=== eeejay is now known as eeejay_is_afk [03:32] Bah! What has suddenly caused Do to no longer be able to register +space as a keybinding? UNITYYYYYYY! [03:33] RAOF: register or use ? [03:33] Register, it seems. [03:33] also, screaming on #ubuntu-desktop doesn't make your bug get fixed any faster ;-) [03:33] No, but it *does* make me feel better :) [03:34] RAOF: someone is using super-space then, so its not a unity thing as unity doesn't register keybindings with gnome [03:35] Hm. [03:39] Hah. Of course, the other option is that Do's broken :/ [03:39] did Do ever work ? [03:39] also, synapse ? [03:39] smspillaz: i've found screaming on #ubuntu-desktop to be an effective way of getting bugs fixed :) [03:39] desrt: sadly, it seems to be the case [03:40] desrt: just /nick ${some_important_person} STACKING!!!!1111one!!1one!1!(shift+1) and my adrenaline levels go through the roof [03:40] :) [03:41] :) [03:41] :( [03:41] * RAOF screams STACKIIIIIING! at the moon. [03:41] Bottoms. [03:41] you know what [03:42] smspillaz: You're familiar with the abomination that is cmake - how do I get it to #define a variable at build time? [03:42] I'm going to go an entire day at UDS where all I say is either "STACKING" "REPARENTING" "CRASHING" [03:43] RAOF: 1) it is not an abomination 2) configure_file (config.h.in config.h) 3) in that config.h.in #cmakedefine SOME_THING ${CMAKE_VAR} [03:43] You'll probably need a larger vocabulary than three words. Might I suggest that you add CORRUPTION, DUAL-HEAD, and GL_MAX_TEXTURE_SIZE? :) [03:43] no, those are your three words [03:44] :) [03:44] RAOF: http://www.cmake.org/cmake/help/cmake-2-8-docs.html [03:44] also, I am trying to use QtQuick [03:44] it is really hard to use for some reason [03:45] I wrote the same thing in GObject/GTK+ in less than an hour [03:45] Hm, I think that maybe -D is the invocation I'm after. [03:45] ohh right [03:45] RAOF: add_definition [03:45] http://www.cmake.org/cmake/help/cmake-2-8-docs.html#command:add_definitions [03:45] add_definitions (-DFOO -DBAR) [03:46] it does it at a source level for the current CMAKE_CURRENT_SOURCE_DIR though, so make sure that you don't have files in the same source dir with different #define requirements [03:47] desrt: oh, I was going to ask, is there a release of glib out with the key maximum size bumped ? [03:47] err key name maximum size ? [03:47] smspillaz: yes. should be the one in oneiric [03:47] cool [03:47] I can remove the truncation in the gsettings backend [03:48] patch was included in 2.29.12 [03:48] so 2.30.0 should definitely have it [03:48] this next two weeks is like "distro week", the only purpose is to do stuff for didrocks :) [03:48] eg, upstreaming distro patches etc [03:48] since apparantly I finished my SRU lists really early [03:48] desrt: awesome :) === eeejay_is_afk is now known as eeejay === eeejay is now known as eeejay_is_afk [05:18] Good morning [05:19] Morning pitti. [05:26] Hi pitti, TheMuso! Great job on Oneiric! [05:27] Nafai: Thanks, already starting to plan for Precise. [05:32] Morning. [05:36] hey Nafai, thanks === Wellark_ is now known as Wellark [05:36] yeah, got my first two uploads in :) [05:37] good morning [05:37] bonjour didrocks [05:37] guten morgen pitti, how are you? [05:37] pretty well, thanks! how about you? [05:37] feeling precise today [05:39] pitti: heh, I'm fine thanks! No more oneiric stories? ;-) [05:39] there's still SRUs :) [05:40] of course :-) speaking of which, once I can get dx adding test case to all their bugs, probably a nux and unity SRUs for today! [05:40] I finally have the crach fix and reverted the UI change [05:50] didrocks: FWIW, if some of the fixes are hard to reproduce, a regression test suffices [05:51] didrocks: i. e. "ensure that maximizing firefox window still works as expected" or something like