=== chriadam|away is now known as chriadam === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Namidairo`bnc is now known as Namidairo [02:48] I wonder if Mark will pitch in a few million to the Edge project if things get close. [02:49] As at the current projected donations. It's about to fall below the required sum (linier) http://ubuntu-edge.info/ [02:50] Hello [02:59] Woofcat, he (fortunately or unfortunately) said that he wouldn't: http://www.reddit.com/r/IAmA/comments/1j166z/hi_im_mark_shuttleworth_founder_of_ubuntu/cba4b3n === Ursinha is now known as Ursinha-afk [03:03] Woofcat: he won't pitch into the 32M, but he'll do what he needs to to make sure the backers get their phones if the project succeeds [03:09] Well damn, I kinda feel if the Ubuntu Forums were up it would have a better chance. === Ursinha-afk is now known as Ursinha [03:10] Since lots of people have questions, etc. what will the bootloader do etc. That could be roughed out, (obviously everything is subject to change in a project) [03:11] A since I don't want to seem like a whiney bitch, I already ordered mine. [03:13] SOrry if this has been answered by anyone else, but since updating the repos a little bit back I've been seeing this: http://pastebin.com/rUKyPsFh during building. Not sure if it's related at all to the bootloops I'm getting these days or not (yes I know they've switched to the flipped images), just figured I'd see if anyone might know. [03:22] Woofcat: I don't think you're being whiney, you're trying to help [03:22] Plus, the last few days have been weekend [03:22] which are usually slower for campaigns [03:22] dmj_nova: I just fear it getting to 15-20 million and stalling out. [03:23] That's fairly close, and Mark has 500+ million. Spare some interest, lol. [03:23] Woofcat: if it gets to 20 million, I'm pretty certain it won't stall out [03:24] (unless it's only reached $20 million in the final day) [03:24] It's getting from $7 million to $15-20 million that's hard. [03:32] I doubt it will reach 32 million, hope it does but probably not. [03:33] needs a funder [03:36] wilee-nilee: I'm cautiously optimistic [03:36] it will not be easy [03:36] and we all need to do everything we can to help it succeed [03:36] for sure I would fund it if I had the money. [03:37] but it's very possible [03:37] even small amounts can help [03:37] and spreading the word, explaining why it's important and what the benefits are [03:38] I spend about 12 hrs a day on the irc helping ubuntu users they get my value there in my own business I make 100$ per hour. [03:39] So I work as little as possible, money is not my main goal in life. [03:39] freedom is [04:10] hmm... apparently phablet-dev-bootstrap -c doesn't update the manifest file.. downloaded the most recent version and repo synced... seems to have fixed that issue... [05:21] Rollout plan for ubuntu touch on android devices | http://askubuntu.com/q/325912 [05:29] Wasapenin [05:30] Gooten tag? [05:30] Konickiwa [05:31] I come in peace [05:32] Someone told me ubuntu touch sucks anyways [05:34] That escalated quickly. [05:34] asafegous: can I help you? [05:36] I have a sprint s4 [05:38] And a single core 32bit box [05:38] Ok. [05:39] The problem is that my mobile device doesnt have ubuntu touch on it [05:39] I don't think anyone has ported Ubuntu Touch to any S4 devices [05:40] Does the s4 have the cooties? [05:41] How do i merge device branch to this badboy [05:41] Well, someone needs to do the port. We support Nexus devices, community people develop ports.. as you can see at https://wiki.ubuntu.com/Touch/Devices [05:42] I got that. Right about now i need some si se pude [05:42] si se pude? [05:42] Puede [05:42] Yes i can [05:43] The Sprint S has the additional issue that it's CDMA which we don't currently support. [06:39] can i install ubuntu touch on Samsung Galaxy Note II ? [06:41] Dhiren: https://wiki.ubuntu.com/Touch/Devices is it listed there? [06:44] Edge campaign just broke $7M [06:46] indeed [06:46] \o/ [07:03] good morning [07:16] Hey guys.. [07:16] Need some quick help if anyone is open [07:17] I cannot get past the command brunch [07:17] the error that comes up is something along the