuilds/+archive/py3.3/+build/3929037
[04:03] <pitti> Good morning
[04:16] <pitti> cjwatson: congratulations and thanks for setting up the -proposed stuff! it's great to see this in action at last
[04:58] <pitti> apw: I subscribed you to bug 1070427, I asked a question about the kernel header provides/depends; that seems to break some cases ATM
[05:02] <ScottK> xnox: Is your python3 transition tracker updating?  I see uploads from yesterday that aren't reflected.
[05:43] <ScottK> xnox: How goes cmake?  pykde4 is now waiting on it.
[06:57] <dholbach> good morning
[07:05] <fishor> hallo devs, i'm porting gnome-sound-recorder to gstreamer-1.0 and gtk3. My repository based on gnome-media with removed all deprecated parts. you can find it here https://gitorious.org/gsr/gnome-recorder/commits/master
[07:05] <fishor> my question, do it make sense to target it for 13.04?
[08:03] <mvo> barry: I took the liberty to upload the fix for #846044 now that its part of the upstream git, just fyi
[10:08] <dagoaty> I am trying to write a preinst script for a package that supports the 'backup' functionality. I think I have this working but as far as I can tell only the installer supports the backup feature. Is there any commandline mode for dpkg which I can use to test my script behaves correctly?
[12:11] <infinity> @pilot in
[12:14]  * dholbach hugs infinity
[12:30] <doko> Sweetshark, when do you plan the first libreoffice upload?
[12:51] <Pjotr> Hello, I have a question about ufw.
[12:51] <Pjotr> Where can I find the default rules profile of ufw?
[13:05] <Pjotr> mdeslaur: can you maybe tell me, which default rules profile ufw uses?
[13:06] <mdeslaur> Pjotr: jdstrand ^
[13:06] <jdstrand> Pjotr: look in /etc/ufw/*rules. you can also do 'sudo ufw show raw'
[13:06] <jdstrand> (if it is enabled)
[13:11] <Pjotr> jdstrand: that doesn't show a lot of rules... Is the default profile secure enough for the average home user, or do you advise adding extra rules?
[13:11] <jdstrand> Pjotr: it is default deny. It is recommended for a home user, yes
[13:13] <Pjotr> jdstrand: would ufw with the default profile negatively affect samba or torrents?
[13:13] <jdstrand> Pjotr: the basic idea is that it is default deny for incoming. default allow with connection tracking for outgoing. it allows for things like dhcp and ping. it is a reasonable default but can be adjusted for your requirements
[13:14] <jdstrand> Pjotr: as a samba client, it shouldn't in recent releases. you would be able to fetch torrents, you'd have to open up a port to serve torrents
[13:15] <jdstrand> (you'd also have to open a port to serve clients)
[13:17] <Pjotr> jdstrand: thanks for the information.... :-)  For now, this answers my questions about ufw.
[13:18] <jdstrand> np
[13:19] <ttx> jml: been trying undistract-me, installed from PPA... but it does not seem to trigger. Only works once I do . /usr/share/undistract-me/long-running.bash && notify_when_long_running_commands_finish_install
[13:20] <jml> ttx: yeah. I've noticed that recently too
[13:20] <jml> ttx: I could swear it worked befoe
[13:20] <ttx> jml: I believe you ;)
[13:20] <ttx> only half as fun if it doesn't work though :P
[13:20] <jml> ttx: manually running '. /etc/profile.d/undistract-me.sh' works for me also
[13:21] <jml> ttx: I suspect it's due to some weird login shell interaction
[13:22] <ttx> jml: probably in the [ -z "$BASH_VERSION" -o -z "$PS1" -o -n "$last_command_started_cache" ] test
[13:23] <jml> perhaps
[13:23] <ttx> dmesg
[13:23] <ttx> oops
[13:26] <doko> jamespage, ping
[13:26] <jamespage> doko, pong
[13:32] <ttx> jml: can't see anything obvious, let me know if you find the solution :)
[13:32] <jml> ttx: will do.
[13:51] <doko> barry, xnox: in the py3.3 test rebuild are about ~200 packages, slangasek identified ~500 which might be affected. there's a small difference
[13:52] <xnox> doko: 500 _binary_ packages that slangasek identified.
[13:52] <doko> ok, so once boost is fixed, we should be ready to change the default?
[13:53] <xnox> doko: more or less. (boost and it's python3-like rdeps)
[13:53] <barry> python-oauthlib is pretty important, but i'll look at that (unless a fix is already uploaded but not in the ppa yet)
[13:54] <barry> it needn't block the change though
[13:54] <xnox> barry: testsuite failures for pure-python modules do not block us to switch default.
