| traccart | um.. last update removed unity8-desktop-session-mir package | 08:54 |
|---|---|---|
| traccart | ^^ 16.04 | 09:04 |
| duflu | traccart: What happens if you try to put it back? sudo apt-get install unity8-desktop-session-mir | 09:07 |
| tsdgeos | traccart: xenial? | 09:12 |
| tsdgeos | yeah the current status is a bit borked | 09:12 |
| tsdgeos | the new toolkit hasn't landed yet | 09:13 |
| tsdgeos | so stuff is a bit "on the edge" | 09:13 |
| tsdgeos | apt wants me to remove unity8 too :D | 09:14 |
| duflu | Such is the dependency graph | 09:14 |
| duflu | Gets out of hand some times | 09:14 |
| tsdgeos | yep | 09:15 |
| duflu | Sadly lots of large package dependency graphs are probably avoidable. One or two removals from them can make a dramatic difference. So developers really should treat package dependencies with more caution | 09:21 |
| traccart | duflu, tsdgeos yep 16.04 if i add unity8 session it removes something else :)) | 09:48 |
| traccart | either way i get a broken unity8 | 09:49 |
| Saviq | traccart, yes, you'll need to wait some time for this to get resolved or get a package or two from proposed | 09:49 |
| traccart | i'll wait :D | 09:50 |
| traccart | for snappy muhahaha | 09:50 |
| Saviq | traccart, https://requests.ci-train.ubuntu.com/#/ticket/771 this is where the underlying cause is being handled | 09:50 |
| traccart | i see, thanks Saviq | 09:51 |
| tsdgeos | cimi: the bug in vivid-overlay preventing the filters stuff seems to have landed, could you give it a final review? | 10:17 |
| tsdgeos | cimi: it just landed, so you'll need a dist-upgrade, is not in the image yet | 10:17 |
| cimi | tsdgeos, ok | 10:19 |
| === greyback__ is now known as greyback | ||
| cimi | tsdgeos, can I use silo 54? | 10:38 |
| tsdgeos | cimi: i think it should be fine, anything wrong? | 10:39 |
| tsdgeos | pstolowski: ↑↑ | 10:39 |
| pstolowski | cimi, sure you can. no promise though about it, we were not able to test end-to-end yet | 10:41 |
| maikeul | hi | 12:13 |
| maikeul | i have issue running Xmir on an ubuntu phone, the mir connection is rejected by the server and I just cannot find a log or indications on how to allow this. | 12:14 |
| sil2100 | Trevinho: hey! | 12:23 |
| Trevinho | sil2100: hi | 12:23 |
| sil2100 | Trevinho: I was looking at publishing silo 11 for you, but I would like for a small change before we go on that would require a rebuild | 12:24 |
| sil2100 | *like to ask | 12:24 |
| Trevinho | sil2100: ok, sure | 12:24 |
| sil2100 | Trevinho: so, the thing that worried me is this merge - not content wise, but how the train handled it: | 12:25 |
| sil2100 | https://code.launchpad.net/~unity-team/unity/python3-ready/+merge/276746 | 12:25 |
| sil2100 | Trevinho: because you touched the changelog, the train basically did not include the commit message of this change in the end changelog | 12:25 |
| sil2100 | Trevinho: meaning there's an undocumented change now | 12:26 |
| sil2100 | Trevinho: could you take this branch, do a dch -i in it, paste the commit-message as part of the new changelog entry and try to rebuild unity? | 12:26 |
| sil2100 | Trevinho: since in cases where you touch the changelog, you need to explicitly write the commit message for the merge inside the changelog | 12:27 |
| sil2100 | Otherwise the train assumes you did that and doesn't really do it for you | 12:27 |