lines of "Variant Kernel" and a couple others [07:40] Anyone wanna help a noob? === chihchun is now known as chihchun_afk === schwuk_away is now known as schwuk [08:27] Laney: hi! I fixed the issue with https://code.launchpad.net/~mardy/ubuntu-system-settings/plugin-invocation/+merge/176929, please let me know if it's acceptable (I don't mind your solution either, it's just that I believe this is what potential contributors would be more familiar with) [08:28] mardy: OK, let's take it for now, thanks === chihchun_afk is now known as chihchun [09:07] Good morning all, happy Tiger Day! :-D === schwuk is now known as schwuk_away === schwuk_away is now known as schwuk === vrruiz_ is now known as rvr [09:39] any idea when ubuntu touch will work on n4 with android 4.3 bootloader/radio? [09:51] user82: i dont think we have tested that [09:51] popey, ogra_ said it will break [09:51] mine did not want to flash properly, should i try again? [09:51] probablly a clockworkmod issue [09:52] * ogra_ definitely didnt say that [09:52] i said we dont support it yet [09:52] aah [09:52] then i will try again... [09:52] try what ? [09:53] you wont be able use 4.3.3 without porting work ... and thats currently not a high prio on our side [09:54] (feel free to port and send patches indeed) [09:57] i got someone here who claims his phablet-flash wants to load the image from http://cdimage.ubuntu.com/ubuntu-touch-preview/daily-preinstalled/current/ but it should be http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/current/ . is this just a not uptodate phablet-flash or is this a regular case? [09:57] oSoMoN: nerochiaro can you take a look? https://code.launchpad.net/~sergiusens/notes-app/testing_artifacts/+merge/177351 [09:57] k1l_: old phablet-flash probaly [09:57] k1l_: apt-cache policy phablet-tools [09:57] oSoMoN: nerochiaro diff has been uploading forever though [09:58] k1l_: 0.15+13.10.20130723-0ubuntu1 is what I have [09:58] ogra_, i try the phablet tool again [09:59] popey: yep, will tell him to run the updates and will look for that version he uses [10:01] btw phablet-flash messes up for me. it pushes files to /sdcard/ and then tells to install from /sdcard/0/. is that a bug? [10:02] user82: what device? [10:02] a nexus 4 [10:02] sounds like you have multi-user active [10:02] popey, yes indeed [10:02] can i disable that? [10:02] i suspect so [10:02] in settings somewhere [10:03] allright thanks [10:03] I dont think we have tested flashing on a multi-user setup device [10:03] long time ago it was active...wonder why it still exists [10:03] i might test though and see if i can file a bug abpout it [10:04] sergiusens: looking. it's basically adding test coverage measurements, right ? should have no impact on the actual tests being run [10:04] should not be a big deal.. [10:04] noone really has it active on phones [10:04] nerochiaro: yeah, just coverage.xml and xml for each unit [10:05] mardy: have some time to halp me with my lack of Qt knowledge? I have a QSortFilterProxyModel that's backed by a QAbstractTableModel populated by a worker thread. I want it to be dynamically sorted whenever the thread supplies a new row but the sort isn't being applied. I guess I'm not calling some method correctly to notify the proxy that it needs to apply the sort, but what? lp:~laney/ubuntu-system-settings/timedate-populate-thread [10:05] user82: yeah, i have it active on nexus 7 though, family dablet [10:06] Laney: I think there was some flag to set to have the filter work live, let me check... [10:06] mardy: yes, setDynamicSortFilter(true), which I'm calling [10:06] popey, do you know how to revert it? i guess it breaks to move everything from /sdcard/0/ to /sdcard/? [10:06] it's supposed to sort by the display name by default which indeed is what I want [10:06] but it just doesn't happen [10:07] popey: dablet? A tablet you dabble with? [10:07] welcome back popey [10:07] my lessThan is never called either (but that shouldn't be needed) [10:07] so I guess I'm missing some signal emission or something [10:08] user82: no idea [10:09] allright thanks. [10:09] Laney: did you try to check whether updating your QAbstractTableModel works fine (by