[00:03] probably some of this cpu usage is from writing 100 log messages a second [00:07] \o/ [00:11] oh, i'm sorry, 500 messages a second (!) [00:19] yes, that kinda thing made bzr slow for a whil [00:19] e [03:27] <__lucio__> we ship on INFO, DEBUG is supposed to make it slow. [03:28] <__lucio__> and if you just restarted syncdaemon after all this time, its not only deleting old items but also doing a couple of database format migrations that might take a while. [03:32] __lucio__: how do i get it back to just logging INFO? i've forgotten [03:33] <__lucio__> mwhudson, vi /home/lucio/.config/ubuntuone/logging.conf [03:34] __lucio__: huh, no file there [03:34] <__lucio__> mwhudson, vi /home/lucio/.config/ubuntuone/syncdaemon.conf ? [03:34] <__lucio__> mwhudson, you are on maverick using no ppa? [03:34] that's there, but doesn't have anything to do with logging in it [03:35] __lucio__: natty [03:35] <__lucio__> [logging] [03:35] <__lucio__> level = INFO [03:35] <__lucio__> put that in logging.conf [03:35] <__lucio__> but natty should default to INFO, ill investigate [03:36] i do remember doing something to turn on DEBUG, but i don't at all remember what [03:36] it's default to INFO now [03:36] but during the development cycle it defaults to DEBUG [03:36] <__lucio__> awesome [03:36] mwhudson: did you edit the logging.conf in /etc/xdg/ubuntuone/? [03:37] also, nightlies are always debug iirc [03:37] dobey: ah, maybe, it has "level.default = DEBUG" in there [03:38] mwhudson: what version of ubuntuone-client do you have exactly? [03:38] dobey: 1.6.0-0ubuntu1 [03:38] ok, hrmm [03:38] maybe i should just buy a new laptop :) [03:39] hrmm [03:39] it appears to be DEBUG by default still. wonder why that is [03:40] file a bug and i'll look at it on wednesday :) [03:41] back to getting some "rest and relaxation" til then :) [08:16] buenas! [08:24] 'lo mandel [08:25] duanedesign: hello :) [09:12] good morning! [09:19] Hello, good morning and welcome! [09:21] JamesTait: if you are that happy every morning at uds we are going to have a problem, I'm cracky in the mornings ;) [09:21] mandel: I've had my morning coffee and my breakfast. You should see me first thing. [09:22] haha [09:22] then we are fine :P [11:26] good morning everyone! [11:27] hi ralsina_ \o [11:28] hi karni [12:32] good morning everyone! [12:34] nessita: buenos dias! [12:35] hola mandel, how are you? [12:36] nessita: fine thx, fighting with multithreading, twisted and COM programming on windows [12:36] nessita: and you? [12:36] how are you eating habits doing? [12:37] mandel: herm... compluicated, as usual :-) [12:37] complicated* [12:40] do we know the keyborad shortcut to open the logout dialog in a desktop? [12:40] nessita: in unit? [12:40] unity* [12:40] mandel: yes [12:41] mandel: I logged in with another user and the launcher and top bar never showed up [12:42] nessita: all the ones I know are here: http://askubuntu.com/questions/28086/unity-keyboard-mouse-shortcuts [12:42] I guess you can launch the terminal and log out like that... [12:42] mandel: thanks! [12:45] mandel: logout like what in a terminal? [12:47] nessita: try logout [12:47] $ logout [12:47] works? [12:47] mandel: won't work, say 'use exit to terminate' or something [12:48] mandel: no prob, I'll kill that session [12:48] that would also work ;) [12:49] try Ctrl + Alt + backspace used to be there [12:49] but I dont longer know.. [12:49] no, tried it already :-( [12:50] I wonder who was the evil person that removed that... [12:50] bastards! [12:50] well, I need to walk the dog, he is here looking at me with an evil face, will be back later [12:50] mandel: enjoy [13:01] hey guys, discussing about desktopcouch in desktop-devel-list@gnome.org [13:01] if you want to jump in, we're talking about providing syncing of contacts via couchdb [13:06] http://mail.gnome.org/archives/desktop-devel-list/2011-April/msg00077.html <- the thread [13:06] * rodrigo_ lunch [13:45] mandel: ping? [13:51] alecu nessita thisfred mandel standup in 10' [dobey in swap, chad on loan, fagan internet-less] [13:51] ack [13:51] ack [13:51] thisfred: HOLA [13:51] Hola! [13:51] how was the vacation thisfred? [13:52] genial! [13:53] thisfred: did you eat a lot? :-) [13:53] We stayed in Palermo Soho, and took it very easy. Went to the film festival twice, ate a lot of superb food and ice cream, and walked around a lot :) [13:53] oh, picked up some extra argentinian :-) [13:53] nessita: yes, the food was *so* good! [13:54] thisfred: I'm glad [13:54] ralsina_: I'm sure it will be gone again tomorrow ;) [13:54] thisfred: come back whenever you need a refresher, you have only seen like 0.05% of the country ;-) [13:55] ralsina_: I know, I totally want to see Mendoza, and Patagonia. I'm pretty sure we will be back. [13:55] Though not flying Delta. [13:56] The flight back was the worst flight of my life. I literally did not fit in the seat, even when the guy in front did not lean his back... [13:56] anyhow, still more than worth it! [13:56] hola alecu! [13:57] hola eric! [13:57] alecu: many thanks for the Film Festival tip! We went two afternoons! [13:57] thisfred, cool! did