| Trevinho | sil2100: what's | 12:27 |
| Trevinho | sil2100: sorrry it was a mistake that "what's"... :-D | 12:27 |
| sil2100 | https://ci-train.ubuntu.com/job/ubuntu-landing-011-2-publish/1/artifact/unity_xenial_packaging_changes.diff <- and in the end package you can see that the mention of changing to python3 | 12:28 |
| sil2100 | Is missing | 12:28 |
| sil2100 | (damn, I have issues typing today) | 12:28 |
| sil2100 | ;p | 12:28 |
| sil2100 | Another issue, but that's a nitpick and I guess we can ignore it, is in bamf: | 12:28 |
| greyback | Mirv: hey, I'm getting a qtcreator crash on startup on vivid+overlay, here's a backtrace: http://pastebin.ubuntu.com/13927939/ | 12:28 |
| sil2100 | Trevinho: https://ci-train.ubuntu.com/job/ubuntu-landing-011-2-publish/1/artifact/bamf_xenial_packaging_changes.diff <- the train suddenly decided to change the 0replaceme in the changelog to a version number ;p ;p | 12:29 |
| Trevinho | oh, nice :-D... | 12:29 |
| Trevinho | mh, I guess I can fix both | 12:30 |
| sil2100 | Trevinho: would be nice to fix that but not required, it's just a nitpick | 12:30 |
| sil2100 | I'm more concerned with the unity one :) | 12:30 |
| sil2100 | Trevinho: just remember that to fix it in bamf, you need to modify the changelog and remove the 0replaceme (or rename to something like 0-replaceme etc.) and then add the commit message to the changelog manually | 12:31 |
| sil2100 | Trevinho: but as I said, this is something we can even fix later | 12:31 |
| Trevinho | oh.... well, no ok I'll fix it later then :-D | 12:31 |
| Mirv | greyback: it's the same that kenvandine reported for system-settings. not deeemed critical as it's desktop only, and started with OTA 8.5 hotfix. you can install https://requests.ci-train.ubuntu.com/#/ticket/761 for a fix. | 12:37 |
| greyback | Mirv: perfect, thanks | 12:38 |
| sil2100 | Trevinho: anyway, thanks! | 12:55 |
| Trevinho | sil2100: thank you! | 12:55 |
| Trevinho | sil2100: poackage is building,,.. | 12:55 |
| bregma | maikeul, you will need to add '-mir APPID' to the XMir command line or else the Mir server will reject the connection | 12:59 |
| bregma | APPID is a valid app-id | 12:59 |
| maikeul | thanks, how do you know about APP-Ids ? | 13:02 |
| bregma | maikeul, the app-id generally matches the .desktop file name of an application | 13:03 |
| tsdgeos | cimi: dednick: what do you think of https://code.launchpad.net/~aacid/ubuntu-settings-components/standarizeImports/+merge/280303 ? | 14:03 |
| cimi | tsdgeos, we should indeed now | 14:13 |
| cimi | tsdgeos, however we need to update the APIs if there are ubuntushape and such... | 14:13 |
| tsdgeos | cimi: do we really need to update the APIs? there otld ones also work, no? | 14:16 |
| cimi | tsdgeos, it works even without updating I think, but we should anyway... | 14:17 |
| tsdgeos | cimi: i see tst_ files | 14:17 |
| tsdgeos | but how do i run them? | 14:17 |
| cimi | if there is a button and we bump to uitk 1.3 and such, we can | 14:17 |
| tsdgeos | make check and make test do not do much | 14:17 |
| tsdgeos | s/much/anything | 14:17 |
| tsdgeos | cimi: also if you see i have not chagned any ubuntu.components | 14:25 |
| tsdgeos | just qtquick and ubuntu.compoonets.listitem | 14:25 |
| tsdgeos | so any ubuntushape is not changed | 14:25 |
