[00:00] or it will be another hardware? [00:00] Neo31: http://www.omgubuntu.co.uk/2014/12/bq-ubuntu-phone-launches-in-europe-this-february/ [00:02] a repurposed version of the its popular Aquaris e4.5 handset preloaded with Ubuntu for Phones. all clear popey thank you :) I will wait for this one [00:02] np [00:03] do you know some friends in Spain that could help me get one? (International credit cards are not allowed in my country Tunisia) But I might be able to manage to send the money in other ways [00:04] I don't know, sorry. [00:04] I wonder if they'll start showing up on ebay :) [00:05] ebay is good, i might be able to get some cash on paypal [00:06] thank you [00:06] :) [00:06] no problem === _salem is now known as salem_ === chihchun is now known as chihchun_afk === salem_ is now known as _salem === chihchun_afk is now known as chihchun [00:49] kenvandine, right, weird failure [00:50] twice in a row, so consistent === chihchun is now known as chihchun_afk === chriadam|away is now known as chriadam === mnr is now known as Guest37553 === marcust__ is now known as marcustomlinson === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk [07:46] Hi everyone! === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk [07:50] I'm fresh Ubuntu Touch user and I wonder if I have the newest version on my Nexus4. In OS Bulld Details there is OS Build Number -> 1, Ubuntu image part -> 20141201 and Device image part -> 20141119 [07:54] I've install it from devel channel [08:08] good morning [08:21] hi [08:53] mvo: How this MR says "Merged" when the change is not in the target branch? https://code.launchpad.net/~bzoltan/click/vivid-transition_mirrors/+merge/246913 [08:54] bzoltan: I think I set it manually as its in lp:click/devel now, but you are right, if the goal is vivid then that actually incorrect, sorry for that [08:54] mvo: no problem, I figured out that it is merged to lp:click/devel [08:55] mvo: I need to make a fix on that. I create a UI switch in the QtCreator to make the mirror archive optional. As in many cases the mirrors are badly out of sync, so we need to offer a fallback to the slower but sure archive. [09:18] can someone explain to me what kind of additional source/drivers are needed for a CM port to a samsung device? after all samsung puts all their source on opensource.samsung.com [09:44] Good morning all; happy Weatherman’s Day! :-D [09:50] Mirv: ping [09:55] Mirv: unping [09:56] tsdgeos: :) [10:23] bzoltan: aha, nice! [10:28] oSoMoN: any hint of why https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1417963/comments/2 might be happening or where in the code do i look for the code that handles that selection rectangle? [10:28] Launchpad bug 1417963 in webbrowser-app (Ubuntu) "webbrowser-app autopilot test regression with Qt 5.4" [Undecided,New] [10:31] tsdgeos, that’s the code for the selection rectangle: http://bazaar.launchpad.net/~phablet-team/webbrowser-app/trunk/view/head:/src/Ubuntu/Web/Selection.qml [10:31] oSoMoN: ok, will have a quick look [10:31] thanks! [10:31] * tsdgeos has the "let's fix the Qt5.4 bugs so we can migrate" hat on [10:32] I have it on my list of things to look at too, but I won’t get to it before tomorrow at the earliest, so if you can fix it in the meantime, that’s good :) === alan_g is now known as alan_g|reboot [10:50] kenvandine: to make a standalone email/irc/chat HTML5 app, with JS socket API in oxide bug 1379303 [10:50] bug 1379303 in Oxide "Need support for a JS socket api" [Wishlist,Triaged] https://launchpad.net/bugs/1379303 [11:13] oSoMoN: i think i know what's wrong in oxide [11:13] just need to recompile it all [11:13] probably will take a while :D [11:14] tsdgeos, so the selection bug is an issue in oxide? [11:14] oSoMoN: Qt 5.4 changed the way QVariants are passed from qml to C++ [11:15] they're now passed as a QVariant that is a QJSValue and you need to unwrap it [11:15] and oxide is doing stuff like [11:15] if (message.type() != QVariant::Map && [11:15] message.type() != QVariant::List && [11:15] message.type() != QVariant::StringList) { [11:15] which will fail [11:15] it's a one line change to unwrap it [11:15] aha [11:15] i've done that fix in various places already [11:16] tsdgeos, ping me when you submit the fix, I’ll happily review it [11:16] oSoMoN: need to compile oxide first, getting this [11:16] oSoMoN: http://paste.ubuntu.com/10071270/ [11:16] any clue? [11:17] i guess it doesn't like my chroot [11:18] tsdgeos, have you run client.py in the root dir first? [11:18] nope [11:18] wait, nevermind, if you’re building from the source package