you saw anything interesting??? [13:57] Saw some great documentaries, and a new Argentinian film, that my wife hated, but I really liked :) [13:58] alecu: las piedras, it's filmed in Tigre and B.A. [13:58] :-) [13:58] Almost nothing happens in it, but it has the best dance scene ever [13:59] oh, modern argentine cinema. "Almost nothing happens" describes 90% of it :-) [13:59] alecu: also we'd just been to Tigre the day before and took a short boat tour [13:59] ralsina_: Modern [anywhere] cinema, though :) [13:59] what's everybody fixation with story anyway! [13:59] me [14:00] me [14:00] alecu: last year I saw a movie filmed from behind a character's ear. Off focus. By Gaspar Noe. Worst thing ever. [14:00] me [14:00] ralsina_, but you've seen irreversible, right? [14:01] me [14:01] alecu: yes, I have. I felt dirty for weeks :-) [14:01] or maybe :-( [14:01] ok, nessita, go [14:01] DONE: intense debug of bug #762004 and bug #762072 (got ricardokirkner and jdo involved). Chat with alecu to define how to really solve bug #759197 and bug #764646 (alecu, I could not make your branch work properly, I need help). Attempted to propose a blind fix for bug #750309 but the issue is not fixed. [14:01] TODO: ask ralsina help with bug #750309, keep fighting u1-servers tests on natty, keep sending a branch to PQM [14:01] BLOCKED: no that much [14:01] NEXT: thisfred [14:01] Launchpad bug 762004 in ubuntuone-servers "Ubuntu One consumer secret differs from Ubuntu SSO leading to authentication failures (affects: 8) (dups: 2) (heat: 30)" [High,Confirmed] https://launchpad.net/bugs/762004 [14:01] nessita: Bug 762072 on http://launchpad.net/bugs/762072 is private [14:01] Launchpad bug 759197 in ubuntuone-client (Ubuntu Natty) (and 2 other projects) "[natty] Ubuntu One Folder bookmark is not created (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/759197 [14:02] Launchpad bug 764646 in ubuntuone-client (Ubuntu) (and 1 other project) "music store widget dies with «TypeError: find_credentials() takes exactly 3 arguments (2 given)» (affects: 1) (heat: 8)" [Critical,Triaged] https://launchpad.net/bugs/764646 [14:02] Launchpad bug 750309 in ubuntuone-control-panel (Ubuntu) (and 1 other project) "Services tab layout doesn't work right (affects: 2) (heat: 241)" [Medium,Triaged] https://launchpad.net/bugs/750309 [14:02] DONE: sprint & vacation [14:02] TODO: get caught up with mail and see what there is TODO [14:02] BLOCKED: no [14:02] thisfred: I sent you an email with 3 bugs for you to look at [14:02] DONE: swap day because I was useless. [14:03] nessita: awesome, I'll get to it eventually, I'm sure [14:03] TODO: SSO UI for windows, talk with clarita&mandel about dashboard, magmt call, etc. [14:03] BLOCKED: no, swamped... a bit. [14:03] alecu? [14:03] DONE: A branch that fixes bug #764646 but that reverts the (broken) fix for bug #759197. A branch to fix the later bug in a cleaner way. [14:03] TODO: push for this branches to land ASAP. [14:03] BLOCKED: no [14:03] NOTE: away thu and fri as national holidays [14:03] Launchpad bug 764646 in ubuntuone-client (Ubuntu) (and 1 other project) "music store widget dies with «TypeError: find_credentials() takes exactly 3 arguments (2 given)» (affects: 1) (heat: 8)" [Critical,Triaged] https://launchpad.net/bugs/764646 [14:03] Launchpad bug 759197 in ubuntuone-client (Ubuntu Natty) (and 2 other projects) "[natty] Ubuntu One Folder bookmark is not created (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/759197 [14:04] me [14:05] sorry! [14:05] mandel: go [14:05] DONE: Implemented async solution for file system notifications on windows. Talked with facundo to see what is the best approach [14:05] TODO: talk with ralsina and clarita about SSO ui and control panel. Get the changes proposed [14:06] BLOCKED: no, but the dog has a tendency to get hurt.... [14:06] ralsina_: I need your help on bug #750309, I've tried the fix you propose (show/hide) and still won't work. Sadly, the issue is not happening to me so I can't test crazy ideas. Would you be able to help? [14:06] Launchpad bug 750309 in ubuntuone-control-panel (Ubuntu) (and 1 other project) "Services tab layout doesn't work right (affects: 2) (heat: 241)" [Medium,Triaged] https://launchpad.net/bugs/750309 [14:06] sorry again, the beast decided to get itten [14:06] bitten* [14:06] nessita: sure, I can take time for it in, say 1 hour? [14:06] mandel: by what/whom? [14:06] ralsina_: perfect [14:07] nessita: I could even give you VNC into my VM, if you want [14:07] ralsina_: though, don't you have the team leads call in one hour? [14:07] ohhhh yeah [14:07] ralsina_: yeah, that can work [14:07] ralsina_: german shepherd, not too bad but I had to clean the wound etc... [14:07] nessita: 11:30 ART I am free [14:07] ralsina_: perfect [14:10] ralsina_: do we have the talk with clarita now? [14:10] mandel: in 2' [14:10] ok [14:11] mandel: actually she was having lunch 20' ago, so I don't know [14:11] ralsina_: ok, cool, s just that I wana do a mumble with facundo about the sd so I need to plan them accordingly [14:12] mandel: facundobatista first, then clarita [14:20] mandel: I keep forgetting my team leads call, it's in 10' and on the phone. It will probably take one hour or so. are you still around at that time? [14:20] ralsina_: me, of course :) [14:21] can you ask clarita? If she can't, just go ahead without me and I'll check with you later. I basically like the new dashboard. [14:22] ralsina_ mandel I need to join team leads too so can we catch up at 3.30? [14:22] clarita: hahaha sure :-) [14:26] clarita: how many of your days do we have? [14:26] is it 1 or 2? [14:31] mandel 2 [14:34] alecu: ping [14:34] nessita, pong [14:35] alecu: did you read my standup? I'm not able to have the U2 bookmark added, I need your help to confirm I'm following the proper steps [14:35] nessita, U2? are we doing music shows now? [14:35] nessita, ok, let's check it. [14:35] alecu: U1, sorry :-) [14:36] alecu: so, yesterday, I built your branch and I did: [14:36] * inside libubuntuone, sudo make install [14:36] * inside nautilus, sudo make install [14:36] alecu: I restarted the computer, and today I logged in as a user that didn't have a U1 account setup [14:37] alecu: and after the credentials were emitted, the U1 bookmark was not in place [14:37] nessita, well, this is inside the gsd plugin, so no wonder it didn't work [14:37] alecu: argh [14:38] alecu: why not? the changes from your branch are in libubuntuone [14:38] nessita, you need to link in the folder /usr/lib/gnome-settings-daemon-2.0 [14:38] to this file: /home/alecu/canonical/ubuntuone-client/the-other-signal/gsd-plugin/.libs/libubuntuone.so [14:38] let's put this in the wiki [14:40] nessita, also: the file is named "libubuntuone.so", but it's the ubuntuone gsd-plugin, not libubuntuone per se. [14:40] alecu: but your changes are localted in libsyncdaemon, so I'm not following :-/ [14:40] alecu: I mean, why do I need to overwrite the gsd thingy when theoretically didn't change? [14:42] alecu: ok, I also ran sudo make install inside the gsd-plugin dir. Is that all is needed? [14:42] nessita, if I do ldd on /usr/lib/gnome-settings-daemon-2.0/libubuntuone.so I get (a lot of lines) and this particular line: [14:42] libsyncdaemon-1.0.so.1 => /home/alecu/canonical/ubuntuone-client/the-other-signal/libsyncdaemon/.libs/libsyncdaemon-1.0.so.1 [14:43] nessita, I never tested with sudo make install, but I think it should work, yes. [14:43] nessita, I always tested with the first symlink [14:44] nessita, btw: the user you just created is clean, right? [14:44] make sure that this gconf key is clear: [14:44] (one sec) [14:44] alecu: clean as is... [14:45] /apps/gnome_settings_daemon/plugins/ubuntuone/checked_bookmark_file [14:46] nessita, only if that key *does not* exist then is the bookmark created. [14:46] so that's why I insist on a new user. [14:49] alecu: where can I delete/check if that key exists? [14:49] nessita, use gconf-editor [14:50] * nessita uses [14:51] hum unity is being mean to me [14:57] alecu: confirmed that the key is not in place [14:57] nessita, but did it work? [14:58] alecu: nopes, but I remembered I don't have u1-client-gnome installed [14:58] so I'm installing [14:58] nessita, u1-client-gtk, right? [14:58] alecu: nopes, u1-client-gnome [14:59] ubuntuone-client-gnome_1.6.1-0ubuntu1_amd64.deb [14:59] nessita, make sure to do the linking again after that. [14:59] alecu: yeap [14:59] nessita, oh, right. It's the control-panel that ends with -gtk [14:59] yeap :-) [15:00] alecu: it worked! [15:01] * alecu relaxes [15:01] alecu: now, question. What if both signals came in? [15:01] alecu: like the old and the new one, are we getting 2 bookmarks? [15:01] nessita, no, only one. [15:01] perfect, approving [15:02] nessita, remember that the signal will keep coming again and again every time you log in. [15:02] nessita, that's what the gconf key is for [15:02] yes [15:02] ok [15:02] nessita, so, if the user wants to get rid of the nautilus bookmark it won't show up again. [15:02] alecu: unless the gconf key is removed [15:03] but that's ok [15:03] right [15:05] alecu: all you need now is 2 more approves (one for each branch) [15:05] nessita, yes, and I need to propose them against stable-1.6 [15:10] alecu: if you say my jokes are good I'll give you the reviews [15:10] alecu: but you have to mean it when you say it [15:10] ok, I'll be mean while saying it [15:11] (in a mean tone) "mandel jokes are so great!" [15:11] hehehehe [15:11] I can you picture you in Buenos aires with a mena face hehehe [15:11] alecu: I'm on them, can I have the urls? [15:12] mandel, only this one: https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal [15:13] how comes it's already april and there are still mosquitoes around? [15:14] this must have been a great year for Fuyi-Vape. [15:14] "it's cold Mr. Mosquito. Why don't you just die already, please?" [15:14] alecu: any specific instructions to test it? [15:15] dont complain, I can't open my windows because there are bugs the size of a fist [15:15] he funny enough that would also work if I was talking about a computer :P [15:15] mandel, to install it: https://wiki.canonical.com/UbuntuOne/CodeReviewFaq#ubuntuone-client%20gnome-settings-daemon%20Plugin [15:16] * mandel dives to test [15:16] mandel, then, to review it: create a new Ubuntu user, log as him, check that nautilus has not created a bookmark to the "Ubuntu One" folder. Then log into Ubuntu One, with a new user or an existing one, and a few seconds later the bookmark to the "Ubuntu One" folder should have been created. [15:17] * alecu is putting this in the review proposal. [15:17] alecu: do you want me to test it in M too? [15:18] mandel, no need to test this on M too [15:18] ok [15:19] nessita, btw: we should be putting how to test this plugins on wiki.ubuntu.com instead... [15:19] not on wiki.canonical.com [15:19] * alecu makes a mental note to move stuff around [15:19] alecu: I m not sure we want to add this as official doc (is not) [15:20] nessita, hmmm.... but we are putting links to the private wiki in the public launchpad bugs [15:20] alecu: where for example? [15:22] nessita, for example, https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal/+merge/58190 [15:24] alecu: you just added that link, that cheating :-D [15:24] alecu: we shouldn't add those link to public pages, agreed [15:24] nessita, and also on the other nautilus-plugin or gsd-plugin pages [15:25] * mandel adds lots of canonical wiki link everywhere…. [15:25] alecu: well, we should not be adding that. But I don't think we should move that to the public wiki, since we don't properly maintain the doc [15:27] my internet sucks. [15:27] * alecu restarts his network driver. [15:31] mandel, seems like roman has just finished reviewing that branch [15:31] oh, well! I tried [15:31] mandel, thanks anyways! [15:31] and I got complemented for my jokes [15:31] :P [15:33] found the issue with my internets: yesterday I copied a few files into ~/Videos, and now SD is hogging my BW [15:37] nessita, ralsina_: do you guys know if dobey will be around today? [15:38] alecu: nt today, yes tomorrow [15:38] alecu: he's on a swap day today [15:38] ok. === clarita_ is now known as clarita [15:45] nessita, rye: https://code.launchpad.net/~alecu/ubuntuone-client/revert-dbus-interface-change-stable/+merge/58315 and https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58316 [15:46] finally, phone off my ear. Oh, wait, have to talk to clarita and mandel :-) [15:46] mandel, clarita: ping? [15:46] pong [15:47] Anyone knows how to fix this error on apt-get update? [15:47] http://pastebin.com/Htmde4Ab [15:48] clarita: free for skype? [15:48] ralsina_ mandel: yup [15:49] ok, calling you all [15:49] * mandel launches skype [15:49] rodrigo_, ping [15:49] alecu, pong === kenvandine_ is now known as kenvandine [15:50] rodrigo_, I wanted to ask this to dobey, but since he's on swap today I'll take a bit of your time: [15:50] rodrigo_, is it ok to call twice dbus_g_object_register_marshaller with the same marshaller? [15:50] mandel clarita: I lost everyone :-) [15:51] rodrigo_, as in here: https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal/+merge/58190 [15:51] ralsina_, no idea about that apt-get error [15:51] alecu: this is wrong, you need to propose https://code.launchpad.net/~alecu/ubuntuone-client/revert-dbus-interface-change-stable/+merge/58315 against stable-1-6 [15:51] alecu, hmm, not sure, let me check [15:51] not my best phone day. [15:52] ralsina_: sorry I'm useless I though I was logged in, it turns out I was not … [15:52] alecu: same for the other branch [15:52] mandel: haha [15:53] alecu, I think it's ok, but usually you just need to register the marshallers once, so I'd move that to the lib initialization code [15:53] alecu: ping [15:54] alecu, iirc, in libsd there area other marshallers being registered at init time, so just move it there, I guess [15:54] rodrigo_, it makes sense [15:54] rodrigo_, thanks! [15:54] nessita, pong [15:54] alecu, of nothing :) [15:54] jajajajaaj [15:54] alecu: did you see my comment about targeting stable-1-6? [15:55] alecu: also, can you please commit with --fixes for the branch solving bug #759197? [15:55] Launchpad bug 759197 in ubuntuone-client (Ubuntu Natty) (and 4 other projects) "[natty] Ubuntu One Folder bookmark is not created (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/759197 [15:56] nessita, yes, I saw your comment, but I forgot to target the -stable branches to stable-1-6 when doing the merge proposal :P [15:56] alecu: it happens :-) [15:56] nessita, but I think I did the --fixes the right way on both branches. [15:56] alecu: ah, and I think you did use --fixes but the branch didn't land yet on trunk (I see the status is not Fix Committed and I thought that --fixes was missing) [15:57] alecu: right, sorry [16:02] alecu: did you fix the proposals? [16:02] rye, I forgot to set the proposal to merge with stable-1-6, so I've re-submitted the merge proposal again. [16:02] * nessita is waiting for that before moving to another task [16:04] nessita, yes they are modified, but I asked ro-drigo about the duplicated marshallers (since dobey is not around today) and he suggests moving the marshallers to the init code... this is in regard to the second branch. [16:04] nessita, rye, the first proposal: https://code.launchpad.net/~alecu/ubuntuone-client/revert-dbus-interface-change-stable/+merge/58322 [16:05] nessita: I am postponing your bug untiil 12:30, is that ok, or you want to have lunch first? [16:05] ralsina_: is ok (I usually have lunch after 1pm) [16:05] nessita: ok. sorry about the delays, but I am blocked until then. [16:05] ralsina_: no prob [16:17] rodrigo_, nessita: I'm looking at the marshaller registration code in libdbus-glib, and I see that there is no problem doing the registration twice. [16:17] alecu: thanks for checking [16:17] I prefer to do it like that, because the patch is a lot smaller than doing it in the initialization code. [16:17] and we will get much better chances of it being accepted. [16:19] rye, can I ask for your review on the updated proposal? https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58324 [16:19] hi [16:19] hi hrw [16:20] I am using u1f on nexus s phone. But it does not sync "/mnt/sdcard/u1/Ubuntu One" subdirectories. [16:21] I have rev32 version (so probably a bit old one) [16:21] alecu, ok [16:21] hrw, let's ask karni, he'll surely know. [16:21] karni, ^^^ [16:21] * karni is here. what's up? [16:21] aha [16:21] * karni reads [16:21] karni: cześć [16:21] hrw: hah, let's stick to English on #ubuntuone ;) [16:22] yes [16:22] hrw: so, the deal is [16:22] hrw: the core of the app is being replaced as we move to REST [16:22] hrw: you can say - u1f you have is out of date (or out of order heh) [16:22] hrw: we have dropped sync though, as it was too slow for users with multiple gigabytes of data [16:23] so what use is for u1f? [16:23] hrw: but if you're asking -- as long as you have checkboxes or the sync emblem to the *right* of folders or files [16:23] hrw: checking that should periodically download those files/folders to your phone [16:23] hrw: wanna the latest (and last with sync) build to your e-mail? [16:24] karni: I can say one thing about u1f/rev32 - that it is unusable when it comes to get infos what is synced or not [16:24] karni: yes, please [16:24] hrw: I'll tell you what :) Great integration with the system, sharing capabilities, future push, and possible lightweight sync. :) [16:24] hrw: sure, feedback always appreciated. noted [16:24] hrw: please send it on priv [16:25] sent [16:25] hrw: latest is 1.2.60 [16:25] * karni grins [16:25] ok [16:25] 1.0.32 on phone ;d [16:26] karni: would be nice to have u1f available to download on lp [16:26] hrw: I know, we'll have it on the Market heheheh. Just not yet :< [16:28] hrw: to sum up - change will come. you've got the latest sync-capable build (sync-down). please do not file bugs, as this client version has just been abandoned. [16:28] hrw: ah! please uninstall first. [16:29] karni: noticed by FC [16:30] hrw: heh ;) yeah, incompatible version. just uninstall first. [16:30] hrw: do let me know how it worked for you, though [16:30] :) [16:32] karni: launching browser to login does not launch browser and announces that login failed [16:32] interesting.. [16:32] dolphin hd browser set as default one [16:32] rye: have we fixed oauth bug yet? [16:32] rye: the one which caused to issue wrong token [16:32] thisfred: ping [16:33] and progress ring is rotating all the time in right top corner even when app does not do anything [16:33] nessita: pong [16:33] thisfred: did you get to the bug reports I linked? [16:33] hrw: noted [16:33] nessita: Still going through my mail [16:33] hrw: (kill it and run again. if it fails, the problem is on the server side - try again later in such case. sorry :( ) [16:34] karni: http://paste.ubuntu.com/596088/ [16:34] thisfred: is there any chance you jump to that email? I would like to know if we need to hurry branches for natty [16:34] nessita: sure, let me search [16:34] E/UbuntuOneFiles(19525): oauth.signpost.exception.OAuthExpectationFailedException: Request token or token secret not set in server reply. The service provider you use is probably buggy. [16:34] thisfred: thanks, subject is High priority bugs [16:34] rodrigo_, hey [16:34] updated the bug [16:35] I go on holiday tomorrow, so I need to do any final debugging today on this one [16:35] hrw: yes, it might be serverside. we have one oauth bug reported. please try again later, in the mean time I'll see what's the progress on it [16:36] jono, looking [16:36] karni: ok [16:36] thanks [16:36] hrw: thanks, and sorry it fails at the moment [16:37] hrw: we'll have better login procedure as well heh. basically, 60% of the app will change. [16:37] hrw: so that it's faster and more user friendly. and I hope sync is just matter of time, because I was particuarly happy with it. [16:37] facundobatista: can you please reply to last comment in bug #757532 adding instructions on how to reproduce? for me it works fine :-) [16:37] nessita: Bug 757532 on http://launchpad.net/bugs/757532 is private [16:38] karni: I hope that at least u1f will be usable. when I tried u1 contacts I decided to abandon any hope on contacts part [16:38] jono, you need to let e-addressbook-factory crash before 'thread apply all bt' [16:38] rodrigo_, oh, I see [16:38] I am not sure it is crashing [16:38] let me try [16:39] nessita, do you have desktopcouch-ubuntuone installed? [16:39] jono, the backtrace you pasted indicates it didn't [16:39] facundobatista: right now, yes, but I uninstall it and kill it and re install it from the control panel on a daily basis [16:40] jono, run evolution after hitting 'r' and then try opening the U1 addressbook and wait until it crashes [16:40] jono, when it crashes, you'll get the (gdb) prompt without having to hit CTRL-C [16:41] nessita, what do you kill=? [16:41] rodrigo_, interesting [16:41] facundobatista: becasue uninstalling will not stop desktopcouch-service [16:41] natty, I am only now able to fix my broken apt-get, and am downloading the natty updates, so ... after lunch? [16:41] nessita: , I am only now able to fix my broken apt-get, and am downloading the natty updates, so ... after lunch? [16:41] when I click on the Contacts button in Evo and then click on Ubuntu One - I get this in the term [16:41] (e-addressbook-factory:15727): Json-CRITICAL **: json_object_get_object_member: assertion `node != NULL' failed [16:41] nessita, I have couchdb stuff, but no desktopcouch-service running, should I kill something before I try again? [16:41] facundobatista: so, you need to do: sudo aptitude purge python-couchdb; sudo killall desktopcouch-service beam.smp [16:41] I get lots of those rodrigo_ [16:42] jono, hmm, do you have the latest couchdb-glib? [16:42] nessita, *python-couchdb*? [16:42] facundobatista: beam.smp if it's running, and then purge all couchdb related [16:42] rodrigo_, I have whatever is in natty :-) [16:42] nessita, I wasn't missing python-couchdb [16:43] facundobatista: what does that mean? [16:44] nessita, I had python-couchdb installed, the bug talks about other package [16:44] rodrigo_, so it looks like Evo is trying to parse the contacts - I get a *lot* of those Json-CRITICAL errors, and I have 1000+ contacts [16:44] jono, oh, yes, I fixed those warnings in git but haven't made a release [16:44] facundobatista: I know, but if you uninstall that it will uninstall everything [16:44] facundobatista: everything in a good way [16:44] jono, they are harmless, so don't worry [16:44] and my hard disk is crunching away as it keeps streaming those messages [16:44] jono, but yes, evo-couchdb is reading the contacts [16:44] rodrigo_, I just kicked it off again to see if it crashes [16:45] I am hoping it crashes when it is done processing the contacts [16:45] jono, you have libcouchdb-glib 0.7.2-0ubuntu2 ? [16:46] nessita, attached, thanks [16:46] facundobatista: is still happening the same for you? [16:46] nessita, yeap [16:46] facundobatista: didn't you get a spinner? [16:46] facundobatista: after the installation ended [16:47] nessita, I had to reproduce it because I removed the logs two days ago [16:47] nessita, what is "a spinner"? [16:47] facundobatista: http://developer.gnome.org/gtk/2.24/GtkSpinner.html [16:47] facundobatista: once DC is installed, didn't you get a spinner on top of the "Enable file sync" frame? [16:48] nessita, I'll reproduce it once more, and will keep it open for you to ask, wait [16:48] facundobatista: thanks, be sure to kill dc-service [16:48] ah, a throbber [16:48] facundobatista: that means: sudo aptitude purge python-couchdb; sudo killall desktopcouch-service beam.smp [16:48] rodrigo_, no libcouchdb-glib installed [16:48] nessita, I will not uninstall python-couchdb, thanks [16:48] do I need to install libcouchdb-glib-1.0-2 ? [16:49] facundobatista: then how are you planning on reproduce? [16:49] rodrigo_, btw, I got some new errors after all those Json errors - it does list an auth key - is it safe to pastebin that? [16:49] best url EVER: http://www.independent.co.uk/life-style/food-and-drink/utter-PR-fiction-but-people-love-this-shit-so-fuck-it-lets-just-print-it-2269573.html [16:49] sorry for the spam :P [16:50] nessita, I remove the package that originally found missing [16:50] facundobatista: that is not enough, you need to purge desktopcouch [16:50] facundobatista: becasue not having desktopcouch will make the service startup delay [16:51] nessita, do you want me to reproduce the bug, or alter the context so the bug may not happen? [16:52] facundobatista: aux le contraire, I want you to reproduce as close as possible, and in a clean install, the missing package is the whole couchdb thingy, not just our plugin [16:52] facundobatista: go ahead, reproduce as you think is best, if you could make a screencast I'll be able to understand your issue [16:52] nessita, as close as possible to what? [16:53] facundobatista: do it as you wish, please, if you can, record a screencast [16:53] mandel: http://www.independent.co.uk/life-style/food-and-drink/mandel-is-easy-to-fool-2269573.html [16:54] facundobatista: I'm understanding she wants you to have all the packages setup from scratch to ensure that it is happening for a new user [16:54] mandel, ok, that's a different issue [16:54] jono, not installed?? [16:54] if it's happening or not for a new user is good to know [16:54] thisfred: I prefer my version :( [16:54] jono, then it's using a libcouchdb-glib from other