| cimi | tsdgeos, yeah, I would change it though too | 14:25 |
| cimi | since we use 1.3 in shell | 14:25 |
| cimi | and everywhere else | 14:25 |
| tsdgeos | you would change what? | 14:25 |
| cimi | uitk to 1.3 | 14:25 |
| tsdgeos | it is already there | 14:25 |
| cimi | ah cool | 14:26 |
| tsdgeos | it can not be changed to itself :D | 14:26 |
| cimi | :D | 14:26 |
| cimi | must forgot we did update months ago maybe :) | 14:26 |
| cimi | ok so it's a no brainer for me | 14:26 |
| cimi | tsdgeos, we need the checklist in the description | 14:27 |
| tsdgeos | cimi: which checlist? | 14:27 |
| tsdgeos | unity8's? or there's one for ubuntu-settings-componetnts? | 14:28 |
| cimi | tsdgeos, https://wiki.ubuntu.com/Process/Merges/Checklists/Ubuntu-Settings-Components | 14:28 |
| tsdgeos | k tx | 14:28 |
| tsdgeos | cimi: i only need to test the indicators, right? or there's something else i need to test? | 14:29 |
| cimi | just indicators I think | 14:29 |
| cimi | there is a test file | 14:29 |
| cimi | example | 14:29 |
| cimi | qmlscene something.qml in the tree | 14:30 |
| cimi | or unity8 indicators | 14:30 |
| tsdgeos | k | 14:37 |
| Trevinho | sil2100: I think you can publish that now :-) | 14:47 |
| ChrisTownsend | Hey guys, any guesses when things will be straight in the archive so apt will no longer want to remove unity8 in xenial? | 14:54 |
| tsdgeos | ChrisTownsend: i think https://requests.ci-train.ubuntu.com/#/ticket/771is it | 15:08 |
| tsdgeos | https://requests.ci-train.ubuntu.com/#/ticket/771 | 15:08 |
| ChrisTownsend | tsdgeos: Ah, ok, that's the culprit. Thanks! | 15:10 |
| Saviq | ChrisTownsend, oh, was it you that unity8 locked up for without gstreamer registry? | 15:13 |
| * Saviq vaguely remembers trying to debug that with someone... and it could've been you | 15:14 | |
| ChrisTownsend | Saviq: I think that was me. | 15:14 |
| Saviq | ChrisTownsend, good news: not just you | 15:14 |
| ChrisTownsend | Saviq: lol, good | 15:15 |
| maikeul | bregma: thanks, I still had the issue and i finally found out that the issue was with the default software of the phone. I reflashed it and now everything works fine | 15:16 |
| Saviq | ChrisTownsend, it seems today it's 100% reproducible: delete the registry file, you're stuck | 15:17 |
| Saviq | ChrisTownsend, so that's actually a good thing :) | 15:17 |
| * Saviq in the process of filing bug | 15:17 | |
| ChrisTownsend | Saviq: My question is, why is the registry file being removed???? | 15:18 |
| Saviq | ChrisTownsend, it might never have been there | 15:18 |
| Saviq | ChrisTownsend, or needs updating | 15:18 |
| Saviq | ChrisTownsend, something of the sort | 15:18 |
| bregma | it shouldn't need to be there, it's just a cache | 15:19 |
| ChrisTownsend | Saviq: Hmm, ok. Well, it seems if there is a bug in the Unity8 desktop, I'll find it:) | 15:19 |
| Saviq | ChrisTownsend, good going! | 15:19 |
| bregma | ChrisTownsend is *the* unity 8 desktop usere | 15:19 |
| ChrisTownsend | bregma: The only one, right?:) | 15:19 |
| bregma | I suspect | 15:20 |
| bregma | everyone else just install it, brings it up, says "OK, cool" and moves on | 15:20 |
| bregma | distro-hoppers | 15:20 |
| ChrisTownsend | lol | 15:20 |
| Saviq | ChrisTownsend, bug #1525285 | 15:20 |
| ubot5 | bug 1525285 in qtmultimedia-opensource-src (Ubuntu) "/usr/bin/unity8:6:__GI_ppoll:ppoll:gst_poll_wait:exchange_packets:plugin_loader_free" [Undecided,New] https://launchpad.net/bugs/1525285 | 15:20 |
