[05:47] <didrocks> good morning
[06:37] <pitti> Good morning
[06:38] <didrocks> guten morgen pitti! How are you?
[06:42] <pitti> didrocks: un peu fatigue, mais bien, merci! et toi?
[06:43] <didrocks> pitti: bien bien! :-)
[06:43] <pitti> didrocks: quand commence-tu tes vacances?
[06:44] <didrocks> pitti: vendredi soir, et toi?
[06:44] <pitti> à demain (looks up Thursday) jeudi
[06:44] <didrocks> oh super! :)
[06:45] <didrocks> j'espère que tu en profiteras bien :)
[06:45] <pitti> "profiteras"?
[06:46] <didrocks> "enjoy"
[06:46] <didrocks> (in future)
[06:46] <didrocks> from "profiter"
[06:46] <pitti> ah
[06:46] <didrocks> it can have a bad connotation sometimes though
[06:47] <didrocks> like:
[06:47] <didrocks> you can "profite" from someone
[06:47] <pitti> didrocks: looking forward to some quiet days next week to get to some of my research projects for mocking hardware, etc.
[06:47] <pitti> as I guess many folks will already be on holiday
[06:47] <didrocks> but apart from that case, it's a possitive one :)
[06:47] <didrocks> oh?
[06:47] <didrocks> so you will continue touching computers? :)
[06:50] <pitti> didrocks: I meant "next Thursday"
[06:50] <pitti> "demain jeudi" != "next Thu"?
[06:50] <didrocks> demain jeudi == Thursday as of "tomorrow", so from the 12th
[06:51] <pitti> oh
[06:51] <didrocks> you wanted to tell the one after?
[06:51] <pitti> right
[06:52] <didrocks> ok, it's "jeudi en huit (8)", but only old people (and I) seems to tell that ;)
[06:52] <pitti> oh, interesting -- old people in Germany say that, too
[06:52] <pitti> "Donnerstag in 8 Tagen"
[06:52] <didrocks> heh, most of the time, when I'm telling that, I need to explain what I meant ;)
[06:52] <pitti> but most folks just say "this Thu" vs. "next Thu"
[06:52] <didrocks> I find that useful though
[06:53] <pitti> yeah, it's a really useful phrase -- much more precise
[06:53] <didrocks> ok, we use the exact date to remove confusion
[06:53] <pitti> je commence mes vacances à la 20e
[06:54] <didrocks> héhé, je commence mes vacances bien avant donc, vendredi 14 au soir
[06:54] <didrocks> oh btw, is that a frenchism?
[06:54] <didrocks> like, we always tell that we start the holidays by the evening of the day you are leaving for holidays
[06:54] <didrocks> and getting back by the morning you are back to work
[06:54]  * achiang specifies all dates in unix epoch to avoid above problems... "can we meet for dinner at 1355342049?"
[06:54] <didrocks> achiang: ahah, I'm sure you feel alone most of the time you try to set that up! :)
[06:55] <achiang> didrocks: ah, but the person who comes to dinner, we will have a GREAT conversation! :)
[06:55] <didrocks> I remember that using this kind of "evening/morning" confused my japanese teacher, they just use like "from saturday to sunday"
[06:55] <didrocks> achiang: indeed, that's a kind of natural selection :-)
[06:56] <pitti> didrocks: at least we don't say that in Germany; we say the first and the last full holiday
[06:56] <didrocks> you don't enjoy the full time of your holidays! we do :-)
[06:56] <pitti> didrocks: try that if you start working at 5 :)
[06:57] <didrocks> fair enough ;)
[06:57] <didrocks> oh oh oh my launchpadlib, why do you keep corrupting the cache?
[06:58] <didrocks> (because it's multiprocess, I know, and you don't support that…)
[07:01] <pitti> you just don't hug it enough
[07:04] <didrocks> seems so :)
[07:04] <didrocks> it strikes back: https://jenkins.qa.ubuntu.com/view/cu2d/ :)
[07:51] <bryce> didrocks, oh how familiar I am with that bug.
[07:51] <didrocks> bryce: I fully trust you on that one :)
[07:51] <bryce> didrocks, I used to hit that all the time with arsenal.