that’s not necessary [11:18] I’m not sure what that error is, try asking on #oxide [11:19] gotta run an errand, will be back in ’30min === chriadam is now known as chriadam|away === _salem is now known as salem_ [12:08] oSoMoN: dbarth: I approved the silo. [12:09] rvr, thanks === JamesTai1 is now known as JamesTait [12:13] kenvandine, morning, ping me when you're here? === MacSlow is now known as MacSlow|lunch [12:25] Salve === alan_g is now known as alan_g|lunch [12:53] popey: but seriously, what is the "official" emai-client [12:53] * Chipaca hides [12:55] hey guys. quick question. Where are the screenshots taken with phablet-screenshot stored? [12:58] rvr: thanks [13:00] muyiscoi: wherever you specify [13:00] muyiscoi: Usage: phablet-screenshot [OPTIONS...] [13:00] muyiscoi: filename is not optional [13:03] Chipaca: Hey. A test case is needed for silo 7 (notification flooding). [13:04] rvr: i've added a comment [13:04] rvr: let me know if you need more than that [13:04] Chipaca: Where? [13:04] in the spreadsheet, here let me transcribe [13:04] To test: start everything, set twitter up with user A. Send user A a DM. Call the “Poll” dbus endpoint. Repeat five times. See the notification centre list all five DMs separately. Install update. Compare. [13:04] rvr: ^ [13:05] rvr: to be clear, the "repeat five times" is to the pair "DM; Poll" [13:05] as if you send more than one DM in a single poll it'll collapse them with the old version too === chihchun_afk is now known as chihchun [13:06] Chipaca: How's the "Poll" dbus endpoint called? [13:07] rvr: from the wiki testplan: gdbus call --session -d com.ubuntu.AccountPolld -o /com/ubuntu/AccountPolld -m com.ubuntu.AccountPolld.Poll [13:07] Chipaca: Ack [13:09] Chipaca: somewhat unrelated, what would it take to add a new service to account-polld? [13:09] just an mp? [13:09] cwayne1: I don't think you need to involve parliament [13:10] cwayne1: :) yes, an mp [13:10] cwayne1: what're you wanting to add? [13:10] i'd rather movement was *away* from polld than *towards* it, but such is life [13:32] Chipaca: cwayne1 account-polld should be EOL, but I welcome untappd polling ;-) [13:32] untappd? [13:32] sergiusens: thats what i was considering :P [13:32] that and fitbit [13:33] that works I guess :-) [13:36] cwayne1: they have no sane way of making it a server-side thing and using regular push? [13:38] not that i know of [13:38] maybe i could do it on my server or something [13:40] cwayne1: if they do oauth2, you can probably make it work [13:40] if you also control the app [13:40] polld is only for when you don't control _anything_ :) [13:40] i control nothing except a scope === alan_g|lunch is now known as alan_g [13:47] cwayne1: :( [13:50] cwayne1: you'd need some changes in your scope for it to “handle” the notifications [13:51] cwayne1: and you might be the first scope to attempt that, so we might find bugs in the store verification script whatsits [13:51] cwayne1: but it should both work and be fine [13:52] if I make a qwidget application (just a basic qlabel saying "hello") and I run it on the phone should it show? currently it seems to run but doesn't display anything on the display [13:53] Riddell: how're you running it? [13:53] cwayne1: bottom line, give it a go, holler if you get stuck [13:55] Chipaca: ssh in and run https://paste.kde.org/p3rkm7oot [13:55] Chipaca: where squidge is the random name of my binary and dialer-app.desktop is a random .desktop file [13:57] cwayne1: Chipaca well untappd has an app and a scope [13:57] Chipaca: is there no way to do notifications without requiring a server? [13:57] sergiusens: it's just a webapp though [13:58] Riddell: hm. I know very little about this, however, have you tried making an actual .desktop file for your app, putting it in one of the canonical (xdg) places, and using ubuntu-app-launch ? [13:58] dobey: push notifications? go via the server [13:58] dobey: postal notifications? on the phone [13:58] dobey: the interface between push and postal is messy, to the point where an app wanting to do postal needs to have a push helper, but it enables stuff so [13:58] Chipaca: notification of some event on a web service that i don't control [13:59] dobey: tell me how you expect it to work, and i'll tell you what you've been smoking :) [14:00] Riddell: or, if that's more your thing, inspect the environment of a running app and guess at what all is important for your app to work [14:00] Riddell: or wait for somebody who knows what they're talking about :) [14:00] Chipaca: well, how does twitter work? [14:00] dobey: via