| robert_ancell | desrt, what is the correct way to register two gdbus objects on different bus names? | 01:22 |
|---|---|---|
| robert_ancell | desrt, the bus name seems to be implied | 01:22 |
| === maclin_ is now known as maclin | ||
| === maclin__ is now known as maclin | ||
| === oCrazyLem is now known as CrazyLemon | ||
| === oCrazyLem is now known as CrazyLemon | ||
| darkxst | seb128, !!!! | 07:59 |
| seb128 | good morning desktopers | 08:00 |
| seb128 | darkxst, hey | 08:00 |
| darkxst | seb128, bug 1301712, fixed now, but really we don;t want the u-c-c/uoa stack!! | 08:00 |
| ubot2 | Launchpad bug 1301712 in shotwell (Ubuntu) "shotwell is pulling in unity-control-center and UOA on Ubuntu GNOME" [Low,Fix released] https://launchpad.net/bugs/1301712 | 08:00 |
| seb128 | you realize that without uoa shotwell has non working features, right? | 08:00 |
| seb128 | I just commented on that | 08:00 |
| seb128 | your call if you like to ship buggy softwares I guess | 08:00 |
| darkxst | seb128, we don't ship uoa by default | 08:00 |
| seb128 | I added the recommends because without uoa if you do publish, add an account, you get a nice fail | 08:00 |
| seb128 | right | 08:00 |
| seb128 | so you ship a buggy shotwell | 08:01 |
| darkxst | seb128, wouldnt the right thing be to disable those features if uoa is not installed? | 08:01 |
| seb128 | that's one other option, patches are welcome | 08:01 |
| seb128 | but in the current state things are just buggy | 08:02 |
| seb128 | well, we ship uoa by default so it's not so much an issue for us | 08:02 |
| darkxst | seb128, I would much prefer you discuss these things with me first! | 08:02 |
| seb128 | I added the recommends because we received some bugs about publishing not working | 08:02 |
| seb128 | I didn't even think a recommends would be an issue for you | 08:02 |
| seb128 | sorry about that | 08:02 |
| darkxst | all recommends are seeded | 08:03 |
| seb128 | yeah, I just didn't think that some flavors were using shotwell but not installing what it needs | 08:03 |
| Laney | morning | 08:03 |
| seb128 | Laney, hey | 08:03 |
| Laney | hey, wie gehts? | 08:04 |
| seb128 | good! | 08:04 |
| seb128 | you? | 08:04 |
| Laney | not bad thanks! | 08:04 |
| darkxst | seb128, do you have bugs pointing to the actual issue/ | 08:05 |
| darkxst | ? | 08:05 |
| seb128 | darkxst, open a photo, do file->publish, pick "add more accounts" from the combo box | 08:07 |
| seb128 | notice how the warnings about g_spawn_command failing | 08:07 |
| darkxst | seb128, and publish menu is provided by the uoa patch I presume? | 08:08 |
| seb128 | yes | 08:08 |
| seb128 | or rather modified | 08:08 |
| seb128 | upstream have their own accounts handling (not using goa or uoa) | 08:08 |
| darkxst | yes I noticed there was no goa integration | 08:09 |
| darkxst | seb128, ok, I will get that sorted | 08:15 |
| darkxst | seb128, Laney any chance of getting tracker into main next cycle? | 08:16 |
| GunnarHj | Morning seb128! | 08:16 |
| darkxst | we really need nautilus built with tracker support | 08:16 |
| darkxst | nautilus is the 'search provider' for file searches | 08:17 |
| darkxst | all that is completely broke right now | 08:17 |
| GunnarHj | seb128: | 08:18 |
| GunnarHj | http://blog.canonical.com/2014/04/02/shutting-down-ubuntu-one-file-services/ | 08:18 |
| GunnarHj | I missed the UIFe request. | 08:18 |
| Laney | ISTR we had problems with io load and indexing before | 08:18 |
| darkxst | has it ever been in main before? | 08:19 |
| darkxst | I though the MIR got rejected due to UBuntu using zeigeist, but that was before we became official flavour | 08:20 |
| seb128 | GunnarHj, hey | 08:21 |
| seb128 | GunnarHj, I'm unsure there was an UIFe request | 08:21 |
| seb128 | darkxst, tracker, I doubt it | 08:21 |
| seb128 | well, maybe, you can file a MIR etc for it I guess | 08:22 |
| seb128 | but I'm unsure we want to build e.g nautilus with it | 08:22 |
| seb128 | you need the indexer right? | 08:22 |
| darkxst | yeh | 08:22 |
| darkxst | the other idea would be to add search hooks for extensions, but havent checked with upstream yet if they would take that | 08:23 |
| darkxst | but its certainly not possible with current nautilus extension api | 08:23 |
| GunnarHj | seb128: Me too. ;-) Does it mean that the Ubuntu One client will be removed before final release? Should we remove the references to Ubuntu One from the docs (resulting in a few untranslated strings)? | 08:23 |
| Laney | It's gone, certainly remove it | 08:25 |
| seb128 | GunnarHj, yes, the package/support for it has started been dropped yesterday | 08:25 |
| Laney | The release team were informed in advance but I guess nobody told the docs guys | 08:25 |
| Laney | sorry :/ | 08:25 |
| seb128 | xnox, "Drop account-plugin-aim and account-plugin-yahoo from recommends to suggests, thus removing 21.8MB of gstreamer0.10 from ubuntu desktop CDs." | 08:26 |
| seb128 | that statement seems weird to me | 08:26 |
| darkxst | I never heard nothing either! until xnox removed things from our seeds | 08:26 |
| GunnarHj | Laney, seb128: I think that a hint on ubuntu-devel-announce would have been appropriate. | 08:26 |
| seb128 | GunnarHj, yeah, me too, I didn't even see the announcement! | 08:27 |
| seb128 | xnox, did you drop bluez-gstreamer as well? that use gst0.10 | 08:27 |
| seb128 | xnox, also telepathy-haze pulls gstreamer0.10 in through libpurple | 08:28 |
| Laney | cyphermox: maybe you could comment with your recommendation | 08:30 |
| Laney | empathy> "Both services have declining userbase" I wouldn't have said that ... | 08:32 |
| Laney | Bit weird to make this decision unilaterally but hey | 08:32 |
| seb128 | I might revert it | 08:32 |
| seb128 | but after talking to xnox | 08:33 |
| seb128 | I doubt he's going to achieve what he wanted | 08:33 |
| seb128 | we need to keep haze on the CD and that keeps gst0.10 | 08:33 |
| seb128 | bluez-gstreamer as well (though I'm unsure what that does and what would we miss without it) | 08:33 |
| Laney | cyphermox said that one is ok | 08:34 |
| Laney | I think it lets you use a2dp | 08:35 |
| seb128 | did he add details on what it doeS? | 08:35 |
| seb128 | what is a2dp? | 08:35 |
| Laney | i.e. send audio over bluetooth | 08:35 |
| * seb128 clueless about bluetooth | 08:35 | |
| seb128 | shrug | 08:35 |
| seb128 | that seems like an useful feature to me! | 08:35 |
| Laney | I think it's in mainline gstreamer with 1.0 | 08:36 |
| Laney | so don't know how useful it is with 0.10 | 08:36 |
| seb128 | oh, nice | 08:36 |
| seb128 | well, as said I would like to understand how those things are used | 08:36 |
| seb128 | before we decide to drop stuff a week before release | 08:36 |
| Laney | laney@iota> gst-inspect-1.0 | grep a2dp ~ | 08:36 |
| Laney | bluez: a2dpsink: Bluetooth A2DP sink | 08:36 |
| seb128 | to then notice after release that after all those stuff were needed and that we regressed usecases | 08:36 |
| seb128 | $ gst-inspect-0.10 /usr/lib/i386-linux-gnu/gstreamer-0.10/libgstbluetooth.so | 08:37 |
| seb128 | ... | 08:37 |
| seb128 | rtpsbcpay: RTP packet payloader | 08:37 |
| seb128 | a2dpsink: Bluetooth A2DP sink | 08:37 |
| seb128 | avdtpsink: Bluetooth AVDTP sink | 08:37 |
| seb128 | sbcparse: Bluetooth SBC parser | 08:37 |
| seb128 | sbcdec: Bluetooth SBC decoder | 08:37 |
| seb128 | sbcenc: Bluetooth SBC encoder | 08:37 |
| seb128 | 08:37 | |
| seb128 | no "SBC decoder" in my gst-inspect-1.0 | 08:38 |
| Laney | yeah that's in bad for 1.0 afaics | 08:38 |
| seb128 | k, I don't have that installed | 08:38 |
| darkxst | I could never get a2dp working anyway, probably not a huge loss! | 08:41 |
| darkxst | ^maybe iphone related though, the iphone stack is an outdated mess right now | 08:42 |
| Laney | never tried it | 08:44 |
| Laney | not that I have an iphone, but I guess it should work with android too | 08:45 |
| darkxst | yeh, although a2dp is a standard! works fine in any modern car (maybe there are patent issues though?) | 08:47 |
| darkxst | so probably what I hit was just a bug | 08:48 |
| seb128 | I don't even know how to test that | 08:48 |
| seb128 | is that just "pair your phone as a bt device, play music through it as an output"? | 08:48 |
| darkxst | seb128, yes, so you pair phone with computer, and then the music play get routed through pulseaudio | 08:49 |
| darkxst | ^ the music you play on phone | 08:49 |
| Laney | computer doesn't even see my phone to pair with it | 08:52 |
| seb128 | great, u-c-c segfaulted when I tried to enable the audio for media option on the phone | 08:52 |
| seb128 | in libgnome-bluetooth | 08:52 |
| Laney | this is severely buggy :( | 08:54 |
| Laney | the enabled state isn't remembered or synced with the indicator | 08:54 |
| seb128 | you likely have a bluez/kernel issue | 08:54 |
| Laney | never sees the phone, get operation in progress on stderr when doing things | 08:54 |
| darkxst | Laney, right, I just gave up on the idea :) and plugged some speakers into my phone! | 09:01 |
| sasa84 | hello | 09:06 |
| sasa84 | part of apport is not translated https://dl.dropboxusercontent.com/u/17510489/Apport.png | 09:07 |
| sasa84 | actually there are no strings for translation either https://translations.launchpad.net/ubuntu/trusty/+source/apport/+pots/apport/sl/+translate | 09:08 |
| seb128 | sasa84, hey, that's "normal", hooks are not translatable atm I think | 09:09 |
| xnox | seb128: why would you want telepathy-haze on the images? | 09:10 |
| seb128 | xnox, because things like gadugadu (the most popular im in China) have no other support for it that through libpurple | 09:11 |
| xnox | seb128: poland, qq is in china =) | 09:11 |
| seb128 | oh, sorry | 09:11 |
| seb128 | but the argument still stands | 09:11 |
| seb128 | some protocols are popular in countries and have no telepathy connectors | 09:12 |
| xnox | seb128: and those support/use audio/video via libpurple? | 09:12 |
| seb128 | so they need to use libpurple through haze | 09:12 |
| seb128 | not sure | 09:12 |
| seb128 | they do use libpurple from texting/messages | 09:12 |