that [05:51] didrocks: with yesterday's compiz I noticed a weird regression wrt. resizing [05:51] but haven't investigated it more closely yet [05:51] pitti: yeah, that's the advice I gave them already. I'm just chasing them to at least having them filling the easy to reproduce one [05:51] pitti: indeed, there is one with some maximized app [05:51] sam fixed it [05:51] I got stuck with an orange resize frame on the left of the screen that used to be my firefox window, with no way to recover [05:52] not sure it worth reset the counter though [05:52] oh? [05:52] not that one [05:52] no way to recover? if you take another window and try to snap at the same place? [05:52] ah, I didn't try that [05:53] I tried to click on the firefox icon in the launcher, press alt+f10 etc. [05:53] * pitti plays around with this and sees whether he can reproduce [05:53] I think you should try to resnap, that's what fixed it for me 3 months ago when we had glitches there [05:54] but there was a compiz proposed upload yesterday, perhaps that changed things [05:54] not for that one I guess [05:54] some casual tests work fine [05:54] so perhaps just bad luck and it's not a regression [05:54] yeah, that's more than possible [05:55] it's been a long time I didn't get glitches with snap though [05:56] pitti: sorry, restarting unity crashed my gnome-terminal and so weechat [05:57] I seldomly use it [05:57] I just needed to unmaximize yesterday because of firefox' bug of not scrolling down on text marking when maximized [06:00] pitti: would be interesting to know if you can get that reliably (and if it's only with firefox or not) [06:01] as some apps have weird behavior, like reasking to set their weight/height after a resize [06:01] right, I tried, but I was unable to reproduce it now [06:01] (that's what prevented before the terminal to get well snapped on the right/left) [06:01] doesn't seem scary though :) [06:08] didrocks: we can now drop all the settings distro patches \o/ :) [06:08] didrocks: I've got compizconfig so that it reads /usr/share/compizconfig/defaults/profile.defaults when looking for default values [06:08] smspillaz: great! we'll see that at the pre-UDS :) [06:09] next item is the settings transitions. I'm going to see if I can make that into a generic module, so you just specify the key names that have changed and what they are supposed to map to [06:10] pitti: you can reproduce it by going into expo mode, dragging a window to a workspace and then double clicking that workspace [06:11] pitti: I've fixed it upstream, but yeah, we'll have it in the next SRU [06:11] smspillaz: hm, that seems to work fine here [06:12] it was something along those lines [06:13] didrocks: (we'll have something like this: http://paste.ubuntu.com/707758/_ [06:13] compiz in precise is going to be awesome [06:14] precisely zero distro patches in the goal for me [06:17] great :) [06:48] smspillaz: is https://bugs.launchpad.net/ubuntu/+source/fglrx-installer/+bug/763005 in the latest compiz tarball? [06:48] Launchpad bug 763005 in compiz "Compiz's "Sync to Vblank" makes display stutter/slow with some drivers (like fglrx)" [Undecided,In progress] [06:48] smspillaz: well, the patch isn't [06:48] am I right? [06:57] didrocks: no, it isn't [06:58] hmm... right-click on unity icons is behaving a little weird. one click to open the menu and one more to close it... clicking on a different icon should open the menu of that icon [06:58] right? [07:00] smspillaz: and bug #864963 ? [07:00] Launchpad bug 864963 in unity "Titlebars for semi-maximized windows appear off the workspace" [Low,Fix committed] https://launchpad.net/bugs/864963 [07:03] doesn't look like it, no (just checked all the patches) [07:03] do you want new patches again or do you want to wait till sru1 ? [07:03] smspillaz: will be