polld [14:00] dobey: sadly [14:00] so it's pull notifications, not push? :) [14:01] dobey: twitter on the ubuntu phone does not use push notifications [14:01] at all [14:02] dobey: neither does gmail [14:02] dobey: nor facebook [14:02] rvr, how goes it? [14:02] Chipaca: yeah, this is the problem with push notifications, because it requires buy-in from the companies that actually control the services. and they don't want to do crap until you have a billion phones on the market with a billion users using their service on your phone :-/ [14:02] Chipaca: does anything other than system image updates? [14:02] dobey: yes, that's why polld was created, to bridge the gap for the apps we care about [14:03] dobey: yes, several things [14:05] Chipaca: is there a visual diagram of how push notifications actually work, somewhere? like that shows how much access the third party server has to ubuntu services, and how much access ubuntu services have to those third parties? [14:05] dobey: if you control the app and the service has oauth2 (or you can in any way get credentials onto the device for the backend to use) you can write a server-side thing of your own to bridge the gap [14:05] dobey: yes? maybe? not sure i understood your question. have you looked at the push docs? [14:05] jgdx, hey, i forgot to mention in that comment that i wasn't even changing which modem was 3g [14:05] dobey: http://developer.ubuntu.com/en/apps/platform/guides/push-notifications-client-guide/ [14:05] jgdx, i was just changing the tech pref on the first modem [14:05] push.svg [14:06] that image should be illustrative [14:06] and it is [14:06] just not of what i was wanting to illustrate [14:06] jgdx, i did the same thing on mako and it worked fine [14:06] shouldn't that, be an actual image? [14:06] rather than some html shown as text? [14:06] yes, yes it should [14:07] and if i just add push.svg to the url, it doesn't load... [14:07] jgdx, it's like the tech pref switch takes a really long time, but while it's changing it's disconnected [14:07] dobey: ze somezing, eet ees broken [14:07] yes it is [14:08] dobey: http://bazaar.launchpad.net/~ubuntu-push-hackers/ubuntu-push/automatic/download/head:/push.svg-20140728193339-5font1q8tkladxu0-2/push.svg === marcusto_ is now known as marcustomlinson [14:10] kenvandine, right, and you had not hot swapped sims [14:10] kenvandine, cause I saw the exact same thing after hot swap && no reboot [14:11] jgdx, i had rebooted [14:11] in fact i rebooted a couple times and retried [14:11] same thing [14:11] hm [14:11] jgdx, what i haven't done is tried the same thing without the silo [14:11] jgdx, i'll do that this morning [14:12] kenvandine, that'd be great. [14:12] jgdx, it doesn't feel like a bug in settings, it's like ofono/NM isn't happy [14:12] story of my week [14:12] yeah [14:12] jgdx, the good news is it worked fine on mako :) [14:12] :D :p [14:12] the bad news is krillin is what needs it :/ [14:13] kenvandine, when you test, you could do this outside of u-s-s using ofono scripts, just to rule u-s-s out completely. [14:13] jgdx, cool, can you give me the scripts? [14:13] kenvandine, also, I will try this again on krillin later (at uni) to confirm it [14:13] i'll do that before i revert from the silo [14:14] grr, need to swap the SIM first [14:14] kenvandine, /usr/share/ofono/scripts/set-tech-preference /ril_[0|1] gsm; [14:14] ok [14:14] and umts? [14:14] umts [14:14] right [14:15] ok [14:15] /usr/share/ofono/scripts/set-tech-preference /ril_[0|1] [any|gsm|umts|lte]; [14:15] also have set-3g-slot ril_n [14:15] nifty [14:15] abeato++ [14:16] Chipaca: ah, better. though i don't know if it answers my question. how does the app server know what user to send the notification to? the push server doesn't know about the app's oauth, and the app's oauth doesn't know about the user's u1 token, right? [14:17] dobey: the application calls Register, gets a token, gives it to the appserver [14:17] dobey: the appserver uses the token to talk to the push server [14:17] dobey: with the token, we know the device, user, and application [14:17] dobey: \o/ [14:18] ok, so there's a third token that both know about [14:18] dobey: third? [14:18] dobey: are you counting the oauth credentials as two? :) [14:19] Chipaca: oh i thought the u1 credentials were involved with talking to the push server from the client side [14:19] dobey: yes [14:19] dobey: but the push server doesn't know the app credentials [14:19] dobey: and the app doesn't know the u1 credentials [14:20] dobey: so there's nothing in the dance that knows three credentials [14:20] Chipaca: but the push client knows the u1 credentials, right? [14:20] dobey: yes [14:20] Chipaca: so third token was correct. because you can't have push notifications without a u1 account :) [14:21] dobey: again, i count two tokens [14:21] dobey: anyway, it's probably academic [14:21] you know how it works, more or less, now :) [14:21] \o/ [14:22] Chipaca: u1 token that push client uses, app calling Register gets a new separate token, and then the oauth token for the third party service, right? [14:22] dobey: yes, there are three tokens involved, but nothing knows the three of them [14:23] Chipaca: sure, i'm counting all the things invovled, not what one individual piece might know about [14:23] ok. [14:23] i wanted to understand the architecture, not what just an app does :) [14:23] i thought by “a third token that both know about” you meant that there were three tokens that everybody knew [14:24] oh, no. i meant it like a pyramid scheme :) [14:24] Amway! [14:26] dobey: and then there's the shared email credentials for the official email client [14:26] Chipaca: there is no official email client [14:26] Chipaca: and no, the one that exists does not yet support gpg or x509 [14:27] ;) [14:27] dobey: sorry, i meant "official" emai-client [14:27] i'm going to write an app and call it that [14:27] right popey? :P [14:27] and it's going to have blackjack, and places to hang your hat [14:28] or was it strange pipes [14:28] anyway [14:28] i saw what was done [14:28] Now listen here you trolls! [14:28] Chipaca: https://www.youtube.com/watch?v=BGi6Q1pNbS0 === MacSlow|lunch is now known as MacSlow [14:52] jgdx, ok, i've reproduced this with the script [14:53] jgdx, i changed it to gsm and it took a few minutes to connect [14:53] then i changed it back to umts and now it's been nearly 10 minutes and still doesn't show as connected [14:53] nmcli shows it as disconnected === marcusto_ is now known as marcustomlinson_ [14:54] not even trying to connect [14:54] Wellark_, ^^^ fyi [14:54] i'm going to drop the ppa now and try again === marcustomlinson_ is now known as marcustomlinson [14:55] <__paulo> Hello everyone. I downloaded and installed qt from qt website but qtcreator crashes on Kubuntu 15.05. Can anyone help me understand why? http://pastebin.com/ESGKXjWC [15:00] jgdx, Wellark_: i'm getting the same thing without the silo enabled [15:00] must be NM or ofono [15:00] maybe ofono... nmcli shows the device as disconnected [15:01] if it was NM i'd think maybe it would show as trying to connect [15:03] although, list-modems shows that ofono thinks it is connected fine [15:03] so must be NM [15:03] jgdx, so you can't reproduce this on krillin? [15:04] jgdx, at least it means it isn't a problem in your branch :) [15:04] but... it also means i can't really test it [15:18] Wellark_, can you using set-tech-preference on krillin/vivid-proposed? [15:18] s/using/try using/ [15:26] dholbach, did you see my note re the sdk install instructions [15:26] ah just saw the email [15:27] dholbach, can you also fix https://bugs.launchpad.net/developer-ubuntu-com/+bug/1418435 [15:27] Launchpad bug 1418435 in Ubuntu App Developer site "installing-ubuntu-for-devices no longer needs to enable universe" [Undecided,New] [15:28] kenvandine, back [15:28] kenvandine, I can't say I've seen that, no [15:28] kenvandine, vidid…? [15:29] jgdx, yeah [15:29] sometimes it takes over 10 minutes to connect [15:29] nmcli shows it as disconnected the whole time [15:29] kenvandine, really. Could be the SIM though, going to test to confirm [15:29] but ofono looks fine [15:29] yeah... could be [15:30] and it's not the ipv6 issue [15:30] but... this doesn't happen with the same sim on rtm [15:30] nope [15:30] it does eventually connect and work [15:30] same sim and same device on rtm [15:30] that basically means it's broken on vivid [15:30] yeah [15:30] it being nm, ofono something [15:30] Wellark_ said vivid has a newer NM and there are some known bugs [15:31] but not sure if it's related [15:31] it's almost five but I'll put some coffee on [15:32] :) [15:32] jgdx, at least it worked fine on rtm on tuesday [15:32] i haven't had that sim in the krillin since then [15:35] Hello! I have an error while trying to run apps from QtCreator to my tablet [15:36] The error is: :-1: error: Could not connect to host: SSH Protocol error: Server and client capabilities don't match. Client list was: aes128-cbc,3des-cbc.Server list was aes128-ctr,aes192-ctr,aes256-ctr,aes128-gcm@openssh.com,aes256-gcm@openssh.com,chacha20-poly1305@openssh.com.Is the device connected and set