| xnox | seb128: why do i not see telepathy-haze in the seed sources? | 09:12 |
| seb128 | but libpurple is one lib | 09:12 |
| seb128 | the a/v part is not split out | 09:12 |
| seb128 | xnox, because it's a recommends of empathy | 09:13 |
| seb128 | we could seed all recommends of all softwares | 09:13 |
| seb128 | not sure what's the point though | 09:13 |
| xnox | nah, no point. For ubuntu, we do choose to seed in recommends. | 09:13 |
| seb128 | well, it's already a recommends from empathy | 09:13 |
| seb128 | not sure why you want to duplicate that info? | 09:14 |
| xnox | no, i don't. Just wanted to know how it was seeded in, to make sure it's not just a dep. | 09:14 |
| xnox | so another way is to hunt down upstream patches which port purple to gstreamer0.10, and then the empathy change to downgrade two accounts can be reverted as well. | 09:15 |
| xnox | cause at this point libpurple is the last one holding up gstreamer0.10 on desktop images. | 09:15 |
| Laney | They did it on a different branch, for the 3.0 series. | 09:15 |
| xnox | is last correct statement? | 09:15 |
| seb128 | I think it is | 09:16 |
| seb128 | but pigdin is moving slowly | 09:16 |
| xnox | Laney: right, i'll look to see if it's self-contained to be cherry-picked. | 09:16 |
| seb128 | they are still gtk2 and gst1.0 | 09:16 |
| seb128 | gst0.10 sorry | 09:16 |
| seb128 | they are porting to gtk2/gst1.0 but when I looked at it there was no easy change we could backport to migrate to gst1.0 | 09:16 |
| sasa84 | ok, tnx seb128 | 09:16 |
| seb128 | xnox, thanks | 09:16 |
| Laney | I don't think it's going to be as easy as you think it might be | 09:17 |
| Laney | otherwise we would have done it back then | 09:17 |
| * darkxst would love to see gstreamer 0.10 be gone from our images ;) (not that I have been following the whole story) | 09:17 | |
| Laney | Risky this close to release too | 09:18 |
| darkxst | hmm I was wasnt really talking about this release! | 09:19 |
| seb128 | xnox is ;-) | 09:19 |
| xnox | seb128: i'm not buying this gadu-gadu & qq argument =) let me check the depends again. | 09:21 |
| * darkxst also wonders why I generally have no idea about what is happening until (almost) after the fact | 09:22 | |
| seb128 | xnox, well, also yahoo ... which you decided was not useful anymore, but I don't think we have data on that being true or false | 09:22 |
| xnox | seb128: so to get gadugadu support in empathy, i need account-plugin-gadugadu which is universe and that's not of my doing. | 09:22 |
| seb128 | darkxst, is that comment about the u1 change? or do you have other examples? | 09:22 |
| darkxst | probably mostly, I find out after things break! :( | 09:22 |
| xnox | seb128: and that depends on telepathy-haze | 09:22 |
| seb128 | darkxst, well, I could say the same, I woke up this morning to see that you dropped the recommends I added yesterday :p | 09:23 |
| xnox | seb128: and i don't see a qq accounts plugin. | 09:23 |
| seb128 | xnox, empathy doesn't enforce ubuntu-online-account, you can go to empathy-preferences and add accounts | 09:23 |
| seb128 | sorry, "empathy-accounts" | 09:24 |
| darkxst | seb128, you seeded the entire unity-control-center/UOA stack on our images!!!! | 09:24 |
| xnox | seb128: oh, the "old UI" ?! | 09:24 |
| darkxst | that is bad ;) | 09:24 |
| seb128 | xnox, the only UI to add accounts that are not supported by uoa | 09:24 |
| xnox | seb128: how to open that? F4/empathy->accounts opens online-accounts. | 09:25 |
| seb128 | darkxst, well, I'm just saying, you did a change during my night and I woke up to it, it's not only to you that those things happen | 09:25 |
| seb128 | xnox, alt-f2 empathy-accounts | 09:25 |
| xnox | seb128: that does offer gadugadu. but... there is no desktop/UI way to get there? | 09:27 |
| seb128 | xnox, not that I know offhand, which is a known issue, we force users into an incomplete account list :/ | 09:27 |
| darkxst | seb128, sorry but +59 packages into a seed, is quite the emergency! | 09:27 |
| seb128 | darkxst, I'm not criticizing the change, just saying that people do work and don't always stop to discuss every change | 09:28 |
| darkxst | seb128, sure, guess I am just pushing a bit more cross-flavour considerations, we do (try atleast) to test everything against ubuntu before upload | 09:31 |
| xnox | so we ship haze, and hidden ui to use it, but no anything discoverable =( | 09:31 |
| darkxst | just that doesnt happen in reverse | 09:31 |
| seb128 | xnox, correct | 09:31 |
| darkxst | although when/if we can get gnome-shell autopkg tests running, I guess that can't happen anymore | 09:32 |
| seb128 | xnox, I could see an argument to drop it from the default install saying it's not accessible to normal users anyway, but doing such change a week before the freeze makes me nervous | 09:32 |
| Laney | Would it break the accounts of people upgrading? | 09:32 |
| seb128 | darkxst, it's not easy to test every flavor, image if we had to test kubuntu, xubuntu, lubuntu, GNOME Ubuntu, etc before every upload | 09:33 |
| seb128 | Laney, we wouldn't remove those binaries on upgrade I guess? | 09:33 |
| seb128 | Laney, or is update-manager agressive about those things? | 09:33 |
| seb128 | which makes me thing I still need to talk to mvo_ about that | 09:33 |
| xnox | Laney: if they had online-accounts-gadugadu installed then no, cause that hard depends on haze and it would stay. | 09:33 |
| seb128 | to see if we can remove gnome-control-center on precise->trusty updates | 09:33 |
| darkxst | seb128, its really only Ubuntu GNOME that has overlap ? | 09:34 |
| xnox | Laney: actually no, it will not break on upgrades, cause upon upgrade haze would stay. | 09:34 |
| mvo_ | seb128: about what exactly? sorry missed parts of the conversation | 09:34 |
| darkxst | I supposed edubuntu uses gnome-flashback too | 09:34 |
| xnox | Laney: cause we are discussing dropping -haze from depends to recommends in empathy. | 09:34 |
| seb128 | darkxst, not really, GTK is used in most flavors | 09:34 |
| Laney | Oh, I thought you wanted to get it off the image | 09:35 |
| Laney | Don't know what the point is then | 09:35 |
| seb128 | mvo_, we transitioned Unity from gnome-control-center/gnome-settings-daemon to unity-control-center/unity-settings-daemon in trusty, I'm wondering if we can make the dist-upgrader clean g-c-c/g-s-d on upgrade if nothing depends on those | 09:35 |
| darkxst | gtk bugs are generally not fatal, just annoying | 09:35 |
| seb128 | darkxst, well, adding uoa to your seed was not a runtime issue, just annoying :p | 09:35 |
| seb128 | but yeah, we try to be good citizen | 09:35 |
| xnox | Laney: dropping -haze from recommends to suggests on empathy, will effectively unseed haze/purple/gstreamer0.10 | 09:35 |
| xnox | Laney: we don't install suggests, do we? | 09:35 |
| Laney | what | 09:35 |
| darkxst | we have had gobject-introspection issues that break gdm... | 09:36 |
| Laney | 03/04 10:34:28 <xnox> Laney: cause we are discussing dropping -haze from depends to recommends in empathy. | 09:36 |
| Laney | <xnox> Laney: dropping -haze from recommends to suggests on empathy | 09:36 |
| xnox | Laney: TYPO! | 09:36 |
| seb128 | it's already a recommends | 09:36 |
| xnox | Laney: need more coffee =))))))) | 09:36 |
| seb128 | it's recommends->suggests | 09:36 |
| xnox | that ^ | 09:36 |
| darkxst | seb128, that and a 100 messages why is the unity world being pulled into ubuntu GNOME :( | 09:36 |
| seb128 | darkxst, yeah, that's unfortunate :/ | 09:36 |
| seb128 | darkxst, anyway mistake happen, sorry about that | 09:36 |
| seb128 | and +1 on autopkgtests for gnome-shell | 09:36 |
| seb128 | xnox, I cleaned those u1 sources from trusty | 09:37 |
| darkxst | seb128, next cycle, its hard ;( | 09:37 |
| mvo_ | seb128: yes, that should be possible | 09:37 |
| xnox | seb128: \o/ | 09:37 |
| xnox | seb128: thanks. I think u1 is all done for trusty, but possibly unity (in-silo) and thunderbird. | 09:37 |
| darkxst | autopilot and maybe gnome-shell, need to be patched to play with clutter widgets | 09:37 |
| seb128 | we don't have autopilot integrated with autopkgtest | 09:38 |
| seb128 | and yeah, autopilot and clutter sounds like "fun" | 09:38 |
| darkxst | seb128, pitti mentioned it was possible (apart from the clutter bit) | 09:39 |
| seb128 | cool | 09:39 |
| xnox | seb128: cjwatson did write a proof-of-concept running autopilot tests as autopkgtests =) | 09:41 |
| seb128 | nice | 09:41 |
| darkxst | seb128, I actually have a couple of 'apprentices' now (CS students I guess) that at are good are code, but hopeless at packaging | 09:41 |
| darkxst | seb128, I doubt upstream will care much about in bugs in 3.8... | 09:47 |
| === vrruiz_ is now known as rvr | ||
| seb128 | darkxst, that's what the new RHEL ships, I somewhat think that's going to help a bit | 09:48 |
| darkxst | seb128, getting source from RHEL is a pita | 09:48 |
| seb128 | some of the fixes are flowing back in upstream git | 09:49 |
| darkxst | maybe some but not most! | 09:49 |
| seb128 | but yeah, GNOME upstream doesn't do much stable serie work | 09:49 |
| seb128 | it would be true for any serie | 09:49 |
| seb128 | if we had 3.10 we would soon stop getting fixes for that as well | 09:50 |
| xnox | what component do we use for XMPP in telepathy? it's telepathy-native? cause libpurple uses gstreamer for XMPP only audio/video by the looks of things. | 09:50 |
| darkxst | seb128, watch this space, I will make sure all patches that affect us are cherry picked to the 3.10 branch; ) | 09:50 |
| xnox | but recompiling pidgin without gstreamer, would piss off anybody to uses pidgin for audio/video xmpp | 09:50 |
| xnox | unless we compile that twice..... | 09:51 |
| xnox | sounds risky | 09:51 |
| seb128 | darkxst, ;-) | 09:51 |
| darkxst | except for the stupid gnome-tweak-tool authors who refuse to take python3 pathes | 09:51 |
| seb128 | xnox, telepathy-gabble | 09:52 |
| Laney | I'd really rather avoiding doing any risky gstreamer stuff now | 09:52 |
