=== JanC is now known as Guest76305 === JanC_ is now known as JanC === DeadSquirrel01 is now known as DeadSquirrel0 === DeadSquirrel0 is now known as DeadSquirrel01 === DeadSquirrel01 is now known as Ds01_ [02:41] hey anyone here running ubuntu on surface pro 3 and does it run well (with toutch etc)? === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [09:18] hello [09:18] can we use sougou at ubuntu phone? [09:18] i search all the information by google,but no answers [11:20] Sil2100 Canonical Sytem Image for OTA 14 still ´Expected: 2016-11-14´. that could do with an update I think [11:22] sil2100 if a new realistic date can be set. [11:23] matv1: yeah, let me poke Pat about it [11:23] Right now we're still struggling as oxide is not releasable [11:24] yes I know. And on a personal note I would say https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1478319 [11:24] Ubuntu bug 1478319 in network-manager (Ubuntu) "Wifi doesn't reconnect to AP if it uses the same SSID than another AP" [Undecided,Confirmed] [11:24] could really do with a fix [11:25] which is also on the OTA 14 list [11:27] sorry i meant to say #1598584 [11:28] https://bugs.launchpad.net/ubuntu/+source/indicator-network/+bug/1598584 [11:28] Ubuntu bug 1598584 in Canonical System Image "wifi slows to a crawl when bluetooth is enabled" [Critical,In progress] === _salem is now known as salem_ [11:59] Just tried to install ubuntu touch on my oneplus one. all went well until where ubuntu-device-flash says: Rebooting into recovery to flash. [12:00] The device rebooted, but is stuck in the boot logo (where it says powered by android) for like 10 minutes? [12:00] I'm afraid I might have fried my bacon... :-D [12:03] any ideas what I can/should do now? === marcusto_ is now known as marcustomlinson === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [14:06] bfiller, heyo -- for bug 1629009 both messaging-app and dialer-app are marked as fix released or fix committed. But I can't find the branches for them or see them fixed. Do you know where the MPs are for those fixes so that I can include them in the silo? [14:06] bug 1629009 in unity-scopes-api (Ubuntu) "Does not work inside a snap due to hardcoded paths" [Undecided,In progress] https://launchpad.net/bugs/1629009 [14:10] mterry, let me check, I know the snaps are in the store [14:10] possible the code never got merged [14:10] renato__, ^^^^ [14:11] mterry, actually wanted to ask you about that - I noticed they were preisntalled in the unity8 snap, but shouldn't we be using the snaps instead? [14:11] mterry, same with browser [14:12] mterry, yes there is snap packages for these apps on store. But probably this bug does not appear on desktop, due the lack of modem. But I have plans to test it with boiko helps [14:13] renato__, where are the MR's? and did they land in trunk? [14:13] bfiller, yes they landed already [14:13] bfiller, but not for this bug specific. Was more related with packaging stuff [14:14] bfiller: fair question about snaps/debs. I think we were using debs as bootstrapping. And I'm leery to remove them until we can install snaps from the store and launch them :) [14:14] renato__: yeah then I don't think this bug is fixed [14:15] Although if the snaps work, maybe it won't matter when we switch to using those... [14:17] But at least that bugs tracking status should be correct and the two packages should be set off Fix Committed [14:17] Though I wonder if the snaps do work without such a fix [14:20] mterry, I'll fix that and we'll take a look today to see if we can fix the issues with launching as deb [14:23] bfiller: though I suppose dialer and messenger in particular can be dropped as built-in debs. Likely uninteresting for most desktop users. I think they were just added so that we'd have some apps to open [14:24] mterry, right, dialer and messaging aren't too interesting on desktop until we add SIP and other messaging (non-sms) support [14:31] kgunn: any objection to dropping dialer-app and messaging-app from the u8 snap? seem like uninteresting apps for desktop [14:36] mterry: yeah, i think that's fine.... is webbrowser already in ? [14:37] if not we should work to add that [14:37] bfiller: ^ fyi [14:37] kgunn: yes and no. It's