up for network access? [15:36] boghison, I've had that. You need the sdk from PPA [15:37] boghison, see http://askubuntu.com/questions/561694/ubuntu-ui-toolkits-interaction-with-my-device-fails-due-to-ssh-protocol-capabil [15:37] boghison, better instructions https://launchpad.net/~ubuntu-sdk-team/+archive/ubuntu/ppa [15:37] jgdx: I do have the SDK, obviously [15:37] boghison, but perhaps not installed from a ppa [15:39] jgdx: I am pretty sure I did install it from the PPA (as per the guide). How can I check? [15:39] mardy, hey, so we've been talking about the trust session reversal here with Thomas, and while we know where we need to get [15:40] mardy, for you we actually have a solution today... ${MIR_SOCKET}_trusted is always there, no need to open trusted sessions where you know there's not going to be UI [15:40] and that will not cause any splashes to appear [15:41] boghison, grep ^ /etc/apt/sources.list /etc/apt/sources.list.d/*|grep sdk-team [15:45] pmcgowan, yep [15:46] jgdx /etc/apt/sources.list.d/ubuntu-sdk-team-ppa-trusty.list:deb http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu utopic main # disabled on upgrade to utopic [15:46] oops, didn't paste full [15:46] in a moment [15:46] boghison, wait [15:46] jgdx http://pastebin.com/C9e92AQf [15:46] boghison, use a pastebin if its > n lines [15:46] :) [15:47] jgdx it's not my first time on IRC :) [15:47] boghison, :P. Are you on trusty or utopic? [15:48] jgdx uptopic [15:48] utopic* [15:48] and the device is on vivid [15:48] looks fine to me.. might have to upgrade after an apt-get update [15:48] ok, looks like qtcreator is in the upgrade list [15:49] right, that should do it [15:50] jgdx, so when monitor-ofono shows it has registered and attached = true [15:50] NM should go online right? [15:51] depends on the mood of nm I think [15:51] nm is in a bad mood [15:51] i'm watching monitor-ofono while waiting for it to connect [15:52] jgdx, it got to Attached = True quickly [15:52] and now just strength changes [15:52] but disconnected [15:53] anything in syslog from NM? [15:53] i'm looking to see if ofono outputs anything when NM decides to connect [15:53] jgdx my devices tab isn't loading [15:54] kenvandine, I'm seeing http://pastebin.ubuntu.com/10075103/ [15:54] nm, that was my own fault [15:54] {ConnectionContext} [/ril_0/context1] Active = True [15:54] that was just over 10 minutes after i changed tech pref === slangase` is now known as slangasek [15:55] * jgdx tries [15:55] oh... whoopsie... [15:56] looks like whoopsie was trying to report a crash [15:56] when it timed out, because it couldn't connect to daisy.ubuntu.com [15:57] then NM started activating the context [15:57] kenvandine, I'm seeing the same thing [15:57] but this has to be very new [15:57] also got an operation not permitted, but it never the less pulled down the context [15:57] i have crash files for ofonod and indicator-network [15:58] jgdx, oh... so you reproduced it! [15:58] great [15:58] hold on [15:58] need to see if it reactivates [15:59] mine activated the context right after whoopsie timed out [15:59] whoopsie timing out makes sense, since i had no data [15:59] i bet this wouldn't happen if wifi was enabled [16:00] well, the crash might happen :) [16:00] * kenvandine reboots [16:01] but it might connect quicker [16:05] kenvandine, so that we're on the same page: 1. Reboot 2. Confirm cellular data 3. Change tech preference on online modem 4. Confirm cellular data gone (for 10 mins?) [16:05] with this syslog http://pastebin.ubuntu.com/10075286/ [16:06] interesting... no crash files but still whoopsie restarting [16:06] jgdx, yes [16:06] Feb 5 17:05:14 ubuntu-phablet kernel: [ 224.126977]init: wait-for-state (whoopsienetwork-manager) main process (5220) terminated with status 100 [16:06] what's with that stuff? [16:07] i moved my crash files out of the way [16:07] and this time no new ones were created [16:07] but still not activating the context [16:08] kenvandine, what's your modem path? [16:08] /ril_0 [16:08] Feb 5 16:05:00 ubuntu-phablet NetworkManager[1359]: (/ril_0): could not activate context, modem is not registered. [16:09] http://pastebin.ubuntu.com/10075310/ [16:10] jgdx, that's whats been logged since i changed it [16:11] ogra_: I made udev file and it didn't changed anything for now, have you maybe other idea? :) [16:12] pmcgowan, the universe question could be relevant for the sdk though, right? [16:13] dholbach, how so? we get everything from main archive or ppa afaik [16:13] dholbach, could check with bzoltan [16:13] pmcgowan, how about the