SRU1, trying to close all your bugs and ensure the right components are subscribed [07:03] smspillaz: can you take care of that for next time? [07:03] smspillaz: they were still a lot of downstream unity task [07:03] I can move all the stuff that we didn't ship to SRU1 [07:04] right .. my understanding was that we were meant to leave the unity tasks there ? [07:04] or do we need to mark the downstream task invalid and leave the upstream task there ? [07:04] smspillaz: not the downstream one [07:04] ok [07:04] smspillaz: the dowstream is there because it was opened [07:04] smspillaz: so rather than adding a compiz downstream task [07:04] take the unity one [07:04] and rename it [07:04] ok [07:04] my think script knows what to do :) [07:04] sync* [07:05] smspillaz: ok, I have a reproduceable bug where a window can disappear, let me restart to check [07:05] didrocks: I wonder if its possible to make "team" milestones, hijacking the unity milestones for compiz bugs doesn't seem right [07:05] didrocks: ugh, just can't win with these bugs [07:05] smspillaz: no, other we wouldn't have those unity milestones [07:05] otherwise* [07:06] but yeah, that's what we need, team milestones… [07:06] right, but they are milestones for the unity proejct not the unity team [07:06] smspillaz: yeah, we need to create another meta-project IMHO [07:06] dx-whatever [07:06] ok [07:06] but dx people have to be rigorous then [07:06] and remove/invalid all invalid tasks then [07:07] (like a compiz bug opened against unity -> should be set to invalid) [07:09] smspillaz: can definitively reproduce, can you test that: [07:09] 1. open two terminal, one on ws 1, one on ws 2 [07:09] 2. minimize one [07:09] 3. click on the launcher -> no terminal visible [07:10] * didrocks downgrades to unity 4.22 [07:10] uhh, not right now. I'm on upstream compiz and cant run unity [07:11] smspillaz: same with it [07:11] so not due to latest unity [07:11] maybe compiz? [07:11] it's probably been there for a while. let me give it a try on my netbook [07:11] smspillaz: yes please :) [07:11] smspillaz: try with the archive version firsrt [07:11] smspillaz: and then, the -proposed version [07:12] didrocks: things like this are very likely a unity thing as that's where the minimization hack happens (why oh WHY did design make me do that) [07:12] this one means: if you have more than one window opened and one of them is maximized, you can't see it anymore [07:12] (in scale) [07:13] worse, if it has focus, you can't bring it back [07:13] minimized or maximized ? [07:13] minimized* [07:13] I give up with this minimization hack [07:13] I really want to go back to having no previews for minimized windows [07:14] didrocks: oh, eh [07:15] didrocks: that's been around for ages === jincreator is now known as jincreator_away === jincreator1 is now known as jincreator [07:15] smspillaz: really? and nobody targeted it as something to fix? [07:15] it seems a major issue IMHO [07:15] * didrocks finds that there is too much things dx passes under cover, until distro finds the bugs… [07:15] didrocks: guess nobody has noticed it [07:16] how do you know it's been around for ages then? [07:16] but I know that's probably been a problem for a while, considering how it works [07:16] smspillaz: any workaround? it's pretty serious, and it was working on natty (even if it was a different behavior) [07:16] I can try fixing it today [07:17] smspillaz: great! I will delay the unity upload on that I guess [07:17] worrying that we let that pass finale :/ [07:17] didrocks: well, you didn't see it till just now ;-) [07:17] smspillaz: because I never minimize windows [07:17] smspillaz: users usually do [07:17] * smspillaz never saw any bugs about that either [07:18] smspillaz: are you looking unity bugs? [07:18] yes [07:19] smspillaz: isn't