normally in, but I think we hit a bug in snapcraft the last silo snap build that excluded it. Working to reproduce and file if so [14:38] kgunn, mterry : once we can launch snaps, the browser is snapped and in the store [14:38] cool, that would make sense ya [14:39] bfiller: nice, so just install from terminal [14:39] kgunn: I'm guessing we'd want to continue keeping USS and u8 together? They are tightly coupled [14:39] or could install from store scope [14:39] mterry: yeah uss should def stay in u8 [14:39] mterry, yes for sure about USS and u8, was always the plan [14:39] imho [14:39] +1 [14:39] alright, will just drop the dialer and messenger apps for now [14:40] pmcgowan: ^ just fyi [14:40] and keep an eye on being able to drop the browser [14:40] pmcgowan: i guess that's a question, not sure what the expectations are about browser and u8-snap...should it always "be in the u8 snap" [14:40] kgunn, sounds fine [14:40] or ok to remove it, and have user install it [14:41] heh, looks like we're installing a few apps I didn't realize -- becuase they don't show up in a scope. Wonder why [14:41] Contacts and Media [14:41] kgunn, I would put it in for now? [14:42] mterry, media probably has a no show in the desktop file, but contacts should show [14:42] pmcgowan: ack === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [16:51] tedg: in the u8 snap, address-book-app doesn't show up in App scopes because its Exec= line is an absolute path. Is that UAL's fault or the click scope's fault? [16:52] mterry: I think that's UAL's fault, it should probably prepend the $SNAP in that case... [16:52] Can't think of a reason you'd want to call a binary outside of your snap [16:52] tedg: OK. It does for the icon right? Makes sense to do so for Exec [16:52] I think /bin/sh maybe? [16:53] tedg: yeah /bin/sh is use case [16:53] It's the only one I can think of though, even stuff like Python would be worst case bind mounted into your snap. [16:53] tedg: we could also make address-book-app not use absolute path... But your call [16:54] tedg: you could check if prepended path is valid, else leave it alone [16:54] Well, we should probably do both. People are likely to crib from our snaps, but I think that the apps we've got are a good example of bugs we should fix :-) [16:54] mterry: The problem is that "valid" is so hard in to calculate the mount container that snaps are put into :-/ === dandrader is now known as dandrader|afk [16:55] mterry: We'd have to look in their base snap, because it could be different than the "/" for us. [16:56] tedg: I'm just looking at legacy case [16:56] mterry: Hmm, I think I might be changing my own mind... we can't have apps with / in their desktop file because there's no way to resolve that. [16:57] mterry: Ah, I see. [16:57] mterry: Legacy in the U8 snap. [16:57] right [16:57] So really just for system settings long term ;-) [16:58] Yeah, let's just fix that then. We can detect it easily enough. === salem_ is now known as _salem === _salem is now known as salem_ [17:11] renato__: you also have address-book-app marked fix released for bug 1629009, but it still has problems looking up Unity.InputInfo due to a non-prefixed UNITY8_QML_PATH [17:11] bug 1629009 in messaging-app (Ubuntu) "Does not work inside a snap due to hardcoded paths" [High,In progress] https://launchpad.net/bugs/1629009 [17:11] er, fix committed [17:13] renato__: also, its desktop file uses an absolute path for its Exec when maybe it shouldn't. We can work around that in UAL but might be easier to just use "address-book-app" instead of full path [17:15] mterry, this is how it work in the past. if you have another solution now I can update it [17:15] renato__: for the desktop file? I mean my suggestion is just switching "/usr/bin/address-book-app" for "address-book-app" [17:15] renato__: for UNITY8_QML_PATH, needs actual code patch [17:16] mterry, ok for dekstop I can update [17:16] mterry, all ours apps uses full path. because this is the way that was working in the past [17:16] mterry, and about the icon? [17:17] renato__: we have code for icon to prefix correctly, I believe [17:17] just harder to do it for path [17:17] Exec path [17:17] When it might be /bin/sh or whatever [17:17] mterry, our