| Laney | if you wanted to try that it should have been a couple of months ago | 09:52 |
| seb128 | +1 | 09:54 |
| xnox | seb128: i'm confused, reading the buildlog empathy is compiled with --enable-gst1.0 option and it does link against gstreamer1.0 in the archive. | 09:55 |
| xnox | reading wrong project buildlog | 09:56 |
| darkxst | seb128, you would probably be amazed at what the the super secret RHEL teams backport, but its not generally available until it filters into CentOS | 09:56 |
| Laney | do they do the monolithic patches thing for all packages? | 09:57 |
| darkxst | I don't know, I have never been able to find a single patch to see! | 09:57 |
| seb128 | darkxst, I've to admit I never tried to look much at what they do indeed | 09:58 |
| Laney | If not some kind of patch tracker for rhel would be interesting ;-) | 09:59 |
| Laney | seb128: Trevinho: want to try getting https://code.launchpad.net/~3v1n0/unity/forcequit-dialog/+merge/213832 in? | 10:00 |
| seb128 | Laney, it's in silo 5 if you want to test the ppa | 10:01 |
| Laney | oh cool | 10:01 |
| seb128 | https://launchpad.net/~ci-train-ppa-service/+archive/landing-005/+packages | 10:01 |
| Laney | how can you trigger a hung window? | 10:01 |
| seb128 | gedit & gdb -p `pidof gedit` | 10:02 |
| seb128 | N | 10:02 |
| seb128 | ? | 10:02 |
| Laney | oh yes good point | 10:02 |
| Laney | I was thinking stuff like that would close the window for some reason | 10:02 |
| * Laney is broken | 10:02 | |
| seb128 | Trevinho, Laney, bregma: those changes to the close dialog seem buggy, I get 2 dialogs to show, the old and new ones | 10:06 |
| * seb128 restarts session in case alt-f2 -> unity was not a proper restart | 10:07 | |
| Laney | urgh I broke everything | 10:07 |
| seb128 | what did you do? | 10:07 |
| Laney | I logged out to start a new session with it | 10:08 |
| Laney | which took ages waiting for some timeout | 10:08 |
| Laney | so I tried to use loginctl session-status to see what it was | 10:08 |
| Laney | that was segfaulting | 10:08 |
| seb128 | urg | 10:08 |
| Laney | and then it hung logging out the next time and restarting lightdm hasn't got it back up | 10:08 |
| Laney | weeeeeeeeee | 10:08 |
| Laney | oh there it is, in a messed up resolution, wtf | 10:09 |
| * Laney restarts | 10:09 | |
| ochosi | Laney: what was in a messed up resolution? just asking cause we recently got a lightdm-gtk-greeter bug assigned from robert that complains about the greeter not being displayed in the correct resolution (could never reproduce that so it's a bit poking in the dark for me..) | 10:11 |
| Laney | lightdm | 10:11 |
| Laney | after I restarted it | 10:11 |
| Laney | it had ignored the settings I think because it was mirrored too | 10:12 |
| Laney | seb128: seems fine to me, I only got one dialog | 10:13 |
| seb128 | Laney, I still get both after a restart :/ | 10:15 |
| Laney | weird | 10:15 |
| ochosi | Laney: so unity-greeter was shown in the wrong resolution? | 10:17 |
| Laney | yes | 10:17 |
| Laney | hmm it's spotify holding the session open | 10:18 |
| seb128 | greeter sessions are left open because of indicators here :/ | 10:19 |
| ochosi | Laney: mind to add whether this should affect unity-greeter too then? https://bugs.launchpad.net/bugs/1300153 | 10:19 |
| ubot2 | Launchpad bug 1300153 in lightdm-gtk-greeter (Ubuntu) "login Screen is not showing correctly" [Undecided,New] | 10:19 |
| ochosi | seb128: well from what i've seen, they get quite brutally killed in unity-greeter (and gtk-greeter, cause we mostly ported unity-greeter's indicator behavior) | 10:20 |
| Laney | ochosi: It's unlikely to be the same problem | 10:21 |
| Laney | or at least I can't say if it is | 10:21 |
| seb128 | ochosi, well, at least indicator bluetooth/sound are managed by upstart | 10:21 |
| seb128 | so it looks like unity-greeter fails to send the signal that is used to stop the jobs | 10:21 |
| seb128 | stop on desktop-end or indicator-services-end | 10:21 |
| seb128 | so I guess the greeter should send indicator-services-end at login and doesn't | 10:22 |
| === MacSlow is now known as MacSlow|lunch | ||
| seb128 | ok, time for some errands and getting lunch on the way, back in 45 minutes or so | 10:27 |
| ochosi | Laney: afaik unity-greeter and lightdm-gtk-greeter use the same logic to get the screen size/resolution | 10:36 |
| mlankhorst | afternoon | 10:47 |
| mlankhorst | not feeling well, running a fever. :( | 10:47 |
| === ara is now known as Guest11257 | ||
| seb128 | ochosi, unity-greeter uses unity-settings-daemon xrandr plugin | 11:21 |
| seb128 | so things might be different | 11:21 |
| Laney | jdstrand: can I add /usr/local/share/glib-*/schemas/ to telepathy's profile? | 11:30 |
| Laney | I get some denials for that when using empathy | 11:30 |
| seb128 | Laney, /usr/local? | 11:31 |
| seb128 | Laney, what did you do! | 11:31 |
| Laney | $ empathy | 11:31 |
| Laney | dmesg | 11:31 |
| seb128 | ls /usr/local/share/glib-*/schemas/ | 11:31 |
| seb128 | /usr/local should be empty | 11:31 |
| seb128 | did you sudo make install thingS? | 11:31 |
| Laney | yes I have some schemas there | 11:31 |
| Laney | it's a legal thing to do :-) | 11:31 |
| seb128 | lol | 11:31 |
| seb128 | well, if you can local install you can tweak your apparmor profiles ;-) | 11:32 |
| seb128 | or do we usually include /usr/local paths to those? | 11:32 |
| * seb128 checks | 11:32 | |
| Laney | other profiles handle /usr/local | 11:32 |
| seb128 | k | 11:32 |
| bregma | seb128, I have removed the close dialog MP fro the current U7 silo because that needs to land for other reasons, we'll try it again in the next landing | 11:40 |
| * bregma takes the dog for a walk | 11:40 | |
| GunnarHj | seb128: Is remote login affected by the Ubuntu One shutdown, or is it login.ubuntu.com that is part of the remote login feature? | 11:42 |
| GunnarHj | https://help.ubuntu.com/14.04/ubuntu-help/sharing-remote-login.html | 11:42 |
| GunnarHj | "The shutdown will not affect the Ubuntu One single sign on service, the Ubuntu One payment service, or the backend U1DB database service." | 11:42 |
| seb128 | bregma, ok | 11:43 |
| seb128 | GunnarHj, hum, I would have say it should not be affected but I'm unsure | 11:45 |
| seb128 | dbarth, ^ can you reply to that? | 11:45 |
| GunnarHj | seb128, dbarth: I was about to remove that page, but then I started to hesitate... | 11:46 |
| seb128 | GunnarHj, I think we stopped installing-by-default the remote login package earlier in the cycle, you can probably drop that page | 11:51 |
| dbarth | GunnarHj, seb128: confirmed, you can drop it | 11:52 |
| seb128 | dbarth, k, it would still be useful if you replied to the question though ;-) | 11:53 |
| dbarth | seb128: eh, which question exactly? | 11:53 |
| dbarth | (just reading the backlog after lunch) | 11:53 |
| dbarth | so, the remote login feature is not affected by the change | 11:54 |
| seb128 | dbarth, is lightdm-remote-session-uccsconfigure relying on u1? said differently, is it still a working package without the filesync service? | 11:54 |
| dbarth | not affected by the U1 shutdown to be more precise | 11:54 |
| seb128 | dbarth, the documentation GunnarHj pointed states "The system is designed so that Ubuntu One securely stores your login information " | 11:54 |
| dbarth | the remote login feature is not enabled by default anymore | 11:55 |
| seb128 | dbarth, ok, good, thanks ;-) | 11:55 |
| GunnarHj | seb128, dbarth: Thanks, then I know how to proceed. | 11:55 |
| Laney | If you want to document it then change it to tell people how to install the package instead | 11:55 |
| seb128 | dbarth, right, still if he stopped working we should remove the package for Ubuntu as well | 11:55 |
| Laney | :-) | 11:55 |
| dbarth | and, now that we dive into this, i think we should also drop the uccsconfigure part | 11:55 |
| seb128 | he->it | 11:55 |
| dbarth | i thought that was the case already | 11:55 |
| seb128 | https://launchpad.net/ubuntu/+source/lightdm-remote-session-uccsconfigure | 11:55 |
| seb128 | that is still in trusty | 11:55 |
| dbarth | but really we only want to keep the technical bit around, in case enterprises want to use that | 11:56 |
| seb128 | if we want to stop supporting it, maybe we should delete it before release | 11:56 |
| seb128 | k | 11:56 |
| dbarth | seb128: think i still have time to request a drop? | 11:56 |
| seb128 | yes | 11:56 |
| dbarth | i will sync up with jason & robert as well | 11:56 |
| seb128 | k | 11:56 |
| GunnarHj | Laney: Well, yes explaining how to install instead of how to remove would of course be an option... | 11:59 |
| === alan_g is now known as alan_g|lunch | ||
| ochosi | seb128: oh, indeed. i thought it was mainly using gdk to determine screen-size etc | 12:13 |
| seb128 | ochosi, it could be, the xrandr plugin is handling some of the work, I don't think it does much on start though (the screen config is done by xorg afaik) | 12:19 |
| ochosi | yeah, i thought so too | 12:19 |
| ochosi | even moving the login-window to different monitors didn | 12:19 |
| ochosi | t seem to involve randr | 12:20 |
| seb128 | no, xrandr basically set the screens config, e.g which one is primary, if it's mirror mode or xinerama, and their resolution | 12:21 |
| seb128 | so it's "xorg server config" more than "greeter work" | 12:21 |
| ochosi | ah, that's what you meant | 12:22 |
| jdstrand | Laney: I guess it would be ok, but I have a feeling it wouldn't be enough in the general case. that said I would suggest adjusting /etc/apparmor.d/local/usr.lib.telepathy which is designed for site-specific additions | 12:23 |
| Sweetshark | seb128: I talked to Rene, he is not interested in the fix as LO42 is unlikely to end up in their stable. So please review http://people.canonical.com/~bjoern/trusty/accessodf_0.1-1.3ubuntu2_amd64.changes and upload. diff is here: http://people.canonical.com/~bjoern/trusty/accessodf_0.1-1.3ubuntu2.diff | 12:31 |
| seb128 | Sweetshark, ok, thanks ... amd64.changes, we do source uploads here ;-) | 12:32 |
| Sweetshark | seb128: ahh, this open source stuff is a hype that will soon pass. ;) | 12:33 |