[07:52] <didrocks> yeah, I wonder how easy/hard, it will be to directly fix it
[07:52] <bryce> finally just wrote a tool to locate the bad cache file and delete it.  :-P
[07:52] <didrocks> like writing in .tmp
[07:52] <didrocks> and move it
[07:53] <bryce> hmm, I never could get the LP folks interested, but I dug into it a bit myself and it looked like maybe you could wrap the call that's trying to read from cache in a try, and on except just pretend the cache didn't exist and requery from the network
[07:55] <bryce> didrocks, https://bugs.launchpad.net/launchpad/+bug/545401
[07:55] <ubot2`> Launchpad bug 545401 in Launchpad itself "httplib.IncompleteRead: IncompleteRead received from launchpad" [High,Triaged]
[07:56] <didrocks> bryce: yeah, and redownload that in the cache
[08:00] <bryce> # Clear broken cache files from launchpadlib cache hourly
[08:00] <bryce>  01  * * * * cd $LPLTK_CACHE && json_check $LPLTK_CACHE | xargs --no-run-if-empty rm -v
[08:00] <bryce> didrocks, https://lists.launchpad.net/arsenal-user/msg00008.html
[08:02] <didrocks> bryce: thanks! I think I'll rather use one cache per job
[08:03] <bryce> didrocks, by that you mean discard the cache after every run?
[08:03] <didrocks> bryce: no, I just set a dir by different process
[08:03] <didrocks> as my jobs are named differently, I can differentiate on that
[08:03] <didrocks> and having one cache dir for each
[08:16] <bryce> didrocks, was that a recommendation by the LP folks?
[08:16] <didrocks> bryce: not sure, but it works for most of my processes
[08:17] <didrocks> bryce: cleaning the cache every hours won't, as it will fail the whole process and I need to wait for the next day
[08:17] <didrocks> or relaunch manually
[08:20] <bryce> didrocks, hmm, interesting.
[08:20] <bryce> ok, vacation time for me, cya all in a week.
[09:01] <seb128> hey desktopers
[09:01] <didrocks> salut seb128!
[09:01] <seb128> lut didrocks
[09:02] <seb128> brrrr, it's cold today
[09:02] <seb128> -6°C this morning
[09:02] <didrocks> -6 here
[09:02] <didrocks> oh! :)
[09:02] <seb128> ;-)
[09:02] <didrocks> feeling the pain then :)
[09:05] <Laney> hey
[09:05] <Laney> only -3 "Freezing Fog" here ;-)
[09:05] <didrocks> Laney: some sun here! So I guess it's relatively better :-)
[09:07] <Laney> heh
[12:48] <solancer> anyone who can fix this PPA https://launchpad.net/~ikarosdev/+archive/unity-revamped
[13:47] <solancer> yo
[13:47] <solancer> anyone here into C++
[13:48] <didrocks> solancer: you already asked here and on #ubuntu-unity, Mirv answered you
[13:51]  * rvr_ wonders why the save dialog has "Recently Used" as the default 'folder', if nothing can be created there
[14:07] <didrocks> pitti: can you please bump the priority for https://launchpad.net/~ubuntu-unity/+archive/daily-build/+build/4057051? It started 14 hours ago, got down to 8 minutes and now 39 minutes. Then, we have to wait for 3 publisher round to get it in main (it's a source package rename) to not break anything during the night
[14:08] <pitti> didrocks: done
[14:08] <didrocks> thanks :)
[14:12] <jcastro_> mdeslaur_: I think that reddit post about that empathy picture is actually a gtalk service feature.
[14:13] <jcastro_> mdeslaur_: on their windows client if you update the picture it transparently just updates it across the service.
[14:13] <jcastro_> I think empathy is just replicating the standard expected behavior there.
[14:13] <jcastro_> (I haven't used the window client in a long time, so my information might be out of date)
[14:16] <mdeslaur_> jcastro_: he's saying empathy automatically picks up your user icon, but I can't see it doing that
[14:17] <mdeslaur_> jcastro_: unlike pidgin, which does in fact use your ~/.face icon
[14:17] <jcastro_> ah
[14:17] <jcastro_> is .face the one we set in the installer?
[14:17] <mdeslaur_> yes
[14:38] <Sweetsha1k> seb128: http://skyfromme.wordpress.com/2012/12/12/libreoffice-test-marathon-bibisect-4-0-and-ubuntu-packages/ <- the long version of the status update yesterday.
[14:39] <seb128> Sweetsha1k, k
[15:18] <micahg> Laney: any idea what happened to webkit on powerpc?
[15:18] <Laney> timed out, retried
[15:19] <Laney> hoping ross is better than adare ...
[15:19] <micahg> hrm, I'm tempted to leave -gstabs in for this upload then an just push it (I had trouble with timeouts without -gstabs on oneiric as well, didn't actually test my theory yet though)
[15:20] <micahg> I wouldn't mind iterating so much if it wasn't a 12(powerpc)-24(arm*) hour wait :)
[15:20] <micahg> thanks
[15:22] <czajkowski> micahg: need something done sooner ?