icons uses full path too [17:17] But I think tedg is working on something there anyway [17:17] renato__: right. UAL prefixes $SNAP for you there [17:18] mterry, do you prefer ted to finish the work. Than we can test? Because this was changed in the past already [17:19] renato__: most apps I can see (USS, terminal, dialer, messager, etc) use non-absolute Exec paths [17:19] mterry, I remember that the store reject the app if the path was wrong [17:19] mterry, ok I will try [17:19] renato__: so I don't know what the state of art is. But you can either fix the desktop or not. Just telling you that it doesn't show up in app scope right now [17:20] renato__: but having it show up isn't much good if you don't also fix the hardcoded unity8 path [17:20] mterry, yes probably because something has changed. In the past was working nice [17:20] renato__: working nice in the unity8 snap? I'm sure it all works fine in normal contexts (phone, desktop) [17:20] renato__: but u8 snap is new, I don't think it ever worked well in there [17:20] mterry, unity debian [17:21] renato__: right. That's fine. I'm not talking about that though. Specifically the unity8-session snap we've been trying to get working [17:21] That's what bug 1629009 is about for example [17:21] bug 1629009 in messaging-app (Ubuntu) "Does not work inside a snap due to hardcoded paths" [High,In progress] https://launchpad.net/bugs/1629009 [17:23] mterry, ok, we tested this against unity8 debian. [17:23] mterry, I will update the apps [17:24] mterry, probably I app that I fixed has hardcoded paths on desktop file [18:32] tedg: https://code.launchpad.net/~mterry/policykit-unity8/snap-root/+merge/311057 === dandrader|afk is now known as dandrader === JanC is now known as Guest70154 === JanC_ is now known as JanC [18:47] mterry: Does that work? Are we putting policykit in the snap itself? [18:47] mterry: Seems like it'd have to be in core? [18:47] (or something else non-unity8) [19:15] tedg: haven't tested end-to-end yet. Just realized that the snapd login service is policykit-protected. And u8 needs that package to talk to polkit [19:16] tedg: there are several other services we rely on that need policykit, we usually just special-cased phablet in Touch [19:16] mterry: K, thinking that we'll probably need another way to install that conf file. [19:16] mterry: like it can't live in our snap [19:16] tedg: that MP is unity8 specific, why wouldn't it live in our snap? [19:16] tedg: policykit itself, sure, that can be outside [19:16] mterry: The agent would be in our snap, but the conf file needs to install outside. [19:17] mterry: To wherever the outside policykit wants them. [19:17] tedg: which conf file? [19:17] the one in the MP is for upstart, which is inside the snap [19:18] mterry: Oh, wait, you're right. Sorry I was confused. [19:18] mterry: I thought we had to install one in /usr [19:18] mterry: Makes sense, +1 :-) [19:19] tedg: any objection to me pushing it through a silo by itself to xenial+zesy? [19:19] mterry: nope [19:19] cool thx [19:48] mterry, tedg: is it known/expected that I can't run a snap from the terminal in unity8 snap session? I can install the snap but don't see it on the path [19:49] bfiller: interesting... I haven't tried that yet, but I would have guessed /snap/bin would be on the path [19:50] bfiller: in general running from the terminal won't work anyway, you'll need to use ubuntu-app-launch to start apps. [19:52] tedg: but for non-graphical apps... [19:52] It does seem like /snap/bin isn't in the PATH. But should be, I'd wager [19:52] I guess, but you can't run anything from it anyway. [19:52] It is the same snap-on-snap problem. [19:53] tedg, I guess when we can launch them from scope it won't be an issue [19:53] tedg, just trying to figure out a way to test snaps until we fix the launching from scope [19:55] I think I might have just cracked the nut on that one... so hopefully soon. [19:56] mterry, about the desktop files. Which one are you using on unity8 snap? That one that is shiped inside of setup/gui/ or the one that is installed with the app? [19:56] Though, honestly, systemd is like being handed a bowl of nuts. [19:56] renato__: setup/gui/appname.desktop [19:56] renato__, are you