| * Sweetshark uploads the source changes quickly ... | 12:33 | |
| Sweetshark | seb128: http://people.canonical.com/~bjoern/trusty/accessodf_0.1-1.3ubuntu2_source.changes | 12:35 |
| seb128 | Sweetshark, thanks ;-) | 12:35 |
| * Sweetshark listens to 'While my guitar gently weeps' in the ukulele version. There might be some irony hidden in there. | 12:42 | |
| brainwash | indicator-sound installs /usr/share/upstart/xdg/autostart/indicator-sound.desktop which contains the line "hidden=True", this file seems to override the normal one located in /etc/xdg/autostart/ | 12:43 |
| brainwash | so, xfce4-session shows the indicator sound autostart launcher as unchecked | 12:44 |
| seb128 | tedg, ^ | 12:44 |
| brainwash | recent indicator-sound change in trusty | 12:44 |
| brainwash | I assume that it prefers the upstart launcher, because we are running an upstart user session | 12:45 |
| === jhernand1z is now known as jhernandez | ||
| === mjohnson15_2 is now known as mjohnson15 | ||
| === alan_g|lunch is now known as alan_g | ||
| seb128 | Laney, do you have an opinion on https://bugs.launchpad.net/ubuntu/+source/rhythmbox/+bug/1296334 ? | 13:12 |
| ubot2 | Launchpad bug 1296334 in rhythmbox (Ubuntu) "Update to bug-fix release 3.0.2 in Trusty" [High,Confirmed] | 13:12 |
| seb128 | dpm, Laney, tedg: https://code.launchpad.net/~attente/indicator-keyboard/1291962-2/+merge/213346 is changing a "..." to "…" which changes the string/invalidates the translations, do you think it's ok to do now or is that too close from release? | 13:20 |
| Laney | seb128: looks ok, what do you think? | 13:20 |
| seb128 | +1, I would prefer have the items right/consistent for the lts, it's an easy translate update, I'm going to drop a note to the translation team | 13:21 |
| tedg | Yeah, I think that's best. | 13:21 |
| seb128 | Laney, or are you speaking about rhythmbox? ;-) | 13:21 |
| seb128 | which I think is ok as well | 13:21 |
| Laney | rb | 13:21 |
| seb128 | k | 13:21 |
| Laney | breaking translations not so sure | 13:21 |
| seb128 | I'm going to do that update now | 13:21 |
| seb128 | rb | 13:21 |
| dpm | seb128, could we do it after release? I've had experience with exactly this change bringing lots of confusion and frustration because translations break | 13:22 |
| Laney | jdstrand: that's the only profile I have installed here which references the schemas | 13:22 |
| Laney | I'll upload that, cheers | 13:23 |
| seb128 | dpm, doing after release, like in a SRU? | 13:23 |
| seb128 | dpm, or next cycle? | 13:23 |
| Laney | yeah I can imagine how it'd be annoying to have to rush to re-translate something like that before release | 13:23 |
| dpm | seb128, I'd say next cycle | 13:24 |
| Laney | I guess there's no way to say 'copy the old translations for this new string' | 13:24 |
| seb128 | not in launchpad | 13:24 |
| dpm | actually the developer could do that and reupload the package with the new English string and the old translations | 13:25 |
| seb128 | we were able to do that when translations were in the source | 13:25 |
| dpm | LP would still import the translations uploaded in the .mo files | 13:25 |
| seb128 | right | 13:25 |
| Laney | so import them and then hack the po files or something | 13:25 |
| dpm | so the developer would have to hack all the .po files | 13:25 |
| dpm | and the .pot file | 13:26 |
| seb128 | that would mean doing a full export, adding those to the source, doing some seeding/intltool-merge, commit that, CI train, clean the source back | 13:26 |
| dpm | to replace the original English string | 13:26 |
| seb128 | I can't be bothered for one string | 13:26 |
| Laney | haha | 13:26 |
| dpm | yes, it's quite a lot of work | 13:26 |
| Laney | yes indeed | 13:26 |
| seb128 | it's easier for a launchpad admin to click through the locales in the webui and revalidate the translations :p | 13:27 |
| brainwash | tedg: any idea why the recent indicator-sound update broke it? | 13:44 |
| Laney | You might want to give a bit more detail than that | 13:44 |
| brainwash | backlog | 13:44 |
| tedg | brainwash, I probably don't have enough backlog :-) | 13:45 |
| brainwash | "indicator-sound installs /usr/share/upstart/xdg/autostart/indicator-sound.desktop which contains the line "hidden=True", this file seems to override the normal one located in /etc/xdg/autostart/" | 13:45 |
| brainwash | "so, xfce4-session shows the indicator sound autostart launcher as unchecked" | 13:46 |
| tedg | Yes, and that should only be in your XDG_DATA_DIRS if you're running an upstart user session, no? | 13:46 |
| brainwash | yes, it's an upstart user session | 13:46 |
| brainwash | used by default in xubuntu | 13:46 |
| tedg | K, so does it start? | 13:46 |
| lenny | tedg: no | 13:47 |
| brainwash | it only starts if you tell xfce4-session to override "hidden=True" | 13:47 |
| Laney | I bet nothing emits the event | 13:47 |
| tedg | I'm not sure how the xfce4-session works, but if it's only using desktop files that's not really going to work with Upstart user sessions. | 13:47 |
| Laney | what happens if you run initctl emit indicator-services-start | 13:47 |
| Laney | ? | 13:47 |
| lenny | Laney: Nothing | 13:48 |
| Laney | status indicator-sound | 13:49 |
| * tedg is upset that Laney is too fast | 13:49 | |
| lenny | indicator-sound stop/waiting | 13:49 |
| Laney | Fishy | 13:50 |
| Laney | start indicator-sound | 13:50 |
| brainwash | Laney: the command starts all indicators | 13:50 |
| lenny | indicator-sound start/running, process 2437 (no audio indicator in tray though) | 13:50 |
| Laney | brainwash: it works for you? | 13:50 |
| brainwash | Laney: yes, but it did start the whole indicator stack (datetime, sound,..) | 13:51 |
| Laney | is that bad? | 13:51 |
| brainwash | it's a change | 13:51 |
| brainwash | we were used to enable/disable indicators via xdg/autostart | 13:52 |
| Laney | oh right, that's going to happen then | 13:52 |
| Laney | if you 'echo manual > ~/.config/upstart/indicator-foo.conf' it should override it | 13:52 |
| === ara_ is now known as ara | ||
| Laney | So you should make some part of your desktop emit this event | 13:53 |
| Laney | You could do it from an upstart job (unity7) or code (lightdm-gtk-greeter) or a .desktop file (gnome-panel) | 13:54 |
| tedg | Technically unity7 does it from code, it emits the event in unity-panel-service when it's ready for the indicators. | 13:55 |
| tedg | But yes, all of those work. | 13:55 |
| Laney | oh yeah, just read the summaries on http://162.213.35.4/search?weighted=1&q=indicator-services-start | 13:55 |
| brainwash | yes, but I'm not sure if the xubuntu team wants to change the way the indicator start | 13:55 |
| brainwash | starts | 13:55 |
| Laney | then change the override file to drop XFCE | 13:56 |
| Laney | Not sure what the point of having upstart sessions is if you don't want to use them though | 13:56 |
| brainwash | upstart user sessions were enabled by default | 13:57 |
| seb128 | tedg, so, question for you | 13:57 |
| Laney | Someone listed it in /etc/upstart-xsessions | 13:57 |
| tedg | seb128, Heh, and I bet you're going to want an answer too! ;-) | 13:57 |
| seb128 | tedg, loginctl lists some lightdm sessions for me, those keep running because of indicator processes | 13:58 |
| seb128 | tedg, should we have unity-greeter emitting indicator-services-end at login so those get stopped? | 13:58 |
| tedg | seb128, It shouldn't have to, sending sigterm to Upstart should do it. Do you have upstart's running as well? | 13:58 |
| tedg | Upstarti? | 13:59 |
| seb128 | CGroup: systemd:/user/119.user/c10.session | 13:59 |
| seb128 | ├─13876 init --user --startup-event indicator-services-star... | 13:59 |
| seb128 | ├─13961 /usr/lib/i386-linux-gnu/indicator-bluetooth/indicat... | 13:59 |
| seb128 | ├─13965 /usr/lib/i386-linux-gnu/indicator-sound/indicator-s... | 13:59 |
| brainwash | Laney: ok, I'll try to explain the situation to the xubuntu team. thanks :) | 13:59 |
| seb128 | tedg, yes | 13:59 |
| Laney | brainwash: If you get removed from that file then you can stop using the user sessions | 13:59 |
| Laney | Those are all of the options that I know of :-) | 13:59 |
| seb128 | tedg, doesn't happen for you? (try to start a guest session and log back to your user) | 13:59 |
| seb128 | oh, urg, settings meeting time! | 14:00 |
| tedg | seb128, Hmm, let me see. | 14:00 |
| Laney | is that now? | 14:00 |
| seb128 | tedg, Laney, charles, meeting :p | 14:00 |
| seb128 | well, according to my google calendar | 14:00 |
| Laney | I got tz confused | 14:00 |
| Laney | what evs, now is fine | 14:00 |
| seb128 | we never defined if that was utc constant | 14:00 |
| seb128 | I'm fine shifting 1 hour if you guys prefer | 14:00 |
| Laney | 1 minute, refilling tea | 14:00 |
| tedg | I don't care either way. | 14:01 |
| seb128 | let's do it now/keep that time then | 14:02 |
| tedg | But I do have 1 additional Upstart. Which is odd, seems to not be consistent :-/ | 14:02 |
| xnox | seb128: Laney: cherrypicked/ported pidgin to gstreamer1.0 and that compiles, but fails at runtime. Uploaded into unapproved empathy that drops haze from recommends to suggests, since there is no default/discovarable UI to configure any of the accounts that haze/libpurple provide and upgrades are unaffected. | 14:02 |
| tedg | I guess there are two options, Upstart isn't getting the SIGTERM or it's ignoring it. | 14:03 |
| tedg | xnox, Is there anyway that Upstart could ignore a SIGTERM? | 14:03 |
| Laney | haze: whatever (maybe the docs team want to document this though) | 14:03 |
| Laney | aim/yahoo, not sure at all, might want to revert that | 14:03 |
| xnox | Laney: aim & yahoo hard depend on telepathy-haze... | 14:04 |
| xnox | i should boot an older image to see the net-difference of available options. | 14:05 |
| * Laney shrugs | 14:06 | |
| mvo_ | gar, I made the mistake to check the lkml systemd/debug cmdline thread :/ | 14:06 |
| Laney | if that's the case then we get to keep it | 14:06 |