place [16:54] but to avoid happening to any user, or me particularly, is what the bug is about :)= [16:55] jono, yes, the oauth key is local, so no problem, but replace it with xxxx if you want [16:55] thisfred: http://www.independent.co.uk/life-style/food-and-drink/thisfred-let-mandel-enjoy2269573.html [16:55] jono, or pastebin.canonical.com [16:55] jono, so, the problem seems to really be you don't have libcouchdb-glib installed [16:56] jono, so, can you please run 'ldd /usr/lib/evolution-data-server-1.2/extensions/libebookbackendcouchdb.so' and pastebin the output? [16:57] rodrigo_, https://pastebin.canonical.com/46359/ [16:57] rodrigo_, want me to install libcouchdb-glib-1.0-2 ? [16:58] jono, run 'dpkg -S /usr/lib/libcouchdb-glib-1.0.so.2 /usr/lib/libjson-glib-1.0.so.0' [16:59] rodrigo_, : [16:59] jono@forge:~$ dpkg -S /usr/lib/libcouchdb-glib-1.0.so.2 /usr/lib/libjson-glib-1.0.so.0 [16:59] libcouchdb-glib-1.0-2: /usr/lib/libcouchdb-glib-1.0.so.2 [16:59] libjson-glib-1.0-0: /usr/lib/libjson-glib-1.0.so.0 [17:00] so it's installed [17:00] so now: dpkg -l libcouchdb-glib-1.0-2 [17:00] right [17:00] rodrigo_, ii libcouchdb-glib-1.0-2 0.7.2-0ubuntu2 GLib-based API for CouchDB [17:00] ok, now can you pastebin the output from e-a-f with the oauth tokens in it? [17:00] the full one, if possible [17:01] rodrigo_, btw, here were the errors: https://pastebin.canonical.com/46362/ [17:01] and still no contacts in evolution? all those warnings you mentioned are because evo-couchdb is translating the items in the couchdb db to vcard for evolution [17:02] rodrigo_, it was after the contacts seemed to be parsed - after all those Json errors [17:02] rodrigo_, I removed some of the keys and put in xxxxxxxxxxxx [17:02] ok, so it gets the contacts db correctly, getting 1382 documents [17:02] is that the number of contacts you have? [17:03] karni: btw - version which I got from you does not remember that it shown 'welcome' requester [17:03] hrw: you mean the login screen? [17:04] hrw: once it kicks-in with correct oauth, you should never see it again (unless you unlink from U1 from PReferences) [17:04] * karni checks oauth dance [17:04] jono, also, in the pastebin it shows you hit CTRL-C, can you just leave it running and see what it does? [17:04] jono, that pastebin shows the opening and auth to couchdb is done correctly [17:05] hrw: heh, Login failed, please try again later. awesome ;P [17:05] rodrigo_, one sec [17:05] ok [17:05] karni: http://home.haerwu.biz/~hrw/snap20110419_180444.png [17:06] nessita, sent by mail [17:06] hrw: uuuuuuuuuu it looks nice xD! right. It's only because it's "Thank you for *testing*" :) I'll make sure to include "Do not show this again" once it's released. [17:06] hrw: Thanks for your input! Good catch. [17:07] rodrigo_, ok, one sec - it is currently spewing Json errors again, takes a few minutes to get to the end [17:07] facundobatista: thanks! [17:07] karni: no, it should show only on first run and after app upgrades when it should show whats new [17:07] * nessita -> lunch [17:07] jono, yes, for 1382 contacts, you'll get a lot of those warnings [17:07] I should really add the upstream patch that removes them to the natty package [17:07] rodrigo_, I don't remember seeing any other errors [17:07] I ran Ctrl-C because nothing else was showing [17:07] but can leave it going [17:08] hrw: right, that's even yet something different :) changelog VS instructions VS "beaware, you're testing!" :) But yes, I know what you mean :) Thanks [17:08] karni: check TitaniumBackup for example [17:08] jono, look for lines starting with 'Sending GET ...' [17:08] jono, and the following lines, that show 'Response body' [17:08] hrw: This is *not* the release version :) [17:08] rodrigo_, oh it crashed! [17:08] jono, ok, thread apply all bt now [17:08] karni: this is FC release again [17:09] hrw: during oauth? [17:09] rodrigo_, I added that, nothing [17:09] maybe it didnt crash [17:09] but it went back to the gdb prompt [17:09] what did gdb say? [17:09] hrw: Anyhow, this FC has probably originated in 11k lines I have removed few days ago. Long gone. [17:09] Bye. [17:09] [Thread 0xb77ddb70 (LWP 16199) exited] [17:09] [Thread 0xb7fdeb70 (LWP 16198) exited] [17:09] Program exited normally. [17:09] (gdb) thread apply all bt [17:09] karni: http://paste.ubuntu.com/596105/ - during start [17:09] it should show somethinh like 'Program receveived SIG*...' [17:09] rodrigo_, ^ [17:09] oh [17:09] did you quit evolution? [17:09] nope [17:10] hmm [17:10] hrw: Thanks again. Yes, I did have one report of that before. That's an interesting one. Will file a bug report. [17:10] jono, can I ssh into your machine? [17:10] karni: it happens when sdcard is unmounted [17:10] karni: u1 should check for it [17:11] rodrigo_, not really, I firewall it out [17:11] hrw: correct. and I'm trying to write logs there. [17:11] rodrigo_, what do you want me to run? [17:11] jono, ok [17:11] so not sure why e-d-s decides to quit [17:11] that seems to be the problem [17:11] so evolution never gets the contacts [17:11] rodrigo_, could it be because there are too many contacts? [17:12] or