trying to figure out why the icon is not showing up? [19:57] bfiller, no I am trying to understand which file mterry asked me to remove the hardcoded paths [19:57] the file inside setup/gui/ is not hardcoded [19:58] renato__: I was using the one installed with the app in /usr/share/applications (right now, unity8 snap bundles the address-book-app deb inside of it) [19:58] Ah, sorry. I was confused on the question. My answer was for other snaps. [19:59] mterry, sorry I do not understand. Address-book-app that is on store, contains both desktop files [19:59] renato__, he's using the deb [19:59] from overlay [19:59] yup [19:59] not the snap [20:08] mterry: Okay, rebuilding UAL in the silo. Do you have a snap rebuild planned? [20:09] tedg: not planned no [20:09] mterry: K, I'll kick one off once this finishes [20:09] mterry, are you seeing address book app show up in scope? (I'm not). Or is that the issue you are trying to get resolved with renato [20:09] tedg: oh I did want to build online-accounts [20:09] tedg: but that can happen before or after snap build [20:10] bfiller: yes. Additionally, once it does show up, it doesn't display anything due to hardcoded paths to unity8's qml modules [20:10] ack [20:10] Which I don't know how you solve when the app is its own independent snap... [20:10] But at least in this use case you can solve === boiko_ is now known as boiko [20:56] jdstrand: I'm getting this from media-hub while trying to play a song from the dash in the in-progress unity8 snap: "Error while executing the underlying io_service: apparmor::ubuntu::Context: Invalid profile name snap.unity8-session.unity8-session" -- what would a valid profile name look like? [21:06] mterry: that is a valid profile name for snappy. I suspect that the checks it has need to be adjusted from the click world (__) to the snap world (snap..) [21:07] jdstrand: oh huh it's worse than that -- media-hub has some particular apparmor integration and is expecting unity to represent as "unity8-dash" [21:07] jdstrand: ok so not anything to do with you, I jumped the gun :) thx [21:08] np [21:09] mterry: note that jhodapp_ is aware of some libapparmor changes that could cleanup this code. I don't know the status of that work but before you go and fix it, you might want to sync with him [21:10] k, though I think this is just internal to media-hub [21:10] mterry: yes, that is what I mean. media-hub can be cleaned up to use the libapparmor more effectively [21:10] mterry, yeah there's some work to be done there [21:10] jdstrand: ah got ya [21:11] there were some changes to the libapparmor api that came in after media-hub implemented those checks [21:12] jhodapp_: is there an existing bug for media-hub? I'm looking for something to track [21:12] I can make one [21:12] mterry, hang on, let me finish the meeting I'm in and I'll search quickly [21:12] jhodapp_: cool no rush [21:12] mterry, there might be one but it might not longer describe what you need [21:42] mterry, ok I think this bug will cover what you're asking for...I was trying to find an email that jdstrand sent me explaining the situation better but can't find it...https://bugs.launchpad.net/ubuntu/+source/media-hub/+bug/1489489 [21:42] Ubuntu bug 1489489 in apparmor (Ubuntu) "The org.freedesktop.DBus.GetConnectionAppArmorSecurityContext() method is deprecated" [Medium,In progress] [21:44] jhodapp_: I don't think that's the right bug [21:44] jhodapp_: https://trello.com/c/ftSp1Ogp/649-implement-aa-query-file-fd-to-fix-toctou-issues-with-using-aa-query-file-path [21:44] jdstrand, thanks! forgot where you had added that info [21:45] jhodapp_: hmm, I'm not sure if that bug will fix my issue (maybe during the cleanup, the offending code will be removed...) [21:45] jhodapp_: thanks for pointer [21:45] mterry, ok if you wouldn't mind filing a new bug referring to that trello card that should work [21:46] jhodapp_: ok will do. tomorrow though, gotta sign off soon [21:46] mterry, same here, cheers [22:09] mterry, hey I am getting this error while trying to snapcraft unity8: E:Unable to correct problems, you have held broken packages. [22:09] mterry, any I idea what could cause that? === salem_ is now known as _salem === yorick_ is now known as yorick