| xnox | Laney: how should i propose the change for discussion? ubuntu-devel post? | 14:08 |
| Laney | Can we just think about it after release? | 14:08 |
| xnox | Laney: imho we should be providing whatsapp & facebook chat login options as together they outnumber 10x-20x active users of aim&yahoo | 14:09 |
| Laney | I get the goal of dropping gstreamer but we've lived with it this far and I'd rather not rush it through now | 14:09 |
| xnox | Laney: dropping gstreamer is very technical/internal change, dropping IM networks is user-visible. | 14:10 |
| Laney | yes I know | 14:10 |
| Laney | you are trying to achieve the former by doing the latter | 14:11 |
| Laney | (facebook chat works already btw) | 14:11 |
| xnox | Laney: and if the goal is to drop gstreamer out of main, then pidgin pitivi qt-phonon all need porting, none of which will be done for trusty. | 14:11 |
| xnox | Laney: yeah, i guess reject proposed empathy and revert the previous upload. | 14:11 |
| Mirv | re: pitivi bug #1253009 still open though, pre-requirements were synced seemingly a few days ago | 14:12 |
| ubot2 | Launchpad bug 1253009 in pitivi (Baltix) "[FFe] Please sync latest upstream release (0.9x) from Debian experimental - Pitivi developers recommends to use 0.92 or later" [Medium,Triaged] https://launchpad.net/bugs/1253009 | 14:12 |
| * xnox ponders if new pitivi uses gst1.0 or not =) | 14:12 | |
| Mirv | yes it does | 14:12 |
| xnox | good, one donw. | 14:13 |
| Mirv | s/experimental/unstable/ now, fixed | 14:14 |
| Laney | xnox: yeah that's what I thought, thanks | 14:14 |
| xnox | i'll give cherrypicking gst1.0 for pidgin another try, the patch was not that large and possible propose that if i'll get the darn thing do audio calls =) | 14:15 |
| Laney | I remember that previously I had it built but failing at runtime | 14:16 |
| xnox | same. | 14:16 |
| Laney | I remember using the test voice/video thingy in preferences | 14:16 |
| Laney | but even that didn't work | 14:16 |
| Laney | back then I didn't know pidgin 3.0 was going to take a lifetime :-) | 14:16 |
| seb128 | xnox, speaking of pitivi, https://bugs.launchpad.net/bugs/1253009 | 14:17 |
| ubot2 | Launchpad bug 1253009 in pitivi (Baltix) "[FFe] Please sync latest upstream release (0.9x) from Debian unstable - Pitivi developers recommends to use 0.92 or later" [Medium,Triaged] | 14:17 |
| xnox | i should try out just pidgin 3.0 and see if that at all works first for audio/video =) | 14:17 |
| seb128 | xnox, is pidgin3 a thing or is that the work in progress for the next/gtk3 version? | 14:25 |
| xnox | seb128: the hg tip of their mainline development repository has options to compile with gtk3/gst1.0 merged sometime between 2012 and 2013 | 14:26 |
| ChrisTownsend | seb128: FYI, I have a fix for the Gtk scrolling issues in Compiz. | 14:27 |
| tedg | Laney, Can you pastebin your url-dispatcher* upstart logs to see if there's anything in those? | 14:30 |
| seb128 | my dir is drw------- from april 2 | 14:31 |
| Laney | ffs | 14:31 |
| Laney | I keep getting logind into a hung state | 14:31 |
| seb128 | tedg, my logs have ** (process:29048): WARNING **: Unable to open URL database | 14:31 |
| seb128 | that's url-dispatcher.log | 14:32 |
| Laney | makes SSHing to the desktop from this laptop take ages | 14:32 |
| Laney | http://paste.ubuntu.com/7198908/ | 14:32 |
| Laney | tedg: ^ | 14:32 |
| Laney | Some of those are from the <app>-dispatcher | 14:33 |
| tedg | Laney, That's the url-dispatcher.log or from the -refresh and -update jobs as well? | 14:34 |
| Laney | tedg: -update ones too, don't have any -refresh | 14:36 |
| tedg | Hmm, bother. Was hoping it was them :-) | 14:38 |
| Laney | seb128: do you have any files inside the directory? | 14:39 |
| seb128 | no | 14:39 |
| seb128 | it's empty | 14:39 |
| Laney | ok | 14:40 |
| tedg | Throw out an idea, perhaps it got created somehow previously? | 14:47 |
| tedg | i.e. it's not being created by the current code | 14:47 |
| Laney | That's what I'm guessing | 14:48 |
| Laney | Don't know this area well enough to suggest what | 14:48 |
| tedg | Thinking about putting a pre-start hook to detect the case and delete the dir. | 14:52 |
| tedg | It'd "solve" the problem for most. | 14:52 |
| tedg | Perhaps we could pull a recoverable error as well, get more data. | 14:53 |
| Laney | I was expecting you'd chmod it in code, but whatever works | 14:53 |
| tedg | Hoping it's a temporary fix. | 14:56 |
| === m_conley_away is now known as m_conley | ||
| tedg | Laney, Just to be curious, what's your umask set to? Anything funky? | 15:26 |
| seb128 | tedg, stop looking for weird configs, that error tops e.u.c today, it's not a weird config one | 15:27 |
| seb128 | tedg, and my umask is not tweaked | 15:28 |
| seb128 | $ umask | 15:28 |
| seb128 | 0002 | 15:28 |
| tedg | seb128, Hmm, I just don't see anything standard doing that. Thought it was click, but it's calling g_mkdir_with_parents(777) | 15:28 |
| seb128 | btw I can't reproduce, removing the dir and running the commands from the upstart job leads to correct permissions | 15:29 |
| Trevinho | seb128: hi, bregma told me that you had issue witht the force-quit dialog thing | 15:29 |
| seb128 | Trevinho, hey, yes, I'm getting both the old and new dialogs | 15:30 |
| Trevinho | seb128: it looks weird to me btw... are you sure that for some reason your unity decor plugin is still active (and thus gtk-window-decorator) | 15:30 |
| Trevinho | seb128: it can't happen on new installs at all | 15:30 |
| Trevinho | unless someone don't run gtk-window-decorator | 15:30 |
| seb128 | I don't know | 15:30 |
| seb128 | how do I check? | 15:30 |
| Trevinho | seb128: check ccsm for active plugins or gsettings on unity profile | 15:30 |
| Trevinho | seb128: also ps aux |grep decorator | 15:31 |
| seb128 | tedg, btw, the the .cache/url-dispatcher timestamp doesn't match any login, so it seems like that directory was not created at login... what else could create it? | 15:31 |
| tedg | seb128, That's kinda why I was thinking click. But looking through the click code, it seems pretty clean. | 15:32 |
| tedg | Nothing fancy there. | 15:32 |
| tedg | I thought perhaps I created the dir, but it seems to work fine on a clean install. | 15:33 |
| seb128 | Trevinho, /usr/bin/gtk-window-decorator is running | 15:33 |
| Trevinho | seb128: yeah, that's why you get it, but it should not run | 15:34 |
| Trevinho | there's something that runs it | 15:34 |
| Trevinho | seb128: and it's not in default config | 15:34 |
| Trevinho | in theory I added migration scripts to prevent that | 15:34 |
| Trevinho | but.... | 15:34 |
| Trevinho | seb128: I would actually drop compiz-gnome package from unity installations, not know for what is needed | 15:35 |
| Trevinho | seb128: I mean, we should provide a compiz-unity instead, without that process, but not sure we can do it now | 15:36 |
| Trevinho | seb128: so gsettings get org.compiz.core:/org/compiz/profiles/unity/ active-plugins shows decor for you? | 15:37 |
| seb128 | Trevinho, I mentioned it several times a month ago that you might want to drop the old decor plugin out of the default binary, to clean the gtk2 depends... | 15:37 |
| seb128 | but you guys ignored me :p | 15:37 |
| Trevinho | seb128: oh, I actually didn't read any of your mentions.... maybe I wasn't directly pinged :) | 15:37 |
| seb128 | Trevinho, yes it does | 15:38 |
| Trevinho | but, I thought we couldn't not to break flashback sessions | 15:38 |
| Trevinho | seb128: you also get the decor plugin mentioned on lsof -p $(pidof compiz)| grep decor ? | 15:39 |
| seb128 | Trevinho, well, you could move the plugin to extra and have those session depends on that | 15:39 |
| Trevinho | seb128: is that something we're still in time to do? | 15:40 |
| seb128 | yes, it's loaded | 15:40 |
| seb128 | $ lsof -p $(pidof compiz)| grep decor | 15:40 |
| seb128 | compiz 29394 seb128 mem REG 8,1 174288 1978581 /usr/lib/compiz/libdecor.so | 15:40 |
| Trevinho | seb128: mh, weird as it should be discarded, but still... | 15:40 |
| Trevinho | now that's why you get that | 15:40 |
| Trevinho | wondering why the migration script didn't work for you | 15:40 |
| Trevinho | it's in /usr/share/session-migration/scripts/00_remove_decor_in_unity_session.py | 15:41 |
| Trevinho | seb128: if you run it now, (a part from a probabile crash), does it work? | 15:41 |
| seb128 | Trevinho, I might have re-enabled it manually when didrocks was testing/having his decoration pixmap issues | 15:42 |
| Trevinho | seb128: ah, i see. | 15:43 |
| didrocks | hum, we do use as well upstart for session-migration on desktop? | 15:43 |
| seb128 | yes | 15:43 |
| didrocks | ok | 15:43 |
| * didrocks gets out his theory for some failing cases | 15:43 | |
| seb128 | Trevinho, I just ran it, compiz segfaulted but the decor plugin is still in my gsettings config | 15:44 |
| Trevinho | didrocks, seb128: is the "gsettings values not being updated without using dconf" issue still there? | 15:44 |
| Trevinho | seb128: well... it shouldn't really... I mean if you call gsettings from that cmd you get it, but.... if you run the migration script again it should say it's not | 15:45 |
| Trevinho | seb128: so maybe the shell call is missing something and loading the default config instead | 15:45 |
| dbarth | charles: ping? | 15:45 |
| Trevinho | seb128: while your lsof of compiz should say none about decor | 15:45 |
| seb128 | Trevinho, oh, indeed, lsof doesn't have it anymore now | 15:45 |
| seb128 | weird | 15:45 |
| Trevinho | seb128: yeah, maybe the call is just wrong :) | 15:46 |
| Trevinho | while the python call is correct | 15:46 |
| Trevinho | seb128: aaaaaaaaaaaaah, the right path is gsettings get org.compiz.core:/org/compiz/profiles/unity/plugins/core/ active-plugins | 15:49 |
| didrocks | Trevinho: not sure what issue you are talking about :p | 15:49 |
| Trevinho | sorry :P | 15:49 |
| Trevinho | didrocks: there was an issue that using gsettings.set_blah_blah("...") on migration script, then the scripts where not actually saved | 15:49 |