it linked to https://bugs.launchpad.net/unity/+bug/868185 ? [07:19] Launchpad bug 868185 in unity "cannot unminimize windows" [Critical,Fix committed] [07:19] didrocks: no, that was something different [07:19] I broke something for a day and then fixed it :) [07:19] anyways, try to understand that the minimized window hack is tricky :) [07:20] smspillaz: bug #682769 [07:20] Launchpad bug 682769 in unity "the launcher doesn't focus minimized dialogs on empty workspaces" [Medium,Triaged] https://launchpad.net/bugs/682769 [07:20] as it's linked [07:20] because its not really supposed to work :p [07:20] didrocks: nope, different bug [07:20] didrocks: and that's fixed :) [07:21] didrocks: by fixing that it caused a bunch of other bugs, again, this stuff is obscenely difficult to get right ;-) [07:22] bug #863114 ? [07:22] Launchpad bug 863114 in unity "Cannot raise window from panel after minimize" [Undecided,New] https://launchpad.net/bugs/863114 [07:22] smspillaz: ^ [07:22] smspillaz: this is the case if you don't have any other app on the same workspace [07:23] so yeah, we got some warning about minimization issue in unity bugs, despite you looking at them [07:23] that bug is still really unclear, but whatever, I'll fix it [07:23] keep in mind that I can't see and fix everything ;-) [07:24] smspillaz: this is the bug, I'm rephrasing it [07:24] "After minimize a window, i can't re-raise it by clicking on its icon on the panel. I have to use super+w or click on another icon before... [07:24] For example, if I minimize firefox, first i have to click on nautilus icon and next i'm able to raise the firefox window by clicking on its icon." [07:24] smspillaz: yeah, but don't claim nobody reported it when people are reporting it :) [07:24] doesn't say anything about switching workspaces [07:24] anyways, I'm going to stop arguing about this and just fix it [07:24] smspillaz: I'm adding the two test cases now [07:27] smspillaz: see bug #863114, is it more clear now? [07:28] didrocks: Error: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/863114) [07:29] I'll file a separate bug for the other thing [07:32] smspillaz: the other? [07:32] smspillaz: I put both test cases [07:32] as they are linked [07:33] ok [07:36] hum, nobody in dx subscribed the SRU team on any of the 20 bugs… :/ [07:36] * didrocks headdesk [07:36] more than 2 hours of just fixing bug tasks and papework… and counting… [07:38] hi pitti didrocks ... Oneiric is out, what's the word on the street? [07:38] rickspencer3: heise.de had a nice review yesterday [07:39] it's been working great on both of my computers, fwiw [07:39] rickspencer3: hey, people seems to like the new version. A lot of positive comments on the french forum contrary to 6 months ago :) [07:39] behaving here :) [07:39] \o/ [07:39] working well here - except on my eee1001 where the installer hangs [07:39] (on keyboard selection) [07:40] Oneiric comes with the didrocks and pitti seal of approval [07:40] didrocks: dx should use my autobug-script :) [07:40] pitti: FYI, I finished the French CD and it's going to be pressed on 10 000 CDs right now [07:40] didrocks: lp-im-working-on bugno [07:40] didrocks: wohoo! [07:40] smspillaz: it doesn't check that the right downstream components are subscribed though [07:40] smspillaz: can you fix that? 45 minutes to fix everything just for compiz there :/ [07:40] and it subscribes sru, assigns the right person, subscribes the right downstream desk [07:40] *task [07:40] (only wrote it yesterday though) [07:40] RAOF, bryceh: it's sooo true! http://xkcd.com/963/ [07:41] didrocks: um, sure [07:41] * didrocks is bored of doing dx paperwork because they don't care, not real interesting engineer work [07:41] didrocks: I think the solution to that is to find a way so that nobody has to do the paperwork ;-) [07:41] pitti: just a small issue, the CD check is failing [07:41] eg, get some clever launchpad scripts in the hands of DX [07:41] didrocks: got a bug report about that yesterday, yes [07:41] didrocks: bug 873401 [07:41] Launchpad bug 873401 in ubuntu-defaults-builder "Check disc for defects failed with a iso images built using ubuntu-defaults-image" [Undecided,New] https://launchpad.net/bugs/873401 [07:42] didrocks: also, I can't reproduce your first "bug" [07:42] smspillaz: well, when someone subscribe the wrong component and don't attach any branch to it (and it happens on compiz as well), how the script will know what to target? [07:42] smspillaz: are you on an empty workspace? [07:42] yes [07:42] with oneiric up to date with compiz proposed ? [07:42] I just switched to trunk unity now [07:42] smspillaz: I'm on 4.0 [07:42] ok, I'll try 4.0 [07:42] smspillaz: and it happens with older version [07:42] didrocks: I can reproduce the second one though, and have fixed it locally [07:43] pitti: can you try reproducing bug #863114, please? [07:43] Launchpad bug 863114 in unity "Cannot raise window from panel after minimize" [High,Triaged] https://launchpad.net/bugs/863114 [07:43] didrocks: out of curiousity, does the terminal window stay" active after its minimized ? [07:43] or does the focus go to the desktop [07:43] smspillaz: no, the desktop is focused [07:43] and you can't unminimize the terminal [07:43] right [07:44] clicking on the launcher icon do nothing [07:44] ok, let me give 4.0 a try then [07:45] smspillaz: your script doesn't seem to work well btw: bug #868930 [07:45] Launchpad bug 868930 in unity "Automaximization happens on unminimize" [High,Fix committed] https://launchpad.net/bugs/868930 [07:45] no team subscribed [07:45] didrocks: I suppose I need to do that with not having any previous terminals open? [07:45] didrocks: probably didn't run it on that bug ;-) [07:45] pitti: no, you can even if you have other terminals opened there [07:46] smspillaz: if you have to specify on every bugs, there is something wrong :) [07:46] didrocks: ah, can reproduce -- nothing at all happens if I click on the button [07:46] I suppose I can make it run on milestones === eeejay_is_afk is now known as eeejay [07:46] smspillaz: ^^ [07:46] no need to ping me :) [07:47] smspillaz: you are quite unlucky to be ever *never* able to reproduce bugs :) [07:47] I'm using trunk unity [07:47] give me the 20 minutes it takes to rebuild 4.0 please :) [07:56] smspillaz: why did you set the downstream task on bug #865051 "Fix Released" ? [07:56] Launchpad bug 865051 in unity "unity launcher loses track of deja-dup windows" [Medium,Fix committed] https://launchpad.net/bugs/865051 [08:03] didrocks: was running through my bug lists and thought that was was released. it wasn't, sorry about that [08:03] smspillaz: ok, fixing then [08:04] smspillaz: so, to sum up, SRU bug list cleaned, just waiting for you to reproduce the 1. case [08:04] awesome, currently rebuilding unity (going to take a bit) [08:05] didrocks: so in future, I guess the point is that we should open the downstream task, but don't touch its status ? (except marking it in progress) ? [08:05] that's what I do now [08:05] smspillaz: yeah, and ensure we only have the right downstream tasks attached [08:05] smspillaz: I think I'll raise that at the mini sprint and we should merge your tool and unify [08:05] ok [08:06] kind of, my tool is supposed to be run whenever you work on a bug [08:06] smspillaz: I guess we can automate that if you link your branch with a bug when merging [08:07] right, although you don't want to subscribe the SRU team for everything :) [08:07] at the moment, I'm doing stuff like [08:07] bzr make-bug -m 0.9.6 -s ubuntu-sru -d compiz -a unity 4.26.0 [08:08] which opens a bug, subscribes the