removing the proxy)? [10:09] i would end up googling for the answer [10:09] mardy: I see all of the results through the proxy [10:09] Laney: maybe you are missing "beginInserRows"/"endInsertRows" and similar [10:09] Laney: ah, and they simply are not sorted? [10:09] have those [10:09] correct [10:09] sergiusens: looks ok to me, if you made sure everything still builds fine i'm just going to approve it [10:11] thanks davmor2 [10:11] Laney: is the model empty, when you connect it to the filter, or has it some data already? [10:12] mardy: I guess it's empty [10:12] popey, i moved everything from /sdcard/0/ to /sdcard. let me see if it breaks! [10:12] mardy: the constructor of the model starts the worker thread [10:13] Laney: and add a Q_DECL_OVERRIDE next to the declaration of the lessThan() method, to make sure that the signature matches that of the virtual method :-) [10:13] popey: It's only so I don't have to kill my phone first with the daily updates ;) [10:13] Laney: other than that, I wouldn't know... can I see the code? [10:13] mardy: I gave you the branch earlier ;-) [10:13] lp:~laney/ubuntu-system-settings/timedate-populate-thread [10:13] Laney: oh :-) [10:14] lessThan shouldn't even be needed though if I want to use the default display name sorting [10:14] IIUC [10:14] takes _ages_ to run all the autopilot tests on a device [10:15] over an hour so far [10:15] apw, could someone from the kernel team look into https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-unity8-autopilot/2/console ... seems that "WARNING: at /build/buildd/linux-mako-3.4.0/net/wireless/scan.c:590" causes a network re-connect which in turn breaks tests [10:17] ogra_, see what we can do. old poc kernels you guys insist on [10:17] ogra_, which one is mako ? is that N4 [10:17] not we ... vendor blobs do [10:17] yeah [10:18] ogra_, using vendor blobs i blame on you, therefore ... QED ... you are to blame for global warming :) [10:18] haha [10:18] nerochiaro: it shouldn't, I tested locally, but we can wait for -ci to return === chriadam is now known as chriadam|away [10:22] ogra_, it works with android 4.3 stuff. my mistake with the multiuser chaos [10:23] sergiusens: i comment-approved it, when jenkins is done with CI feel free to apporove it yourself if everything passes [10:23] nerochiaro: great, thanks [10:23] user82, oh, you havent been clear that you wanted to flash *onto* an android 4.3 device, i thought you asked when ubuntu touch will use 4.3 [10:23] Is there a bug for the new phablet images going adb-AWOL? The Nexus4 keeps vanishing from the "adb devices" list [10:23] user82, sorry for misunderstanding [10:24] oSoMoN: your MR is approved too [10:24] nerochiaro: thanks [10:24] ogra_, yeah that was what i did. with bootloader 20i and radio 84 (i believe) [10:24] duflu, works fine on a g-nexus here [10:24] Hmm [10:24] zsombi: hi, welcome back. when you have a few minutes can you please ping me ? there's an MR i submitted for fixing IMA issues that I'd like to discuss with you [10:24] and on my n7 [10:25] Well, a lot has changed since I did any Android hacking. But I would have thought a stable adb connection wouldn't [10:27] Actually, it looks like the kernel isn't detecting any USB connect/disconnects now I've updated saucy === jdstrand_ is now known as jdstrand === gusch is now known as gusch|lunch [10:33] I see. It's the Nexus which keeps disconnecting. Something's up with the new phablet image I got today. That would be a good place to end for the day [10:35] nerochiaro: howdy, I have ~25 mins till standup, so if that time is enough, we could handle that too [10:36] nerochiaro: do you have the link to the MR? [10:41] oSoMoN: https://code.launchpad.net/~sergiusens/webbrowser-app/testing_artifacts/+merge/177357 [10:44] Laney: FYI, I'm quiet because I'm investigating. :-) [10:44] mardy: :-) [10:44] It's foxed me a bit - I hope someone else can get further === sil2100_ is now known as sil2100 [10:45] Laney, i don't think you can modify a model in a separate thread while a view is attached to it [10:45] sounds very thread-scary [10:47] mhr3_: the worker thread passes the items back to the originating one as they become available [10:47] sergiusens: looking, thanks [10:49] Laney: so, one thing I noticed is that if I remove the call to setFilterRegExp in the constructor, the lessThan() method is called [10:49] Laney, ah, missed that, then it's safe indeed [10:49] Laney: *many times*, making the app unresponsive for several seconds [10:50] mardy: right, I'd expect it to only operate on small sets [10:50] using the proxy should make this ok [10:50] Laney: yep. Though it's weird, according to the docs doing both both filtering and sorting should be fine [10:50] nerochiaro: https://code.launchpad.net/~sergiusens/share-app/testing_artifacts/+merge/177359 [10:51] indeed [10:51] Laney: it actually works [10:52] is there some problem telling QML to update its view? [10:52] Laney: I mean, when I type something in the search box, results are sorted [10:52] wtf?! [10:53] Laney: I mean, by removing setFilterRegExp in the constructor [10:53] ah [10:53] Laney: maybe that's because the set is sorted initially, and then the filtering is easy [10:55] ogra_, those warnings are literally that, warnings it is not entirely clear they would trigger anything, are we sure they are ? [10:56] Laney: ok, I got some good news: remove both the call to setFilterRegExp and sort() in the constructor, and add m_timeZoneFilterProxy.sort(0); at the end of timedate::setFilter [10:56] nerochiaro: one more https://code.launchpad.net/~sergiusens/share-app/testing_artifacts/+merge/177359 [10:56] Laney: for some reason then it works, though it's really slow if you type just one letter [10:57] Laney: actually it's not slow, if you remove the qDebug() line in lessThan() === MacSlow is now known as MacSlow|lunch === pete-woods1 is now known as pete-woods [11:02] mardy: Was hoping to not have to do explicit sorting, but oh well [11:03] now how do I get back the initial regexp? [11:03] Laney: it seems you need to call sort() just once, in fact [11:03] could I connect to a signal when the first item comes in then? [11:04] Laney: maybe, let me try a bit more [11:04] ok [11:04] Laney: actually, you don't need to remove the calls to setFilterRegExp [11:05] oh yeah, that does work [11:05] let me try remembering if I've sort()ed before and only doing that once [11:06] Laney: I actually didn't test if that would work, mine was just a guess; it maybe that changing the filter disrupts the sorting [11:07] anyone open to talk today? [11:08] murgero: wassup? [11:08] I am having some issues compiling CM10.1 for ubuntu touch [11:08] * popey points murgero at sergiusens or rsalveti [11:09] @sergiusens hello? [11:09] I am also new to IRC :/ [11:09] * murgero [11:09] * murgero sergiusens hello [11:09] mardy: I pushed to the branch, have a look now [11:09] * Laney does an armhf build [11:10] murgero: I'd waitr fo [11:10] bah [11:10] murgero: I'd wait for them to reply, they may not be awake right now, in other timezones [11:10] murgero: what's the issue? [11:10] murgero: but agree with popey, don't ultra ping ;-) [11:10] Sorry! I am new to IRC [11:10] anyway [11:11] I am getting a variant kernel error, along with error code 1 and 2 from make. [11:11] I have started from sctratch like 6 times no luck [11:11] I am trying to follow the tutorial too, but I am new to porting on linux. [11:12] murgero: please pastebin the compile output [11:12] seb128 (or anyone else), I'm trying to debug system-settings to figure out how it connects to PulseAudio. Is there a way I can get log information (or similar out of it) ? [11:12] I would love to, but I am currently at work, progress can be checked at my web page: http://urgero.org [11:13] murgero: offramp.urgero.org.wsadns.com unknown host [11:13] specifically.. "Description: Unable to locate the server named "offramp.urgero.org.wsadns.com (2)" --- the server does not have a DNS entry. Perhaps there is a misspelling in the server name, or the server no longer exists. Double-check the name and try again." [11:13] murgero: well I can't locate that host it seems and make 1 and 2 error could be anything [11:13] type it manually : http://urgero.org [11:13] Laney: seems to work fine [11:13] I am using that server now to use IRC. [11:13] indeed [11:14] murgero: nope, still