| Trevinho | so most of migration scripts actually used dconf to write the settings value | 15:50 |
| didrocks | Trevinho: nothing changed AFAIK | 15:50 |
| Trevinho | seb128: can we retry a landing for the forcequit then? | 15:51 |
| seb128 | Trevinho, sure, still the upgrade path seems flaky, we should ensure the decor plugin is unloaded, maybe make it conflicts with unity? | 15:51 |
| Trevinho | seb128: it's already conflicting with it and there are multiple migration scripts | 15:52 |
| Trevinho | seb128: the one you launched + the ones in ccsm | 15:52 |
| seb128 | k, dunno how I ended up with both though | 15:52 |
| seb128 | I still think we should move libdecor.so to another binary which we don't install by default | 15:53 |
| seb128 | to clean out some of the binary depends it creates | 15:53 |
| Trevinho | seb128: yeah, I agree, if we can still do it, I can figure it out | 15:53 |
| seb128 | Trevinho, seems it might be a bit late, which is a shame | 15:53 |
| Trevinho | in some spare time (in theory I'm in holiday now, but the weather is pretty bad here in Barcelona right now :°() | 15:54 |
| seb128 | Trevinho, enjoy your holidays! | 15:54 |
| seb128 | sorry I didn't know you were off work ;-) | 15:54 |
| Trevinho | seb128: yeah, I tried to, but I ended to be completely wet :D | 15:54 |
| Trevinho | seb128: so... since this is still a "quite busy" period I've still some time to fix things | 15:55 |
| l3on | Hi all ! | 16:10 |
| l3on | I'm going to remove this patch in nautilus http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/trusty/nautilus/trusty/view/head:/debian/patches/ubuntu_titlebar_css.patch | 16:10 |
| l3on | and apply changes to themes. darkxst around ? | 16:10 |
| seb128 | l3on, he's likely sleeping at this time, he's in .au | 16:11 |
| seb128 | l3on, also I guess you want to say "you would like to suggest replacing it", right? | 16:11 |
| seb128 | l3on, because you don't have commit access so I don't see how you are going to do it without asking us first... | 16:12 |
| l3on | seb128, yep. Right. That patch is wrong. You're are going to hardcode style properties in nautilus code. What's happen if someone use gnomeshell with Ambiance/Radiance themes ? | 16:14 |
| seb128 | l3on, no idea, darkxst did it this way because it had issues doing it in the theme iirc | 16:15 |
| seb128 | l3on, if you know how to do please open a bug with patches for review and subscribe ubuntu-sponsors | 16:15 |
| l3on | at same time patch has border-radius set to 0 that causes also on unity on 14.04 | 16:15 |
| l3on | yep, seb128.. working on. | 16:16 |
| mhr3 | didrocks, ping? quick question - when bumping soversion, do we have to change the pkg name too? ie soversion is going from 0->1, do we have to rename libunity-scopes0 to 1? | 16:16 |
| seb128 | thanks | 16:16 |
| seb128 | mhr3, yes | 16:16 |
| mhr3 | seb128, and could we actually jump back at some point from 33 to 1? | 16:18 |
| seb128 | mhr3, that would be backward ;-) | 16:18 |
| seb128 | why would anyone do that? | 16:18 |
| mhr3 | we're forced to release with pretty much every commit | 16:19 |
| seb128 | mhr3, you could, at the risk of screwing things still using the real/previous "1" (assuming your new "1" has a different abi) | 16:19 |
| mhr3 | and we haven't really reached official 1.0 | 16:19 |
| seb128 | mhr3, one days you guys are going to learn to stop changing the apis in incompatibles ways right? | 16:20 |
| mhr3 | seb128, unlikely | 16:21 |
| mhr3 | :P | 16:21 |
| seb128 | k, so you just keep bumping the soname with every upload | 16:21 |
| seb128 | and don't bother about going back to "1" one day ;-) | 16:21 |
| mhr3 | just feels wrong to release a lib with soname 33 as the first real release | 16:21 |
| seb128 | yeah, well go back to "1" | 16:22 |
| seb128 | I doubt you are going to still have users of the previous abi you called "1" by then | 16:22 |
| seb128 | mhr3, or just skip "1", go from "0" to "2" | 16:22 |
| seb128 | mhr3, you can also rename the lib to "libunity-scopes-dont-learn-what-stable-abi-means-yet.so.<n>" | 16:23 |
| seb128 | mhr3, and rename the lib the day you do learn ;-) | 16:23 |
| kenvandine | haha | 16:23 |
| mhr3 | seb128, heh, but yes, exactly, at that point we're pretty sure noone is using anything <33 | 16:24 |
| seb128 | mhr3, you guys are writing a book on "how to not do thing" right? | 16:24 |
| mhr3 | seb128, actually it's called "how ubuntu developers are forcing us to do the wrong things" :P | 16:25 |
| seb128 | mhr3, you are the ones are not able to design a proper api and stick to it... ;-) | 16:25 |
| seb128 | we just make sure you don't punch users in the face by doing incompatible changes without telling your users about those | 16:26 |
| happyaron | seb128: I reported that bug, :( | 16:27 |
| happyaron | seb128: for the firefox default search engine | 16:27 |
| happyaron | seems it's bug 800304, but I'm not very sure. | 16:27 |
| ubot2 | Launchpad bug 800304 in firefox (Ubuntu) "browser.search.defaultenginename does not work from distribution.ini" [Low,Triaged] https://launchpad.net/bugs/800304 | 16:27 |
| seb128 | happyaron, right, I'm unsure it's a ubuntu-defaults-builder issue | 16:34 |
| seb128 | seems rather a firefox one | 16:34 |
| seb128 | chrisccoulson might be able to help you, at least to reply to questions | 16:34 |
| happyaron | ok | 16:35 |
| Laney | seb128: turned out to be an upstart bug ;-) | 16:53 |
| seb128 | Laney, which one? the url-dispatcher permission thing? | 16:53 |
| Laney | seb128: yep | 16:55 |
| Laney | echo "manual\nexec sh -c umask" > ~/.config/upstart/test.conf; start test; sudo telinit u; start test; cat ~/.cache/upstart/test.log | 16:56 |
| Laney | or something like that, should show the difference before and after | 16:57 |
| seb128 | no cookie for upstart! | 16:58 |
| seb128 | Laney, where did you discuss/debug it? | 16:58 |
| Laney | #-touch | 16:58 |
| seb128 | shrug, I close that one by error again ;-) | 16:58 |
| seb128 | glad to see you figured it out though | 16:58 |
| Laney | touch hater | 16:58 |
| seb128 | haha, you got me! | 16:59 |
| Laney | yeah, jod h got pinged so he should take a look | 16:59 |
| * seb128 wears his "desktop4ever" badge | 16:59 | |
| ogra_ | seb128, dude, its the future ! | 16:59 |
| Laney | meanwhile ted ought to add some defensive code there ;-) | 16:59 |
| === alan_g is now known as alan_g|EOD | ||
| Laney | chmod it back and fix up the umask | 16:59 |
| seb128 | ogra_, the desktop is the futur, I know! | 16:59 |
| ogra_ | lol | 16:59 |
| * Laney → climbing | 17:08 | |
| Laney | byesie bye | 17:08 |
| * Laney forgot to patch pilot today :( | 17:09 | |
| * Laney moves it to tomorrow | 17:09 | |
| seb128 | Laney, have fun! | 17:13 |
| seb128 | bregma, andyrock: what's the status of the screenlocking/screen dpms issue? did you guys talk with robert_ancell to figure out what to do next (adding interfaces to unity?) and who is doing it? | 17:29 |
| andyrock | i'm adding the interfaces | 17:29 |
| andyrock | and the fading effect | 17:29 |
| seb128 | k | 17:29 |
| bregma | andyrock, is Trevinho there too? | 17:30 |
| seb128 | did you/somebody let robert_ancell know? | 17:30 |
| bregma | andyrock, do you think you'll have something to propose today? | 17:30 |
| seb128 | bregma, Trevinho said earlier that in theory is on vac day today? | 17:30 |
| bregma | seb128, yeah, but he's planning to visit andyrock at some point, with his laptop :) | 17:31 |
| seb128 | ;-) | 17:31 |
| seb128 | bregma, Trevinho, andyrock: anyway, please keep robert_ancell updated on what he happening, he started looking at those issues the other day, I would prefer avoiding duplication (e.g not having him continuing while you guys also work on the same thing) | 17:32 |
| andyrock | bregma, he was here before | 17:53 |
| andyrock | ok so to keep you guys updated, i've already implemented the fade effect | 17:54 |
| andyrock | the screen now turns off | 17:54 |
| andyrock | but sometimes stop working | 17:54 |
| andyrock | seb128, ^^^ | 17:54 |
| seb128 | andyrock, thanks | 17:57 |
| seb128 | andyrock, stop working, like doesn't turn back on? or is unity not liking it? | 17:57 |
| === m_conley is now known as m_conley_away | ||
| andyrock | seb128, just unity emits the ActiveChanged signal | 17:57 |
| andyrock | but power-manager fails to turn off the screen | 17:58 |
| seb128 | weird | 17:59 |
| seb128 | well, working with some bug is better than not working ;-) | 17:59 |
| seb128 | so maybe submit that for review so things keep moving | 17:59 |
| andyrock | well just need to finish it | 18:04 |
| andyrock | it's under unity-team | 18:04 |
| andyrock | so others can work on it | 18:04 |
| elfy | seb128: do you have a couple of minutes? | 18:05 |
| camako | I'm installing ubuntu on macbook pro.. Basically I got it running, but wifi drivers missing... | 18:31 |
| camako | MBP model-id=10,1 | 18:31 |
| camako | ubuntu 12.04 | 18:31 |
| camako | this mac has no wired interface, so I'll have to download to flash and install from there | 18:32 |
| camako | What packages do I need? | 18:33 |
| czajkowski | aloha | 18:33 |
| czajkowski | so today I keep getting https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1202754 but it's not adding any logs to it | 18:35 |
| ubot2 | Launchpad bug 1202754 in update-manager (Ubuntu Saucy) "update-manager crashed with SystemExit in exit(): 0" [High,Confirmed] | 18:35 |
| czajkowski | am on Trusty | 18:35 |
| czajkowski | which does run very nicely | 18:35 |
| czajkowski | thank you :) | 18:35 |
| robert_ancell | larsu, do you have a link to the XDG_CURRENT_SESSION change? | 18:49 |
| robert_ancell | larsu, note we are currently using XDG_CURRENT_DESKTOP | 18:49 |
| seb128 | robert_ancell, XDG_CURRENT_SESSION? | 19:01 |
| seb128 | robert_ancell, hey btw ;-) | 19:02 |
| robert_ancell | seb128, hello | 19:02 |
| seb128 | robert_ancell, I wonder if that's a typo, https://bugzilla.gnome.org/show_bug.cgi?id=727546 has XDG_CURRENT_DESKTOP | 19:02 |
| ubot2 | Gnome bug 727546 in general "gdm-session: support XDG_CURRENT_DESKTOP" [Normal,Unconfirmed] | 19:02 |