dependencies? [16:14] dholbach, I "assumed" not but good question to confirm [16:14] right [16:14] I'll doublecheck [16:14] thanks === marcusto_ is now known as marcustomlinson_ [16:22] dholbach: what is the question? [16:23] john-mcaleely, pmcgowan, there's no ubuntu-device-flash for vivid in the ppa, nor a ubuntu-sdk [16:23] rvr: any news for me? [16:23] let me try it for utopic [16:23] bzoltan, it's all right - I'm just looking at the install instructions again [16:23] dholbach, the ubuntu-sdk package comes from main ? [16:24] Chipaca: I'm now on it [16:24] thats fine, but we should have all the tools in the ppa for vivid [16:24] to unify the instructions [16:24] pmcgowan, I'm afraid not: [16:24] daniel@daydream:~$ apt-cache showsrc ubuntu-sdk | grep ^Dir [16:24] Directory: pool/universe/u/ubuntu-touch-meta [16:24] daniel@daydream:~$ [16:24] dholbach, so why is that not in main? [16:25] most of the stuff is not in main - I don't know what's the reason behind it [16:25] dholbach, ok then we indeed need to keep the insturctions for universe [16:25] pmcgowan, pool/universe/u/unity-scope-click - etc [16:25] right [16:26] I'll close the bug then [16:26] thanks [16:26] rvr: ta [16:28] cyphermox, i think network-manager=0.9.10.0-4ubuntu5 has broken some stuff, when we change the tech pref it takes ages for NM to decide to reconnect [16:28] cyphermox, yesterday this worked fine, only thing i see since then was network-manager update [16:28] no reason why it would change anything of tech pref [16:29] not that specifically [16:29] but [16:29] it was a pure core/ipv6 change [16:29] but ofono reconnects [16:29] it isn't activating the context [16:29] what I'm saying is you probably just didn't hit that issue before, but it was already there [16:29] we can reliably reproduce it now [16:29] dholbach: pmcgowan: I quickly went down two steps on the ubuntu-sdk dependency tree. 55 packages come from univers and 138 from main [16:29] on mako and krillin in vivid [16:29] ages = ? [16:30] about 10 minutes [16:30] nmcli shows it as disconnected [16:30] should be 5 as per the automatic reconnect timers [16:30] eventually we see NM activate it [16:30] it could be missing signal/taking time to correctly attach when you touch tech though [16:30] cyphermox, to repro https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1418077 [16:30] maybe a little more than 5, as other things might be happening too [16:30] Launchpad bug 1418077 in network-manager (Ubuntu) "NM fails to create a cellular data connection for 10 minutes" [Undecided,New] [16:31] jgdx, oh... you filed this bug yesterday? [16:31] this would need careful watching of ofono's states as well [16:32] kenvandine, the symptoms were the same [16:32] but I could not connect it to such a benign change, as well as being patient enough to wait for it to come back [16:37] what is a reason for not staring ssh service at start up? [16:37] pngo: security? [16:38] popey; desktops start that service at start up [16:38] we don't ship ssh on by default in ubuntu [16:39] got it [16:39] do we even ship the ssh server by default? [16:39] cyphermox, jgdx: interestingly... on my mako i'm only seeing this when switching from umts to gsm, not switching from gsm to umts [16:40] Chipaca: At the third direct message, it says "multiple direct messages" [16:40] rvr: yep :) [16:40] Nice [16:41] cyphermox, so perhaps this bug has been there longer, yesterday when i tested this i switched from gsm to umts and it worked perfectly [16:41] so i wouldn't have tickled this bug [16:41] kenvandine, I've seen before that you can go one way without 'losing' cellular data, but not the other [16:41] right [16:42] and... i just confirmed something else... when connected on edge, if i turn off wifi [16:42] it isn't connecting to data [16:42] but that's at the ofono level (or the provider level) whether you lose cell data as you switch technology [16:42] Chipaca: no just ssh client you have to install openssh-server iirc [16:42] cyphermox, yeah, but NM isn't handling that change [16:42] so one thing would be to teach NM to handle whatever other signal ofono send out when this particular thing happens [16:43] so just now, indicator showed edge, i turned on wifi, indicator showed connected to wifi, turned off wifi and the indicator changed to no connection [16:43] and confirmed with nmcli, shows disconnected [16:43] so this is the same thing i noticed last week, no data for a while after leaving wifi [16:44] actually, looking back at monitor-ofono output, it looks like it changed to searching at some point [16:44] maybe while i