| ChrisTownsend | Saviq: Thanks, that is indeed the issue I ran into back when. | 15:21 |
| Saviq | ChrisTownsend, do you recall when that was? like 2 weeks ago at least, right? | 15:21 |
| Saviq | ChrisTownsend, and you didn't, by any chance, have the Qt 5.5 packages installed? :) | 15:22 |
| Saviq | was that even xenial? | 15:22 |
| ChrisTownsend | Saviq: Hmm, seemed longer than that. I'll look through my chat logs. | 15:22 |
| Saviq | tx | 15:23 |
| ChrisTownsend | Saviq: Oct. 26 | 15:23 |
| Saviq | ChrisTownsend, ack, almost two months, then | 15:23 |
| ChrisTownsend | Saviq: Yep, and I'm pretty sure I had Wily on that machine then. | 15:24 |
| Saviq | jhodapp, it was ChrisTownsend back then ↑, almost two months, likely with wily still | 15:24 |
| jhodapp | Saviq, ok, I do remember discussing something about a deadlocked Unity8 on startup...seems this is the same thing and person | 15:41 |
| Saviq | jhodapp, might very well be, a bit more detail hopefully | 15:42 |
| Saviq | jhodapp, I wonder if this is specific to a certain plugin | 15:42 |
| * Saviq tries to remove gst-plugins.* packages one by one | 15:42 | |
| jhodapp | Saviq, I believe it was if I remember correctly, but my memory is hazy on this | 15:42 |
| Saviq | or would you say not worth it? | 15:42 |
| Saviq | jhodapp, ok, trying | 15:42 |
| jhodapp | no I think it would be worth trying | 15:42 |
| sil2100 | Trevinho: thanks, let me take a look agian :) | 15:51 |
| maikeul | does this error rings a bell for you : 'error securing buffer...' when running an X11 app (see http://pastie.org/private/n1gdprpgngyaoqhowfdzla ) | 15:51 |
| maikeul | same with -mir or -desktop_file_hint when running Xmir | 15:52 |
| Saviq | jhodapp, down to good and base, still hanging | 15:54 |
| Saviq | not gonna remove those, might not recover ;P | 15:55 |
| Saviq | owait, dpkg -r to the rescue | 15:57 |
| jhodapp | :) | 15:59 |
| Saviq | jhodapp, no plugins, still hanging | 15:59 |
| jhodapp | Saviq, wow | 15:59 |
| jhodapp | Saviq, it might very well be the QtMultimedia and GStreamer integration | 16:00 |
| Saviq | jhodapp, well, ok, there's still some plugins that are not gstreamer1.0-plugins* | 16:00 |
| Saviq | jhodapp, trying to get rid of those too | 16:00 |
| jhodapp | k | 16:00 |
| Saviq | ! progress | 16:06 |
| Saviq | jhodapp, hah! clutter plugin seems to be the culprit | 16:13 |
| jhodapp | Saviq, oh wow | 16:15 |
| jhodapp | Saviq, did it change recently then? | 16:15 |
| jhodapp | we should file that upstream then | 16:15 |
| Saviq | jhodapp, only we'd have to convince them that's their issue ;) | 16:19 |
| jhodapp | :) | 16:19 |
| Saviq | jhodapp, https://launchpad.net/ubuntu/+source/clutter-gst-3.0 not much going on | 16:20 |
| Saviq | jhodapp, but it's difficult to say when this happened, since the registry updates fine outside of the unity8 session, at which point unity8 session won't hang either | 16:20 |
| Saviq | since the registry won't need updating | 16:21 |
| jhodapp | Saviq, yeah | 16:21 |
| jhodapp | Saviq, looks like this was the latest gst related source file to change: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/view/head:/clutter-gst/clutter-gst-video-sink.c | 16:23 |