| seb128 | robert_ancell, yeah, seems like a typo in the commit message | 19:03 |
| robert_ancell | seb128, I thought as much | 19:03 |
| seb128 | czajkowski, hey, how do you trigger it? | 19:03 |
| seb128 | elfy, hey, I'm here now, better to leave some context when you have a question | 19:04 |
| elfy | seb128: yea - sorry - bug 1284635 | 19:04 |
| ubot2 | Launchpad bug 1284635 in ibus (Ubuntu Trusty) "Keyboard layout changes after login" [High,Confirmed] https://launchpad.net/bugs/1284635 | 19:04 |
| elfy | we're getting really itchy and worried about that in Xubuntu land | 19:05 |
| elfy | I talked to happyaron - who said it's on his list, but he'd got higher priority things atm | 19:05 |
| seb128 | robert_ancell, thanks for investigating the lockscreen/dpms issues btw, andyrocks said earlier that he's working on making unity emits the ActiveChanged signal | 19:05 |
| robert_ancell | seb128, cool | 19:06 |
| elfy | RC is next week and we're stuck with a system that loses keyboard layout for anyone not using a US layout ill they go and change it | 19:06 |
| seb128 | elfy, I guess if I say 'patches are welcome' it doesn't help you? | 19:06 |
| elfy | lol | 19:06 |
| seb128 | but reality is that we have our issues as well, and resources/hours in the day are what they are | 19:07 |
| seb128 | so if xfce has issues they probably need to help getting those resolved | 19:07 |
| seb128 | we help when we can but we are currently struggling with our lot of release issues and have no spare room to help you guys | 19:07 |
| elfy | mmm - our issue atm to be honest is that there was no problem till the update to the latest version | 19:08 |
| seb128 | well, we are not upstream for ibus | 19:09 |
| elfy | I understand there's only 60 minutes in an hour - if you're really lucky | 19:09 |
| seb128 | the update also resolved other issues | 19:09 |
| seb128 | we also don't see the issue you describe on GNOME or Unity | 19:10 |
| elfy | I know | 19:10 |
| seb128 | so it's not ibus being totally buggy, just an interaction issue with xfce | 19:10 |
| knome | seb128, it's not a long time ago that canonical wanted to set a policy that makes developers fix what they broke, even if it didn't affect them... | 19:10 |
| knome | seb128, if something is landed late in the cycle, even if it was "just" a new version from upstream, i expect the developer(s) in question to be responsible for at least looking at the issue | 19:11 |
| seb128 | knome, what if the update came from Debian through a sync? | 19:11 |
| knome | i don't know | 19:12 |
| seb128 | there we go | 19:12 |
| knome | does that prove it's our problem now either? | 19:12 |
| seb128 | no, there is not blame/being "$flavor" problem there | 19:13 |
| seb128 | it's just one of a long list of issues to be resolved for release with everybody being busy | 19:13 |
| knome | well, my point of view is: | 19:13 |
| knome | if it came from a non-automatical debian sync (i don't know if that's the case), the one who did the sync should be at least somewhat responsive to questions about the issue | 19:15 |
| knome | if it's an automatical sync... again, i don't know | 19:15 |
| elfy | and I don't know enough to make a comment | 19:15 |
| knome | i'm not a very technical person either, so i probably don't understand *all* the technical implications | 19:16 |
| knome | would it be completely out of question to revert back to the earlier version? | 19:16 |
| seb128 | yes, it is | 19:16 |
| seb128 | the update fixes as important issues on other flavors | 19:16 |
| knome | okay | 19:16 |
| seb128 | so you would just trade a set of issues for a similar one | 19:17 |
| seb128 | but living on an upstream-unmaintained version instead | 19:17 |
| knome | right, i acknowledge and agree the upstream version is better | 19:17 |
| seb128 | the way forward is to fix the issue on the current one, not to dig ourself in staying on a version that is not maintained and has a bugs as well | 19:17 |
| knome | do you know if it's a manual or automatic sync? | 19:17 |
| seb128 | it's a manual merge from happyaron | 19:18 |
| seb128 | who is the who assigned to the bug | 19:18 |
| knome | right, | 19:18 |
| knome | in that case i would expect at least a quick look from him to our issue | 19:18 |
| elfy | seb128: aron xu | 19:18 |
| seb128 | but he's having higher priority work assignments | 19:18 |
| elfy | sorry - catching up ... | 19:18 |
| knome | sure, i understand... | 19:18 |
| seb128 | he had a look from what I know | 19:18 |
| seb128 | it just needs work | 19:18 |
| knome | but it doesn't help us much, or fit to the "the one who breaks it, fixes it" policy | 19:19 |
| seb128 | but he's busy with paid-work priorities and not having free slots to help there | 19:19 |
| seb128 | not a lot we can do... | 19:19 |
| knome | and i do acknowledge he didn't necessarily "break" it... | 19:19 |
| knome | just to be clear, | 19:19 |
| seb128 | right | 19:19 |
| seb128 | that principle is right | 19:19 |
| seb128 | but real world is what it is | 19:19 |
| knome | if a canonical employee breaks something, but has other more high priority work items, the policy doesn't apply? | 19:20 |
| seb128 | lol | 19:20 |
| seb128 | no, quite the contrary | 19:20 |
| knome | how so? | 19:20 |
| knome | he's working on the more high priority things now | 19:20 |
| seb128 | if some community person do a sync and never show up again, what do you do? | 19:20 |
| seb128 | hire people to find that person and force him into work? | 19:20 |
| knome | obviously there are different realities to volunteer people | 19:21 |
| seb128 | you have probably more chance of Canonical employees to be hold responsible and help there | 19:21 |
| knome | yes, i acknowledge that | 19:21 |
| seb128 | well, then your comment doesn't make sense | 19:21 |
| seb128 | it's not because he's a Canoncial employee that he gets a free pass | 19:21 |
| knome | no, | 19:21 |
| knome | i'm not saying that | 19:21 |
| knome | let me rephrase | 19:22 |
| seb128 | we all have priority lists | 19:22 |
| seb128 | usually paid job comes first, because we need to pay for food and stuff | 19:22 |
| knome | sure... | 19:22 |
| seb128 | then we help as free time allows | 19:22 |
| seb128 | same apply here | 19:22 |
| knome | well, was the ibus sync paid work? | 19:23 |
| knome | or voluntary? | 19:23 |
| seb128 | it ends up at "there are more things that need to be fixed than hours in a day" | 19:23 |
| knome | i understand | 19:23 |
| seb128 | it was on work time (I guess) | 19:23 |
| knome | but i also see it as: things that are broken 'only' on flavors are always lower priority issues than issues with ubuntu desktop | 19:24 |
| seb128 | but that work solved issues | 19:24 |
| seb128 | where number-of-users-benefiting-from-the-change > users-having-issues probably | 19:24 |
| seb128 | (that's assuming that Ubuntu has more users than Xubuntu) | 19:24 |
| seb128 | well | 19:25 |
| knome | yes, so in reality it is as i said | 19:25 |
| knome | i'm not saying that's not how it shouldn't be, i completely empathise the canonical employees.. | 19:25 |
| knome | but that's how it is | 19:25 |
| seb128 | the priority is on doing what benefits most users? | 19:25 |
| seb128 | sure it is | 19:25 |
| seb128 | would you recommend on screwing most users for the benefit of a smaller part? | 19:25 |
| knome | and basically it means that for example, this issue will very unlikely be fixed | 19:25 |
| knome | no. | 19:26 |
| knome | i just said i understand why it is as it is | 19:26 |
| seb128 | right, and I understand your frustration | 19:26 |
| knome | please understand that i'm just trying to assess what the situation for the xubuntu team here is | 19:26 |
| seb128 | but the world being what it is I don't have an answer that would make you happy I think | 19:26 |
| knome | and it looks like: nobody has time and this fixed our issues so we will leave it as it | 19:27 |
| knome | *is | 19:27 |
| seb128 | right, I understand that, and I'm on that side of the world as well quite often | 19:27 |
| seb128 | reporting a bug on GTK that doesn't impact GNOME ends up the same way | 19:27 |
| knome | which leads us to the situation where we either try to find a ibus developer that understands the issue and is able to fix it in 2 weeks, or, drop ibus and make xubuntu unusable for many people | 19:27 |
| knome | of which the latter is, unfortunately, more realistic | 19:28 |
| seb128 | yeah, I sadly don't have good alternative to suggest you | 19:29 |
| knome | i understand | 19:29 |
| seb128 | happyaron can probably have a look, but I don't know if that's going to be before release | 19:29 |
| knome | is there anybody else that could possibly at least look at the issue ASAP? | 19:29 |
| knome | and if not... | 19:29 |
| seb128 | so the issue should be at least fixable in a SRU/for LTS .1 | 19:29 |
| knome | the other question is: | 19:29 |
| knome | does it make sense to sync things this late manually, if it's unrealistic to fix regressions the new versions introduce | 19:30 |
| seb128 | the update was made because it fixes issues as said | 19:30 |
| knome | (more of a rhetorical question, yes i know it's a balance with fixing other issues as well) | 19:31 |
| seb128 | there are probably more users happy about the update than ones unhappy | 19:31 |
| knome | too bad there isn't any way to measure that even after the release | 19:31 |
| seb128 | you just happen to be in the unlucky minority | 19:31 |
| elfy | well for us - it seems to break for everyone not using a US layout | 19:31 |
| seb128 | well, on xubuntu | 19:31 |
| seb128 | we got no report on GNOME or Unity | 19:31 |
| seb128 | which have a larger userbase than XFCE | 19:32 |
| brainwash | we don't use gnome-settings-daemon :( | 19:32 |
| seb128 | (which is not an excuse for creating issues for XFCE, but if there is a call to be made) | 19:32 |
| brainwash | or the new fork | 19:32 |
| seb128 | right | 19:32 |
| elfy | seb128: I checked during the time I was looking at it - it affects us, studio (obviously) lubuntu, mythbuntu | 19:32 |