was on wifi [16:45] but it shows registered and attached now [16:45] but nmcli shows it as disconnected still [16:46] and boom... now NM decided to activate the context [16:46] just under 6 minutes after i turned off wifi [16:46] seems it's the second attempt at which nm suceeds? [16:46] *succeeds [16:48] jgdx, ok... i wonder if this has anything to do with testing the apn editor :) [16:49] i do see if fail to connect to context2 before it connects to context1 [16:49] but... that doesn't get logged until right before it activates context1 [16:49] so long after it disconnects [16:49] kenvandine, I confirmed this with the ofono provisioned context, though [16:50] so only that single context? [16:50] yeah [16:50] no custom context created? [16:50] oh... yeah, on my krillin i only have that one [16:50] Noskcaj, I did remove-contexts at some point [16:50] so yeah, rules that out [16:50] Noskcaj, tab fail, sorry. [16:52] cyphermox, mind if i assign bug 1418077 to you? [16:52] bug 1418077 in network-manager (Ubuntu) "NM fails to create a cellular data connection for a unacceptably long time" [Undecided,Confirmed] https://launchpad.net/bugs/1418077 [16:54] kenvandine: I'm not sure what time I have to give it; how rush is it? [16:54] i think it's probably critical [16:54] i'd hate for this to end up in rtm [16:55] especially if it's like it's defined here; seems like there is likely not much I can do about it [16:55] losing data when going off wifi is pretty bad [16:55] ie. ofono needs to say that cellular data is ready (ie. Attached and all), and from there there are delays in NM before autoconnection, which affect everything, not just modem [16:56] cyphermox, but this isn't happening in rtm [16:56] but you're not just going off wifi there -- that would work [16:56] so something's different [16:56] you're also changing the technology [16:56] cyphermox, i just reproduced it by turning off wifi without changing the tech [16:56] last week i was noticing that sometimes when i left the house [16:56] i didn't have data for a while [16:56] was cellular really connected at that time? [16:57] just now when i did it, i think it had actually go to searching [16:57] but it registered and attached long before NM activated the context [16:57] I'm not sure changing the 5 min delay in NM is wise before very careful testing on all arches/all use cases, not just the phone [16:57] and it's the only thing I can think of that is likely to help there [16:57] i think the real question is why isn't this happening on rtm? [16:58] there are obviously some differences between 0.9.8 and 0.9.10 [16:58] on rtm you have 0.9.8 [16:58] yeah... any chance 0.9.10 is going to end up in rtm soon? [16:58] but the underlying logic is largely the same, the 5 minutes delay is there in both [16:58] no [16:58] no chance at all [16:58] but still, we can't say we don't care about vivid [16:58] it explicitly doesn't have to go in RTM [16:59] indeed [16:59] Chipaca: Silo 7 approved. [16:59] * Chipaca frames that [16:59] cyphermox, so maybe not critical, but high at least [16:59] kenvandine: tbh I'd really love if others could start looking at NM too; that way I wouldn't always have to be the one to deal with these bugs [16:59] not saying that it should be you though ;) [16:59] definately not me :) [17:00] but sure [17:00] you're our go to guy for NM though :) [17:00] cyphermox, any suggestions? Wellark_? [17:00] slangasek: there's bug 1418077 to deal with ^, this is one case where it's likely to take some investigation time, like a day or two. [17:00] bug 1418077 in network-manager (Ubuntu) "NM fails to create a cellular data connection for a unacceptably long time" [Undecided,Confirmed] https://launchpad.net/bugs/1418077 [17:01] kenvandine: Wellark knows about mbpi already, not sure whether he has time for this though -- that's part of the problem, I have no idea what other's schedules are :) [17:01] cyphermox, one thing that is interesting there, you said NM waits 5 minutes [17:01] yes [17:01] this has taken 10 minutes for me [17:01] cyphermox, kenvandine, we just discussed this in our standup earlier [17:01] not always... [17:01] that's been the case for years [17:01] but always more than 5m [17:02] our team will pick this up as a focus next sprint [17:02] awe_, so can you get it assigned? [17:02] the ten minutes is worrying, but it could be explained by how things went by for ofono to pick up the data conenction to Attached. [17:02] kenvandine: I'll at least triage the bug now and ask for the information that would be needed to debug this [17:02] cyphermox, the 