[13:54] <xnox> barry: as e.g. python-oauthlib is installable with python3.3 as default.
[13:54] <barry> has anyone looked at ubiquity?
[13:54] <xnox> doko: do you have python3-defaults arch:all -> arch:any done?
[13:54] <xnox> barry: local build is fine.
[13:55] <barry> it only fails on amd64
[13:55] <xnox> barry: ppa build "got killed" in the logs
[13:56] <barry> hung for 150m apparently
[13:56] <barry> in any case, doko i think we should do it
[13:58] <doko> xnox, yes, will remove the header symlink again
[13:59] <xnox> doko: ack.
[14:00] <doko> barry, xnox: ok, then I'll wait for the boost builds to succeed, and the make 3.3 the default (which might be tomorrow)
[14:15] <tsdgeos> is the "raring" part of the archive private on purpose?
[14:16] <micahg> tsdgeos: what do you mean?
[14:17] <tsdgeos> micahg: http://archive.ubuntu.com/ubuntu/dists/raring exists but gives 403
[14:17] <tsdgeos> that's what i mean
[14:17] <micahg> wfm
[14:17] <tsdgeos> does it?
[14:18] <micahg> yep
[14:18] <yofel> wfm too
[14:18] <tsdgeos> you guys in copenhagen?
[14:19] <micahg> not yet
[14:19] <tsdgeos> well, i'm blaming the network somehow
[14:19] <tsdgeos> does not work from here, but works in a server i have somehere else
[14:19]  * yofel isn't going
[14:19] <tsdgeos> and it works for you
[14:19] <tsdgeos> so obviously not a problem in the server
[14:25] <develop7> Hello all. Is there any way to figure out which process handles particular keyboard shortcut?
[14:27] <develop7> I'm going to fix https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1064345 myself, but have virtually no clue where to start from.
[14:55] <Sweetshark> doko: ASAP, which means once I get all 3.7 deps in place.
[15:04] <dagoaty> Is there any frontend to debconf I can use to test the 'backup' feature? I've only seen this working in the installer.
[15:11] <rbasak> pitti: is there an easy way to test an apport hook? It refuses to even --save against my test package because it is "not an official Ubuntu package"
[15:12] <rbasak> I haven't managed to find an option to do this anywhere, and there isn't any discussion of hook testing in https://wiki.ubuntu.com/Apport/DeveloperHowTo :-(
[15:22] <barry> xnox: i think it's still worth having a discussion on 3.3/3.4.  the actual decision will probably be a rubberstamp, but in the past such sessions have been very good at bringing up collateral issues to think about
[15:23] <xnox> barry: well currently it's scheduled in a strange place: clashes with building arm images, rapid archive bringup and a few other cool sessions for me =)
[15:24] <barry> xnox: complain to your track chair :)
[15:24] <barry> xnox: or make your participation essential and let the scheduler sort it out
[15:25]  * barry now goes to check his own schedule
[15:25] <xnox> barry: actually that does nothing. you need to as track chair to special mark somebody "essential" in the summit.
[15:26] <Laney> I don't know that it does absolutely nothing
[15:26] <Laney> but it's not a hard requirement like it appears
[15:26] <slangasek> xnox: right, are you sure it has *no* effect?
[15:27] <slangasek> all I know for sure is that there's an additional knob now
[15:28] <xnox> slangasek: can you please accept this one for uds-r sprint https://blueprints.launchpad.net/ubuntu/+spec/foundations-r-upload-intermediary
[15:28] <xnox> slangasek: there is a session, but summit doesn't "know" the blueprint link.
[15:29] <xnox> although it's kind of done ?!
[15:29] <slangasek> xnox: "accept" it?  That blueprint isn't marked as proposed for uds-r
[15:29] <slangasek> probably because cjwatson doesn't think we need to discuss it? :)
[15:29] <xnox> slangasek: I guess I got confused why it's linked from
[15:30] <xnox> http://summit.ubuntu.com/uds-r/meeting/21333/archive-maintenance-software-hacking/
[15:30] <slangasek> ahh, the url is there just as a reference
[15:31] <xnox> slangasek: this blueprint is still on the schedule http://summit.ubuntu.com/uds-r/meeting/21074/foundations-r-landscape-fde/
[15:31] <xnox> slangasek: but it was rejected from uds-r
[15:32]  * xnox twiddled the "subscriber vs essential" and I'll look how that will affect the scheduler next time it reruns ;-)
[15:36] <Laney> hmm, yeah, quite a number of conflicts
[15:36] <rbasak> pitti: I found my answer by digging in the code. I can set APPORT_DISABLE_DISTRO_CHECK in the environment
[15:37] <rbasak> SpamapS, jamespage, adam_g: FYI ^^
[15:37] <jamespage> rbasak, nice
[15:39] <cjwatson> dagoaty: pretty much any old debconf frontend, including the default of "dialog", supports backup if your script says "db_capb backup".  It is not at all installer-specific.