| seb128 | I'm just saying that reverting would hurt more users than it would help | 19:33 |
| seb128 | so that's not a position I can defend | 19:33 |
| seb128 | well, even without comparing userbases | 19:33 |
| elfy | I realise that :) | 19:33 |
| seb128 | the way forward is not to live in the past, it's to fix the issues | 19:33 |
| elfy | and that ;) | 19:34 |
| brainwash | we basically just need someone who can analysis the situation and maybe explain why it fails to pick the system kb layout and falls back to en_US | 19:34 |
| seb128 | and to reply to your questions, no it's not "unrealistic to fix regressions the new versions introduce" | 19:34 |
| seb128 | it just happens it took a while for the issue to be flagged, and as we come close from release other important issues are raising priority as well | 19:34 |
| knome | so... it's realistic that somebody fixes the bug before release? | 19:34 |
| seb128 | so we would probably have had cycles to fix it earlier | 19:35 |
| seb128 | but it's getting late and everybody is on "we need to fix those for release" and swamped with work | 19:35 |
| knome | yep. | 19:35 |
| knome | which is why i was asking about syncing things late in the cycle | 19:36 |
| seb128 | well, xubuntu doesn't have anybody wanting to do debugging? | 19:36 |
| knome | we don't have anybody who's able to | 19:36 |
| seb128 | e.g you rely on the Ubuntu/Unity team or are screwed? | 19:36 |
| knome | we can do debugging, and we can run as much testing as needed | 19:36 |
| knome | but we are very short on developer resources this cycle | 19:36 |
| knome | so what we can do is limited | 19:36 |
| seb128 | "that late" | 19:36 |
| seb128 | the new ibus was synced a month ago | 19:36 |
| knome | we've even done most of our uploads via the sponsors queue | 19:36 |
| knome | yes, i'm not saying we are perfect either | 19:37 |
| seb128 | the issue was not when the update happened | 19:37 |
| seb128 | it's that it took you guys a while to flag it as an issue | 19:37 |
| elfy | it took a while to realise it was ibus tbh | 19:37 |
| seb128 | you should have users running the devel release and telling you when thing go wrong | 19:38 |
| knome | seb128, we didn't want to tip the bug to the desktop team if we didn't know what was causing it, so we did our own debugging | 19:38 |
| seb128 | so you have a few days of updates to look through | 19:38 |
| knome | the bug is filed in february | 19:38 |
| knome | we've been looking at it since | 19:38 |
| elfy | seb128: we do - I filed it a few days after the update I believe | 19:39 |
| knome | it was escalated to the desktop team when we knew it's likely to be ibus that's causing the bug | 19:39 |
| seb128 | well, that's unfortunate that it didn't get raised up earlier | 19:39 |
| seb128 | anyway, no point holding blames | 19:39 |
| seb128 | not sure what to suggest doing next | 19:40 |
| knome | well the point is, we do try to do our best to make sure we don't point the desktop team to bugs that are caused by, say, xfce packages | 19:40 |
| seb128 | do you have anyone that could test if the same issue is happening atm on Debian (which has the same ibus version)? | 19:40 |
| knome | we can make that happen | 19:40 |
| seb128 | that would help | 19:41 |
| seb128 | or maybe try installing the debian version on xubuntu and see if it has the same issue | 19:41 |
| seb128 | that would rule in/out an Ubuntu specific issue | 19:41 |
| knome | i tried, but it's not that easy, you would need to downgrade a lot of dependencies | 19:41 |
| seb128 | you can probably take the debian source and build it on Ubuntu | 19:41 |
| knome | hmm, sorry, that's not what i tried. | 19:42 |
| knome | i tried to use the version before the sync | 19:42 |
| seb128 | oh | 19:42 |
| seb128 | no, that's not useful, we know that this one works | 19:42 |
| seb128 | what would be useful is to try the version from Debian to see if the issue comes from one of our patches | 19:42 |
| seb128 | that would nail down the problem | 19:42 |
| knome | can you point me to it, and i'll get it tested now | 19:42 |
| seb128 | http://packages.qa.debian.org/i/ibus.html | 19:42 |
| seb128 | http://ftp.iut-bm.univ-fcomte.fr/debian/pool/main/i/ibus/ | 19:43 |
| knome | thanks | 19:43 |
| seb128 | I guess those should install fine on trusty, but I didn't try | 19:43 |
| knome | can i be in touch with you on this in the future? | 19:43 |
| seb128 | sure, that's an open channel ;-) | 19:44 |
| seb128 | you can find me (and others) pretty much any work day (less likely during the w.e) | 19:44 |
| elfy | thanks for the help seb128 | 19:44 |
| seb128 | yw! | 19:44 |
| knome | well i guess the real question is if you are willing to hear back on it | 19:44 |
| seb128 | let me know how that goes | 19:44 |
| knome | or if you will just dismiss (which i'd also kind of understand...) | 19:45 |
| seb128 | yes, I can't promise we can spend days on debugging | 19:45 |
| seb128 | but I'm happing trying to help with the resources we have | 19:45 |
| knome | thanks | 19:45 |
| seb128 | yw! | 19:45 |
| elfy | yep - thanks | 19:45 |
| seb128 | if you find out that the issue is in one of our changes I can try to make we get to the bottom of the issue before release | 19:46 |
| seb128 | if the issue is in Debian/upstream it might be more difficult | 19:46 |
| seb128 | to make sure* | 19:47 |
| knome | yes, obviously | 19:47 |
| ochosi | hey Laney | 21:48 |
| ochosi | Laney: just in case you're still around, i wanted to quickly inquire about the recent change in the indicator-sound desktop file that brainwash asked about before | 21:49 |
| ochosi | it's not really clear to me why the "Hidden=True" setting is being added now (i mean: a bit late in the cycle). or, why it is added to some indicators only | 21:50 |
| ochosi | so i just wanted to know what's the plan with this | 21:50 |
| ochosi | Laney: if i should talk to someone else about this, that's also fine (just let me know who to bug :)) | 22:03 |
| Laney | ochosi: hello | 22:04 |
| Laney | tedg did it, but I can explain | 22:04 |
| Laney | The desktop files in the /usr/share/.../xdg (whatever it is) directory turn off the ones in /etc/xdg/autostart/ when you're using upstart user sessions | 22:04 |
| Laney | they've been added for indicators that have been converted to upstart | 22:05 |
| ochosi | right | 22:07 |
| ochosi | but why the Hidden=True? | 22:08 |
| ochosi | i mean is it really necessary? | 22:08 |
| Laney | That's exactly how you turn off an autostart file | 22:09 |
| Laney | So yes | 22:09 |
| ochosi | humm, i see | 22:09 |
| ochosi | why hasnt that been done for all indicators then? | 22:09 |
| Laney | That directory is only relevant when you use upstart user sessions | 22:09 |
| ochosi | or has the change for the others simply not landed yet? | 22:09 |
| Laney | it has, but maybe they haven't all been uploaded yet | 22:09 |
| ochosi | ok | 22:09 |
| ochosi | well we've started seeing this so late now, it's quite meh to mess with our session so late in the cycle | 22:10 |
| ochosi | and while i was monitoring some of the pending changes to the indicators (for gtk-greeter) i didn't see this one coming | 22:10 |
| ochosi | i mean: it's quite meh for us to mess with our session... | 22:11 |
| ochosi | we don't even have uploaders around atm | 22:11 |
| Laney | I gave a few ways you can fix this earlier | 22:11 |
| ochosi | so changes have to go through the sponsors queue | 22:11 |
| ochosi | right, i guess we can launch all installed indicators in the session | 22:12 |
| ochosi | brainwash told me about that | 22:12 |
| ochosi | but then we'd need to have XFCE stripped from the OnlyShowIn= line | 22:13 |
| ochosi | cause otherwise the non-functional desktop-files show up in our session's autostart manager | 22:13 |
| ochosi | would that be a feasible change for you guys? | 22:14 |
| ochosi | (i.e. we use "loginctl emit indicator-services-start" and -end in our session and the desktop files don't show anymore in xubuntu) | 22:14 |
| Laney | That UI shouldn't be ignoring NoDisplay | 22:15 |
| ochosi | right, but we can't fix that right now | 22:15 |
| Laney | You can't drop OnlyShowIn if you want to use upstart otherwise it won't work | 22:15 |
| Laney | Well, that's what tells the XDG autostart files when to be considered | 22:16 |
| ochosi | hmkay | 22:16 |
| ochosi | last question, how would blacklisting work? | 22:16 |
| ochosi | (that's another option he mentioned) | 22:16 |
| Laney | dunno what you mean | 22:17 |
| Laney | If you mean not using upstart then that's /etc/upstart-xsessions | 22:17 |
| ochosi | "blacklist the upstart autostart launcher" is what he said | 22:17 |
| ochosi | right, but i have no idea what that will result in for us | 22:17 |
| ochosi | that seems like a rather drastic and inconvenient change | 22:18 |
| ochosi | not even sure whether indicators would still work for us then | 22:18 |
| Laney | they ought to | 22:18 |
| brainwash | blacklist might be the wrong term | 22:18 |
| brainwash | override the conf file | 22:19 |
| ochosi | what conf file? | 22:19 |
| brainwash | echo manual ... | 22:19 |
| Laney | That's for local overrides | 22:19 |
| brainwash | .config/upstart/indicator-xyz.conf | 22:19 |
| ochosi | brainwash: that doesn't sound like a very flexible solution though | 22:20 |
| brainwash | it's not | 22:21 |
| ochosi | so i'd rather rule that one out | 22:21 |
| brainwash | just a way to control the indicator started by upstart | 22:21 |
| ochosi | so we have two left: | 22:21 |
| brainwash | on the user side | 22:21 |
| ochosi | 1) don't use upstart user sessions anymore | 22:22 |
| ochosi | 2) start indicators in the session with the signal | 22:22 |
| ochosi | and i guess with 2) we also have to live with them showing in the session manager for now | 22:22 |
| brainwash | or patch xfce4-session? | 22:24 |
| brainwash | hack :) | 22:24 |
| ochosi | well we have to patch our session anyway... | 22:25 |
| === thumper is now known as thumper-gym | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