fails [11:14] diwic, run it on a command line? but no, there is no explicit debug flag for it [11:14] I wonder/hope it actually speeds it up on the device :P [11:14] Laney: if you have some time, I'd recommend building a smaller test case and filing a bug to Qt [11:14] My ISP is updating DNS right now :/ I'll pastebin a typed out copy from a screenshot [11:14] diwic, system settings doesn't do anything special, it just use qtmultimedia [11:14] mardy: ok, will try to do [11:14] thanks for the help [11:14] Laney: yw [11:15] Laney: maybe it's really a bug: http://stackoverflow.com/questions/16222797/sort-a-qabstractlistmodel-derived-model-by-role-in-qml-listview [11:16] diwic, if you want a simpler testcase: http://paste.ubuntu.com/5924884/ [11:16] diwic, just save that as a file and run qmlscene savedfile.qml [11:16] seb128, thanks, will try [11:16] Here is the last like 20 lines output that I can read: http://pastebin.com/WaZTwZSh [11:17] mardy: which part of that do you think is the same as this case? [11:18] sergiusens: so? [11:18] Laney: maybe the bus is that if you don't call sort() after all the items are retrieved, it won't work [11:18] Laney: but I really don't know :-) [11:18] :P [11:19] I guess with this workaround/solution we're doing more sorting than we need to [11:19] maybe not though [11:19] dunno [11:21] murgero: you need to remove the dalvik specific targets for PART2 and I'll need to see your device repo to figure out PART 1 [11:21] D2USC is the repo I am using [11:21] what do you mean remove the dalvik specific targets? [11:23] Hey sorry about that, lost connection. [11:24] ws/help [11:25] gema, not sure this is a QA issue or ui toolkit.. but bug 1206088 and bug 1206086 are causing autopilot tests to fail on my mako device. [11:25] bug 1206088 in Sudoku App "Autopilot tests get wedged in sudoku-app" [Undecided,New] https://launchpad.net/bugs/1206088 [11:25] bug 1206086 in Ubuntu Terminal App "Autopilot tests get wedged in terminal app" [Undecided,New] https://launchpad.net/bugs/1206086 [11:29] Hi folks [11:29] What's up? [11:29] does anyone know where the contacts are in the "latest" preloaded daily? [11:29] I am still trying to figure out how to build it lol [11:30] its not where the release notes say [11:30] Do you have the right release? [11:30] im talking about the phablet image.... [11:30] ah. [11:30] I do not know yet sorrry [11:30] yes, it works well [11:31] what are you building/running? [11:33] I am running a D2USC (S3 US Cellular Variant) [11:33] phablet Saucy build [11:34] ah ok [11:34] woking ok? === hikiko is now known as hikiko_lunch [11:42] I am still trying to build it. I am having Environment issues. [11:45] This is a good question: http://askubuntu.com/questions/326004/will-canonical-disable-amazon-and-similar-lenses-by-default [11:45] If anything, the enabling of such "non-friendly" stuff should be asked upon installation, the same way it asks in Ubuiquity about asking MP3 stuff [11:46] *Ubuntu [11:49] * murgero brb [11:49] * murgero back [11:49] cool [11:49] wtf? Issues with IRC brb [11:52] is it correct that https://launchpad.net/ubuntu-download-manager will provide the ability for apps to download files in the background on Ubuntu touch devices? [11:53] and if so, will there be a qt/qml wrapper or do I need to write one myself? === MacSlow|lunch is now known as MacSlow === hikiko_lunch is now known as hikiko === alan_g is now known as alan_g|lunch === _salem is now known as salem_ [12:23] popey: on what image? [12:27] gema: Ubuntu Saucy Salamander (development branch) - armhf (20130726.1) [12:27] popey: I don't think we are yet running those tests you are describing, right? [12:28] gema: you might not but I am ☻ [12:28] I am running all autopilot tests for all apps [12:28] popey: ok, what do you want me to do with those bugs? [12:28] gema: well, nothing specifically, i wondered if you'd seen the issue at all, and whether it might be an autopilot one or a toolkit one [12:29] so i can assign to the right person [12:29] popey: no, I haven't, you may want to talk to the sdk folks and see if they have [12:29] or tell them to triage === jhodapp|afk is now known as jhodapp [12:29] gema: ok. when you guys run autopilot, and it gets stuck, does someone look at the screen? [12:29] i.e. can they tell it's wedged? [12:30] popey: no, by the time we get to the stuck phone it is a matter of pushing the power button [12:30] gema: that's what I expected ☻ [12:30] thanks === nerochiaro is now known as nerochiaro_food [12:36] cjwatson: here's the output of strace for the packagekitd failure: http://ubuntuone.com/3XWc7La7lJc9DZ3wqgcScQ [12:37] and here's the actual output of packagekitd during that run: from that run with strace: http://pastebin.ubuntu.com/5925102/ [12:40] alecu: I really must do something about the error reporting here. But you're running into https://bugs.launchpad.net/click/+bug/1204523 [12:40] Launchpad bug 1204523 in click "Click packages can't be installed from directories not viewable by the clickpkg user" [Undecided,New] [12:40] alecu: I'll see if I can fix that today [12:41] cjwatson: great, I'll see if I can find a workaround for tomorrow's demo. Thanks. [12:41] alecu: I should be able to fix it before the demo; but a workaround is to make the download directory readable by the clickpkg user. [12:41] (I'd like to avoid having to actually copy the file to a temporary directory, as that could be slow.) [12:44] sergiusens: hi [12:46] slangasek: hey [12:48] sergiusens: hi! so it's been mentioned to me that you might be working on a patch to phablet-tools to let us bootstrap specific (non-current) system-update images [12:49] sergiusens: is that true... and is that something you might have soon? It would be helpful to me to have it before Wednesday, since I'm running a demo and it's difficult to prepare a demo if I can only run any given update once ;) [12:49] slangasek: true, when's that demo btw? I can rush it if urgent [12:49] sergiusens: Wednesday before you wake up [12:50] if not EOD today [12:50] slangasek: good, it should be done EOD [12:50] sergiusens: great, thanks! === dednick is now known as dednick|afk === oreneeshy_ is now known as oreneeshy [13:04] slangasek, just get lots of devices :) [13:07] cjwatson: hey, I just noticed that when I install a new version of a click package (eg, evilapp 0.1 and evilapp 0.3), then I have two desktop files in ~/local/share/applications: [13:07] com.ubuntu.developer.jdstrand.evilapp_evilapp_0.1.desktop [13:07] com.ubuntu.developer.pedrocan.evilapp_evilapp_0.3.desktop [13:08] cjwatson: I'm thinking that is not intended, cause shouldn't the user that installed the upgrade get the latest version? [13:08] jdstrand: They're different package names [13:08] jdstrand != pedrocan [13:08] cjwatson: sorry, yes, those are [13:09] cjwatson: but I do have: [13:09] com.ubuntu.developer.jdstrand.evilapp_evilapp_0.1.desktop [13:09] com.ubuntu.developer.jdstrand.evilapp_evilapp_0.2~unreleased.desktop [13:09] That's indeed unintended. Bug please? === jounih_ is now known as jounih === dednick|afk is now known as dednick [13:12] guys [13:12] ueventd is taking 100% on the phone [13:12] any idea? [13:14] cjwatson: bug #1206115 [13:14] tsdgeos: reboot [13:14] bug 1206115 in click (Ubuntu) "previous version of desktop file not removed on upgrade" [Undecided,New] https://launchpad.net/bugs/1206115 [13:14] tsdgeos: we have a race with ueventd and udev sometimes [13:14] rsalveti: ok, that helped [13:14] cjwatson: note, this is with click 0.2.6 [13:14] we're working on a fix as we speak [13:15] awesome [13:15] rsalveti: now it's sensorsservice taking 100% :D [13:15] is that similar? [13:15] tsdgeos: think so [13:15] oka [13:16] tsdgeos: which image are you using? [13:16] --pending from today [13:16] right, then just should be the race [13:16] oki [13:16] we had an issue last week with the /factory partition, but that's already fixed with latest === alan_g|lunch is now known as alan_g [13:27] popey, ogra_: right how are the images looking today? [13:28] davmor2, so so http://reports.qa.ubuntu.com/smokeng/saucy/ [13:28] ogra_: I'm glad I don't have a manta :) [13:29] well, yeah, the tablet builds are suffering from slight ignorance [13:29] ogra_: :D [13:29] worse is the three failed tests on mako [13:29] it regressed. so cant release it [13:30] without looking i guess