[16:00] <dagoaty> cjwatson: Odd, I only seem to get an OK prompt with dialog.
[16:02] <dagoaty> cjwatson: I even tried the example code at http://www.fifi.org/doc/debconf-doc/tutorial.html#AEN217 with no success
[16:04] <Snow-Man> what's w/ using sftp to upload to a ppa..?  docs indicate using: incoming = ~sfrost/<ppa-name>/ubuntu, but I'm getting an error that launchpad couldn't find a PPA named 'postgis' (even tho I see it on the website..)
[16:06] <rbasak> Snow-Man: is that an email error? There's a known bug that the email error reports for upload failures name a random PPA that isn't yours
[16:06] <rbasak> Snow-Man: (but in that case the error might still be valid - it happens to me if I use the wrong path)
[16:07] <Snow-Man> rbasak: well, yes, I see the random PPA name, but in the body of the email it's saying "Could not find a PPA named 'postgis' for 'sfrost'.", even though when I go to: https://launchpad.net/~sfrost/+archive/ppa it says "postgis" at the top
[16:08]  * Snow-Man randomly tries stuff
[16:12] <Snow-Man> well, that's cute
[16:12] <Snow-Man> Source/binary (i.e. mixed) uploads are not allowed.
[16:12] <Necrosporus> Why there is no GUI option to disable advertising banner in software center?
[16:12] <Snow-Man> missed *that* in the docs..
[16:20] <ppetraki> I've got an upstart task that simply won't start. http://pastebin.ubuntu.com/1307508/
[16:20] <ppetraki> all the parts of the script run fine, put it in a task, fails
[16:24] <slangasek> ppetraki: do you have a /var/log/upstart/opengrok-index.log?
[16:25] <ppetraki> slangasek, http://pastebin.ubuntu.com/1307525/
[16:25] <ppetraki> slangasek, but it all runs fine from the shell
[16:26] <slangasek> ppetraki: chances are there's something different in the upstart environment (which is mostly a blank slate) vs. the shell environment that makes the difference
[16:27] <ppetraki> slangasek, so im sourcing a bash script, and then calling the funcs, is that supported?
[16:27] <slangasek> ppetraki: oh, for one thing when you're sourcing that script at the commandline it's sourced by bash, and when you're running an upstart job the shell is /bin/sh == ash
[16:27] <ppetraki> that's got to be it
[16:27] <ppetraki> slangasek, http://pastebin.ubuntu.com/1307536/ confirmed
[16:28] <slangasek> ppetraki: yeah, so I think you want to put your script into a separate file with an explicit #!/bin/bash, and then put 'exec myscript' in the upstart job
[16:28] <ppetraki> slangasek, got it, thanks
[17:14] <slangasek> Laney: you say you were seeing conflicts for sessions you were marked as 'participation essential' on?  Can you give me an example?
[17:15] <slangasek> Laney: the scheduler seems to be doing an adequate job of warning me if p.e. subscribers have conflicts
[17:17] <Laney> slangasek: I don't think I have conflicts like that, just ones where I subscribed non-essential
[17:17] <Laney> should I expect doing that to influence the scheduler at all?
[17:17] <slangasek> Laney: ok - expected behavior :)
[17:17] <Laney> perhaps I'll have to upgrade some then
[17:18] <slangasek> yes, that's what the flag is there for! :-)
[17:18] <Laney> slangasek: also, how come https://blueprints.launchpad.net/ubuntu/+spec/foundations-r-future-release-brainstorming got changed to declined?
[17:19] <slangasek> Laney: we believe it's covered by the combination of the release-schedule session and the quantal-feedback session
[17:20] <slangasek> http://summit.ubuntu.com/uds-r/meeting/21405/foundations-r-schedule/ + http://summit.ubuntu.com/uds-r/meeting/21047/foundations-r-prior-release-feedback/
[17:21] <Laney> well, not if the latter is entirely backwards looking
[17:22] <slangasek> Laney: having an entirely backwards-looking session that doesn't result in actions for the future would not be a productive use of time :)
[17:22] <slangasek> are there specific things you think are not covered in the other two sessions, that would warrant a third?
[17:23] <Laney> I'll just distribute the topics from the whiteboard
[18:20] <Snow-Man>  c
[18:20] <Snow-Man> erp
[22:28] <jelmer>  /query barry
[22:29] <jelmer> whoops
[22:29] <jelmer> hi barry :)
[23:36] <xnox> fgrep changed?! http://paste.ubuntu.com/1308437/