| jhodapp | Saviq, it's possible the regression is in this change: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/revision/7#clutter-gst/clutter-gst-video-sink.c | 16:24 |
| Saviq | jhodapp, /me installing wily version | 16:24 |
| jhodapp | k | 16:25 |
| Saviq | jhodapp, hangs as well | 16:28 |
| Saviq | jhodapp, I'll try the other wily versions | 16:29 |
| jhodapp | Saviq, ok, I wonder if Unity8 creates just the right scenario for clutter to hang that maybe upstream has never seen | 16:29 |
| Saviq | jhodapp, we're not using clutter at all, that might be it :D | 16:29 |
| jhodapp | ha | 16:30 |
| jhodapp | ok | 16:30 |
| Saviq | jhodapp, even the earliest wily version has the same issue | 16:32 |
| jhodapp | wow | 16:33 |
| * Saviq checks clutter 2.0 | 16:34 | |
| jhodapp | Saviq, that was quite a large change to clutter for that first version | 16:35 |
| jhodapp | http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/changes?filter_file_id=cluttergstvideosink.-20150614041526-zq0eiljaby1bo65k-95 | 16:35 |
| jhodapp | http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/changes?filter_file_id=cluttergstvideosink.-20150614041526-zq0eiljaby1bo65k-95 | 16:35 |
| jhodapp | oops | 16:35 |
| Saviq | jhodapp, well, we'd have to compare with 2.0 | 16:35 |
| Saviq | but that's no longer there in xenial | 16:35 |
| Saviq | owait | 16:36 |
| jhodapp | agreed | 16:36 |
| jhodapp | Saviq, look at line 53: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/revision/1#clutter-gst/clutter-gst-plugin.c | 16:37 |
| Saviq | yay | 16:37 |
| jhodapp | Saviq, this would happen when the plugin is loaded via the registry | 16:37 |
| Saviq | nice one | 16:38 |
| jhodapp | thanks, there's a good chance that's it | 16:38 |
| Saviq | jhodapp, let me compile without that def | 16:38 |
| jhodapp | ok | 16:38 |
| Saviq | jhodapp, check out line 2093 http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/clutter-gst-3.0/wily/revision/1/clutter-gst/clutter-gst-playback.c#clutter-gst/clutter-gst-playback.c | 16:45 |
| * Saviq says the threads call should be protected with that | 16:46 | |
| jhodapp | Saviq, yeah | 16:47 |
| Saviq | there's more instances like that | 16:47 |
| jhodapp | Saviq, so we would no longer have CLUTTER_WINDOWING_X11 be true with Unity8 right? | 16:48 |
| Saviq | jhodapp, that's my thinking | 16:48 |
| Saviq | jhodapp, I mean, we don't | 16:48 |
| jhodapp | Saviq, can you put a #error in there to double check? | 16:49 |
| Saviq | jhodapp, but it's only the _check_windowing... that checks runtime | 16:49 |
| Saviq | the other is build time | 16:49 |
| Saviq | which is obviously wrong | 16:49 |
| jhodapp | ok | 16:49 |
| Saviq | because it goes "if built with X11 support, means you're running under X11" | 16:49 |
| jhodapp | exactly | 16:50 |
| jhodapp | what if we still had the build time requirements satisfied | 16:50 |
| jhodapp | for X11 | 16:50 |
| Saviq | jhodapp, that's where the additional check comes in | 16:57 |
| Saviq | jhodapp, http://paste.ubuntu.com/13935011/ | 16:57 |
| jhodapp | Saviq, what's that from? | 16:58 |
| jhodapp | Saviq, are we missing that? | 16:59 |
| Saviq | jhodapp, that's my patch... unfortunately not enough it seems | 16:59 |
| Saviq | it might be we're actually passing that runtime check, too | 16:59 |
| jhodapp | indeed | 16:59 |
| jhodapp | you already tried commenting out XInitThreads() completely? | 17:00 |