10 minutes was from the time monitor-ofono said it was attached [17:03] kenvandine, I mentioned this very problem to cyphermox this morning. I have to land ofono later today, but will try and look at it tomorrow [17:03] kenvandine, I can't promise I'll be allowed to self-assign, but I will look at it [17:03] so clearly others have reproduced this already :) [17:03] good [17:03] as I've been seeing similar issues [17:03] it's very easy to reproduce [17:03] kenvandine, ack. It's something that needs to be fixed for OTA1 [17:03] on both krillin and mako [17:04] kenvandine, RTM, Vivid, or both? [17:04] just vivid [17:04] right, that's consistent with what I see [17:04] the delay when disconnecting from wifi is aweful [17:04] i happens to me when i walk down the street [17:04] again, all I can promise right now is that we'll put some resources on it, and at least try and root cause the problem [17:05] kenvandine, that said... it *is* our devel release [17:05] great... [17:05] jgdx, ^^ [17:05] indeed [17:05] that's why i suggested high, not critical [17:05] wee [17:05] but again, it's on my teams' short-term radar [17:05] good, i really need to stop thinking about it so i can move on to content hub :) [17:05] kenvandine, I have to run... gonna be late for my lesson; bbl [17:05] jgdx, get that silo prepared for rtm :) [17:05] awe_, thanks! [17:05] np! [17:06] awe_: triaged it [17:06] awe_: good to know it's on the radar then [17:07] jgdx, can you get the logs cyphermox asked for? [17:14] kenvandine, sure [17:24] have someone had a luck to use syncevolution to sync calendar or contacts? I tried https://syncevolution.org/wiki/synchronizing-evolution-http-howto. this howto did not work for me on desktop as well. [17:25] I wrote simple scripts that sync my phone and computer but looks like this would be better way. [17:27] pngo, we already use syncevolution for syncing contacts and calendar events on the phone [17:27] for google accounts anyway [17:33] kenvandine: yes, but its use is currently restriced to that i think. syncing between phone and evolution on a PC are not doable at the moment through standard interfaces on the phone [17:34] nor is it easy to extend syncevolution to enable other remote sources [17:40] cyphermox, kenvandine, added logs [17:40] jgdx, thx === chihchun is now known as chihchun_afk [17:51] Can someone help with: init: ureadahead-touch main process (1447) terminated with status 5 [17:52] I checked ureadahead-touch-log: Counted 0 CPUs ; ureadahead: Error while tracing: No such file or directory [17:53] How can i fix/debug it? Any idea [17:57] (Its port) [18:11] mandel, any progress on the check hash branch you have? [21:09] kenvandine, that tech switch test failure seems to be local to ci. [21:10] passes 100% of the time both isolated and not [21:22] jgdx, thx [21:23] forgot to say that it was tested on mako [21:41] Danchapman: is the search button supposed to search emails in the new Dekko? [21:42] seems like email is a big topic on #ubuntu-touch ;p [21:42] for me when i look say british airways it just re-order my emails chronologically [21:42] Danchapman ^^ [21:43] haha k1l :) [21:59] taiebot atm it only supports searching sender and subject. More complex searching will come in the next week or so [22:03] Danchapman: Progress are amazing. it is very nice :) will it be the official mail app for UT ? [22:15] Hi, I sometimes have an issue with the url-dispatcher where it launches the incorrect app, eg at the moment album:/// is launching the gmail webapp?! (in this case it seems to be looping as well as i cannot move away from the gmail app) ... I suspect this is due to me constantly installing different versions of the music-app, but I was wondering if there are any log files that I should look at to figure out what is going on? [22:16] You think url-dispatcher is launching gmail for albums because you installed the Music player? [22:16] no it just randomly breaks...sometimes it launches calendar or facebook ... when i click "play with music-app" [22:17] but note that i push various different versions of the music-app across to my device daily [22:17] ooo dbus-monitor is going mental... [22:18] nhaines, you can clearly see it trying to launch gmail when it is calling album:/// http://pastebin.ubuntu.com/10080370/ [22:19] but this case seems worse than normal as it is looping ... normally it just does it once [22:33] Laney, hey, could you take a look at [1]? Fixes something that has annoyed me. [1] https://code.launchpad.net/~jonas-drange/ubuntu-system-settings/cellular-insertsim/+merge/248835