its the same issue i pointed out to apw from yesterdays log [13:33] ogra_: from apt-get updates there were issues with both the browser and facebook, so today I'm doing a flash + --wipe just to see if those got resolved, browser seg faulted with some lovely warnings and the facebook still failed to open even after the browser was fixed [13:38] ogra_: but seems weird, unity8 was 100% for maguro, 90.9% for mako === dduffey_afk is now known as dduffey [13:38] and sdk failed for mako as well [13:38] rsalveti, wee the packlog :) [13:38] *see [13:39] there is a kernel warning oops ... [13:39] and it seems that triggers a network re-connect so dbus looses its address [13:39] at least thats my theory [13:40] ogra_, the networking one? [13:40] since the symptom seems to be random dbuss issues [13:40] ogra_: oh, that's weird [13:40] apw, yes [13:40] but how would wifi be related with dbus? [13:40] ogra_, those seem to be completely warnings about bad beacons; and we only show us connecting to the network once as far as i can see before the first one [13:40] if i look at the logs i see the kernel warning and right after the next dbus call fails === nerochiaro_food is now known as nerochiaro [13:40] ogra_, as we log about IPv6 not working at the point we ocnnect [13:40] dbus is local only so why would netwokring matter to it i wonder [13:41] apw: but should be harmless still, right? [13:41] * cjwatson tries to think of a decent way to arrange for an arbitrary user to be able to read an arbitrary file, given that I can run some code as root first, but would rather not have to modify the permissions on the original inode [13:41] cjwatson, can't you open that file as root, then hand the open r-o file descriptor on [13:42] I could create a readable temporary directory and create a hardlink in that, but only if temporary directories are created on the same fs and only if the inode itself is readable [13:42] a hard link would have the same ownership though [13:42] apw: Additional constraint I forgot to mention is that it must be possible to seek back and forward in the file. But maybe that will still work [13:42] cjwatson, you can seek in an open file if it is a file on disk [13:42] I think the reason I mentally discarded that option is incorrect [13:42] ogra_, popey: oh that's an interesting bug, flash the phone with a wipe, Wifi is enabled but not connected, 3g is disabled. Click on the network to connect to and it enables 3g too [13:42] So thanks for correcting my brain-failure :) [13:43] heh :) you do it for me all the itme [13:43] davmor2: i thought 3g was enabled by default [13:44] it should, yeah [13:44] ogra_, so where is that dbus failure between ? ie who was talking to who when they noticed; it is possible we handed the results to NM (after the warn) and it didn't like the content either [13:44] popey: let me reflash ans screenshot as I go [13:44] ogra_, as the warning is about contents which make no sense to the kernel but it take no action other than warning [13:44] apw, my left side of brain was talking to the right one ... nobody else involved :) [13:45] apw: I do have to hand over an actual path, not just an fd, but maybe I can use /proc/self/fd/ if sufficiently careful [13:46] ogra_, heh i meant in the dbus conversation [13:46] cjwatson, yeah that is not an uncommon way to do it [13:46] cjwatson, couldnt you do a bind mount having the mountpoint have different perms ? [13:47] ogra_: Possibly, but I think apw's way is easier, and certainly complexifies the test suite less [13:47] apw, i think i need to wait for doanac to investigate that part [13:49] ogra_, ok let me know what comes of it === dednick is now known as dednick|lunch [13:57] ogra_: hnngh :) [13:58] slangasek, ? [13:58] oh [13:59] heh [14:02] ogra_, popey: post flash http://ubuntuone.com/7NuNZPUmYrj4XSizxjQu0p, enable my wifi connection http://ubuntuone.com/2thsWTGY55o2Saxpi32zyx, after adding password http://ubuntuone.com/0FNN6lTQMFDfhfW2tth9qK [14:03] davmor2: interesting bug [14:03] will MATLAB run on Ubuntu Mobile? | http://askubuntu.com/q/326053 [14:04] davmor2, did you try just öeaving it sit for a moment might just be a timely coincidence [14:04]