right person, affects the right packages, updates the status, milestones it etc [08:33] smspillaz: have you been able to reproduce it? [08:37] morning mvo [08:40] didrocks: nope, still rebuilding :( === eeejay is now known as eeejay_is_afk [08:44] smspillaz: does unity blacklist some .desktop files or something? [08:44] smspillaz: virtualbox isn't showing up in the dash. [08:46] smspillaz: interestingly, neither is pointing devices [08:48] as in gpointing-device-settinsg. [08:49] hyperair: I don't write bamf, so you are best asking DBO that [08:49] ah okay [08:52] hey glatzor, good morning! [08:53] mvo, I am currently working on separating the AptWorker into three threads (download, worker, committer) [08:54] mvo, this will allow to preform the download of queued transactions during install time of another one [08:54] glatzor: !!! [08:54] glatzor: that sounds awsome [08:54] glatzor: anything I can look at already? [08:55] glatzor: it will be task based, right? so ideally instead of "commitpackage" u-m would give you a split up list of tasks? [08:55] mvo, I am currently pushing the work to the threading branch: lp:~aptdaemon-developers/aptdaemon/threading [08:56] mvo, the client API won't change [08:56] mvo, the worker thread use queues to put the currently processed transaction to each other [08:57] see the doc string for aptdaemon.worker.AptWorker [08:57] mvo, the test suite seems to be broken. [08:58] glatzor: right, so in order to utilize this best, e.g. update-manager would submit a bunch of self contained transactions instead of one huge CommitPackages ? [08:59] mvo, this could be a very good use case [09:00] mvo, I missed your point previously. [09:00] glatzor: thanks, and no worries :) just making sure I understand this correctly [09:01] mvo, another problem is ListUpdate. The acquire instance is not accessible. So I would have to reimplment it in python [09:01] glatzor: oh, hm, so python-apt needs tweaking to make this availalbe? [09:03] mvo, actually libapt :/ [09:04] mvo, But ListUpdate is a quite small and simple function. [09:04] mvo, it would be nice to have the RunScripts method availabe in python-apt. But even this one could be reimplemented purely [09:05] in python [09:06] good morning everyone [09:07] hey chrisccoulson [09:07] hi pitti, how are you? [09:07] hey chrisccoulson [09:07] hi didrocks [09:07] chrisccoulson: pretty well, thanks! how about yourself? [09:08] pitti - yeah, good thanks [09:09] glatzor: well, I think we should expose this in libapt/python-apt to make the amount of duplicated code smaller [09:10] hi chrisccoulson [09:11] hi njpatel. is there any reason why unity doesn't pass a launch context to g_app_info_launch when launching applications? [09:11] hi rodrigo_ [09:12] chrisccoulson, because it wreaks havoc with raising of the app window as we're in the window manager [09:12] glatzor: could you file a (small) bug with the requirements? [09:13] mvo, I will do so [09:13] thanks glatzor (and no rush :) [09:13] njpatel, i've been trying to figure out one thing which bugs me constantly (focus stealing prevention never works properly when launching applications), and that is probably why :/ [09:13] nothing adds a DESKTOP_STARTUP_ID to the launched applications environment [09:13] and that embeds the timestamp of the event which triggered the launch, which the launched application uses to initialize _NET_WM_TIME [09:14] so, in the case where i click on the icon for libreoffice and then go and do some other work whilst it's starting, libreoffice shouldn't steal focus when it opens [09:14] but it does everytime in unity [09:14] and i thought it was a compiz bug [09:24] chrisccoulson, riiight, if DBO is feeling better today, you guys can figure it out as it was his decision to drop it due to do the issue we were having [09:26] njpatel: were you at the release party last night [09:33] czajkowski, I was planning to be, then got into a discussion at millbank and it got too late and needed to get home :( [09:33] so, yeah, fail [09:33] or, njpatel fail [09:33] czajkowski, how was it? [09:33] aye I only heard you should have been there when talking to rickspencer3 [09:33] njpatel: great turn out [09:34] lots of people turned up good fun [09:34] sweet [11:01] hi czajkowski [11:03] rickspencer3: morning :) nice talking to you last night. [11:03] rickspencer3: on a quick note would you mind if I drop a quick pm to you [11:05] czajkowski, ditto, last night was fun [11:38] chrisccoulson, just answered to https://bugs.launchpad.net/desktopcouch/+bug/872639 [11:38] Launchpad bug 872639 in thunderbird-eds-extension ""Cannot open book: Could not create DesktopcouchSession object"" [High,Confirmed] [11:39] rodrigo_, thanks [12:15] * rodrigo_ lunch [12:19] have to run out for some errands, bbl [12:21] ok, I just renamed the french unstable forum to "Precise", it really started :) [12:21] chrisccoulson: I like the conversation view in thunderbird [12:22] didrocks, yeah, i've forgotten what it's like without that ;) [12:22] chrisccoulson: but for bug reports, it's not quite well, it shows the launchpad bot as the sender of every emails :/ [12:23] yeah, i noticed that too. i'll ping protz about that [12:29] thanks :) [12:30] apart from that, I love it [12:30] I don't have to expand the discussion view anymore [12:50] mvo, it seems that old gobject crashes quite a lot if using threading with python2.7 [12:50] mvo, the threading code is stable on my squeeze system :/ [12:50] mvo, so I uncommitted the merge of the first threading changes to trunk === m_conley_away is now known as m_conley === zyga is now known as zyga-afk [13:04] glatzor_: ok, the old gobject? did you try with gi.repository already? [13:07] mvo, no. the dbus server side is not ready yet [13:08] * mvo nods [13:23] mvo, oh, it is the flNotDir function that causes crashes apt-pkg/contrib/fileutl.cc [13:25] glatzor_: that is suprising, that is a really small function [13:25] glatzor_: and looks safe at least from first glance [13:27] mvo, Program received signal SIGSEGV, Segmentation fault. [13:27] [Switching to Thread 0x7ffff12bc700 (LWP 8778)] [13:27] 0x00007ffff2e07029 in std::string::rfind(char, unsigned long) const () [13:27] from /usr/lib/x86_64-linux-gnu/libstdc++.so.6 [13:27] (gdb) bt [13:27] #0 0x00007ffff2e07029 in std::string::rfind(char, unsigned long) const () [13:27] from /usr/lib/x86_64-linux-gnu/libstdc++.so.6 [13:27] #1 0x00007ffff30b5b80 in flNotDir(std::string) () [13:27] from /usr/lib/libapt-pkg.so.4.10 [13:27] #2 0x00007ffff30fdaa2 in pkgAcqArchive::Done(std::string, unsigned long, std::string, pkgAcquire::MethodConfig*) () from /usr/lib/libapt-pkg.so.4.10 [13:27] #3 0x00007ffff30e7739 in pkgAcquire::Worker::RunMessages() () [13:27] from /usr/lib/libapt-pkg.so.4.10 [13:27] #4 0x00007ffff30e90a7 in pkgAcquire::Worker::InFdReady() () [13:27] from /usr/lib/libapt-pkg.so.4.10 [13:27] mvo, another one: Program received signal SIGSEGV, Segmentation fault. [13:27] [Switching to Thread 0x7ffff12bc700 (LWP 8778)] [13:27] 0x00007ffff2e07029 in std::string::rfind(char, unsigned long) const () [13:27] from /usr/lib/x86_64-linux-gnu/libstdc++.so.6 [13:27] (gdb) bt [13:27] #0 0x00007ffff2e07029 in std::string::rfind(char, unsigned long) const () [13:27] from /usr/lib/x86_64-linux-gnu/libstdc++.so.6 [13:27] #1 0x00007ffff30b5b80 in flNotDir(std::string) () [13:27] from /usr/lib/libapt-pkg.so.4.10 [13:27] #2 0x00007ffff30fdaa2 in pkgAcqArchive::Done(std::string, unsigned long, std::string, pkgAcquire::MethodConfig*) () from /usr/lib/libapt-pkg.so.4.10 [13:27] #3 0x00007ffff30e7739 in pkgAcquire::Worker::RunMessages() () [13:27]