| Saviq | not yet | 17:00 |
| jhodapp | ok | 17:00 |
| Saviq | trying that now | 17:02 |
| jhodapp | ok | 17:04 |
| Saviq | jhodapp, hmm that this code was wrong I've almost no question, but that doesn't seem to be enough unfortunately | 17:07 |
| jhodapp | Saviq, that sucks | 17:07 |
| jhodapp | I think we're on the right trail though | 17:07 |
| Saviq | jhodapp, FWIW we have a bug that they might actually accept... running gst-inspect-1.0 hangs in vt (without X11) | 17:09 |
| jhodapp | Saviq, ah a very good point | 17:09 |
| sil2100 | Saviq: hey! A quick question - do you have any knowledge about the ubuntu-application-api3-examples package? | 17:09 |
| Saviq | sil2100, little, greyback might be better, but let's try | 17:11 |
| greyback | Depends on the question :) | 17:12 |
| sil2100 | greyback: so my question is very simple - do we need it in our images? Since the xenial touch seeds have it included | 17:12 |
| greyback | sil2100: no reason that I can think of | 17:13 |
| sil2100 | I'm doing some minor cleanup removing some api2 leftovers, was wondering if I should add it to vivid as well | 17:13 |
| sil2100 | Ok | 17:13 |
| sil2100 | greyback: thanks, good enough for me ;) | 17:13 |
| greyback | np | 17:13 |
| Saviq | jhodapp, can you confirm these steps please | 17:17 |
| Saviq | - log in on vt (with no graphical session) | 17:17 |
| Saviq | - rm ~/.cache/gstreamer-1.0/registry.x86_64.bin | 17:17 |
| Saviq | - gst-inspect-1.0 | 17:17 |
| Saviq | - behold | 17:17 |
| * Saviq can't store in LP for some reason | 17:17 | |
| jhodapp | Saviq, sure | 17:19 |
| jhodapp | Saviq, seems to have completed successfully for me | 17:20 |
| Saviq | jhodapp, oh hm, wonder if there's something else that makes this go wrong | 17:21 |
| jhodapp | Saviq, might be, I'm running wily | 17:21 |
| Saviq | ChrisTownsend, ltinklcan you guys check those steps ↑? | 17:21 |
| Saviq | that's enough to exercise the deadlock (which happens in the clutter gst plugin for me) | 17:22 |
| Saviq | jhodapp, OTOH http://tronche.com/gui/x/xlib/display/XInitThreads.html says it will go ack/nack, maybe it doesn't need the wrapping check | 17:25 |
| Saviq | which would ~explain that it doesn't help us | 17:26 |
| jhodapp | Saviq, yeah, the compile time check is practically worthless as well | 17:26 |
| Saviq | jhodapp, it is, on WIN32 ;) | 17:26 |
| jhodapp | :) | 17:26 |
| a1fa | is it me, or is unity blocking arbitrary mutli touch? | 18:01 |
| a1fa | "Unity reserves to itself gestures with 3 and 4 gestures making impossible to Touchégg make use of it. | 18:03 |
| a1fa | is there a config toggle to disable unity gestures, or are we still stuck having to compile it/remove it from source? | 18:10 |
| greyback | a1fa: you referring to unity7? If so, I don't know how to turn that off, aside from not using the unity plugin in compiz | 18:31 |
| a1fa | heh | 19:17 |
| a1fa | the only thing i am missing from unity at this point is "swipe left, swipe right" integration with chrome | 19:18 |
| traccart | ah unity7 | 19:25 |
| traccart | nope | 19:26 |
| a1fa | looks like touchegg can takeover 2 finger gestures | 19:27 |
| a1fa | if you disable all synclient options | 19:27 |
| a1fa | -_- | 19:27 |
| a1fa | but then you loose ability to scroll in unity | 19:59 |
| a1fa | -_- | 19:59 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!