[15:22] <Laney> it's just a long build
[15:22] <czajkowski> ah ok
[15:22] <Laney> ppc is at a link step atm
[15:23] <Laney> this is probably where it will time out if that's going to happen
[15:52] <Laney> not looking good ...
[16:43] <seb128> pitti, still working?
[16:44] <Laney> ooh it moved on
[16:47] <plars> is https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1080437 likely related to the recent move to  compiz?
[16:47] <ubot2`> Launchpad bug 1080437 in ubiquity (Ubuntu) "no background during the 13.04 daily install" [Undecided,Confirmed]
[16:48] <didrocks> xnox: ^
[16:48] <didrocks> plars: xnox is the ubiquity upstream, he would know more :)
[16:48] <xnox> plars: no.
[16:49] <xnox> plars: but troubleshooting would be welcome, because e.g. standard background on raring install in kvm is grayscale.
[16:51] <xnox> plars: I am suspecting kernel, but I have no good ways to discuss it.
[16:52] <plars> xnox: it's greyscale?
[16:53] <plars> xnox: it doesn't look greyscale to me on hardware and on vbox - is this a kvm specific thing?
[16:53] <xnox> plars: installing raring in KVM VM, boot, login try changing the backgrounds - all of them are approx 8-bit grayscale with some distant recollection of intended pattern, note that the unity launcher is still full technicolor.
[16:53] <plars> xnox: ah, are you using the default cirrus driver?
[16:53] <xnox> plars: yes, it might be a separate issue, but the two issues together don't help me debug the background problem.
[16:54] <mvo> glatzor_: when you have a moment, could you check https://code.launchpad.net/~mvo/aptdaemon/send-signal-on-auth-fail please? it will allow a process like unity to know that the transaction failed to authenticate.
[16:54]  * xnox doesn't know if it's there or not there. It's all black....
[16:54] <plars> xnox: yes, that driver is known to have a lot of trouble... that's a whole other story.  there's an easy workaround though, boot kvm with -vga std
[16:55] <xnox> plars: ok, i will set that in virt-manager and see if I can isolate one thing from another.
[16:55] <xnox> plars: even if cirrus has problems, it's still a regression since both precise & quantal come up with color background cirrus or not cirrus.
[16:56] <plars> xnox: iirc, we had lots of trouble with it under quantal, I'll see if I can dig up the bug on that
[16:56] <xnox> plars: yeah a kernel driver was enabled and later disabled in quantal, i wonder if it got re-enabled for raring.
[16:57] <xnox> plars: maybe we should poke #-kernel about it =)
[16:58] <plars> xnox: looking at an upgrade issue right now, but will see if I can hunt down the older bug we had open on this in a bit and poke them.  Using -vga std should help in the meantime though
[16:59] <plars> xnox: I was just hoping that maybe it was yet another compiz related issue with the dm-ubiquity session
[16:59] <xnox> plars: it was present pre-compiz flip.
[16:59] <plars> not that I'm wishing for compiz problems - hopefully it didn't come out that way though
[17:00] <plars> ah, true
[17:07] <seb128> pitti, seems not, anyway the context was: did you see that before:
[17:08] <seb128> $ DISPLAY=:0 python -c "from gi.repository import GObject"
[17:08] <seb128> ...
[17:08] <seb128>     from ._gi import _API, Repository
[17:08] <seb128> ImportError: could not import gobject (error was: MemoryError())
[17:08] <seb128>  
[17:08] <seb128> pitti, it's happening on my nexus upgraded to raring, it could be a "not enough memory" issue since the device has no swap but I freed 160Mb, that ought to be enough to import gobject...
[17:29] <glatzor_> hello mvo
[17:30] <glatzor_> mvo, I don't know if the transaction should fail if the auth failed
[17:30] <glatzor_> mvo, this would allow to a denial of service attack
[17:31] <glatzor_> mvo, since you could block any transactions of any other user by just listening on the bus for new transaction and letting the run() method fail
[17:33] <glatzor_> mvo, oh wait the check forgein method would forbid this. the attacker could only block the transactions of the current user.
[17:46] <mterry> seb128, did I miss the meeting?
[17:47] <seb128> mterry, you did, and I missed that you were taking a vac day yesterday until I checked the admin website today :p
[17:47] <seb128> mterry, I wish we had team holidays integrated in a desktop calendar
[17:47] <mterry> seb128, oh!  haha
[17:47] <seb128> we would need to have a desktop calendar first though :p
[17:48] <mterry> seb128, I thought today was Tuesday still
[17:48] <seb128> lol
[17:48] <seb128> mterry, I think that day is over in all timezones
[17:48] <mterry> seb128, stupid question since it's Wed.  But yeah, sorry I didn't give a heads up
[17:48] <seb128> ;-)