bad data in the contacts? [17:12] jono, in that case, it would timeout [17:12] which I was thinking was going to be the problem [17:12] rodrigo_, I do get timeout errors in Evo [17:12] oh [17:12] karni: colour scheme of app would be nice to adapt to system colours [17:12] I put that in the bug report [17:12] karni: most of my phone is white on black not ubuntuorange on white [17:12] hrw: Design got ZOMG late onto that, themes needs fixing [17:13] jono, yes, but that was before, when you didn't get any output, e-a-f was crashing, so evo got a timeout from dbus [17:13] hrw: noted! [17:13] rodrigo_, I get a timeout Evo error after all the Jason errors I think [17:13] hrw: btw the "ubuntuorage on white" was attempt to streamline it with the WebUI. But we may consider otherwise, sure. [17:13] jono, oh ok [17:14] karni: U1 logo on top bevel is hardly readable due to gradient below it [17:14] so yes, the problem seems to be you have many contacts [17:14] hrw: placeholder ;) [17:14] before, e-d-s used corba so there was no timeout [17:14] hrw: forget the looks, please :) [17:14] but now with dbus there is a timeout indeed [17:14] karni: thats the only thing which works for me now ;D [17:14] rodrigo_, it seems like 1300 contacts isnt that many - can you fix the upper limit? [17:14] I'll have to get the contacts in chunks [17:14] ahhh ok [17:14] hrw: have you been able to log in? [17:14] rodrigo_, do you think this is fixable for natty? [17:14] jono, the upper limit is dbus' timeout [17:14] * karni laughs [17:15] alecu: what are your thoughts on bug #762717? I think it may be better to hide the progress bar when disconnected, but I don't think it's worth a freeze exception [17:15] Launchpad bug 762717 in ubuntuone-control-panel "Launcher icon shows syncing progress when not connected (affects: 1) (heat: 141)" [Undecided,New] https://launchpad.net/bugs/762717 [17:15] jono, I think so, but will have a look [17:15] karni: nope [17:15] rodrigo_, cool - so I guess you could simulate by important a tonne of contacts, at least 1400 so you can see if it crashes for you [17:15] alecu: it seems pretty low priority to me: we aren't lying to the user [17:15] rodrigo_, particularly with Evo auto-adding contacts, I suspect this bug could bite a lot of people [17:15] hrw: I see :< Thanks for those comments on the login screen! :D [17:15] jono, the crash is another thing, that's why I'd need a gdb backtrace [17:16] karni: expect more when/if it will work [17:16] karni: oauth is borked only in this version or in total? [17:16] karni: access to settings/about from login screen menu would be nice [17:16] hrw: It's a nut to crack. I'm replacing almost all of it, so I'm not sure if they will be valid. But please do keep that feedback coming. [17:16] rodrigo_, ok, so do you need anything else from me? [17:17] jono, I'll try to cook up a quick patch [17:17] hrw: good point [17:17] jono, you're out tomorrow, right? [17:17] hrw: oauth -> server side [17:17] rodrigo_, out tomorrow, but back tomorrow at about 6pm [17:17] hrw: so I'm glad it's not released when oauth is broken [17:17] karni: now I can not even say which version I am running [17:17] so I can test anything then, and then I am out from Thu onwards [17:17] hrw: true [17:17] thanks rodrigo_, appreciate all the effort on ths [17:17] this [17:17] jono, ok, then run e-a-f on gdb and if it crashes, 'thread apply all bt' and paste the output on the bug, ok? [17:17] karni: add version info into debug.log [17:18] jono, although as I said, that's not related to the timeout, I guess [17:18] rodrigo_, run what? [17:18] e-a-f [17:18] hrw: it is. you just can't send it :x [17:18] what is e-a-f? [17:18] e-addressbook-factory [17:18] oh right [17:18] will do [17:18] so you will cook a patch and then I will do that [17:19] jono, if you can run it on gdb while I cook the patch, we might know why it crashes and fix the 2 issues in 1 patch [17:19] although I guess it's not crashing, it's just quitting because there are no clients [17:19] as evolution doesn't open the u1 addressbook because of the timeout [17:21] hrw: although we're abandoning this branch, you've got one of the best comments I've heard. I'll certainly incorporate some of the fixes in the new version. [17:21] rodrigo_, yeah, I have not been able to make it crash [17:21] karni: beware me [17:22] thisfred, looks like wishlist to me too [17:22] jono, ok, I'll let you know tomorrow if I have a patch for you to test, ok? [17:22] jono, ping me tomorrow when you're around [17:22] rodrigo_, perfect, thanks! [17:22] alecu, ok, thanks, marked as such [17:22] karni: got browser [17:22] if I have it soon than that, I'll ping you [17:22] hrw: half way there [17:22] rodrigo_, will do, actually, I will be around tomorrow morning pacific [17:22] at about 9am [17:22] ok [17:22] can do some testing then if you have something [17:22] otherwise, won't be back until 6pm [17:25] karni: ok, got there. app is oauthed [17:25] hrw: you can count time ;) [17:25] hrw: I wonder how much GB of data you have [17:25] karni: settings shows "username:unknown, email:unknown" [17:25] karni: current plan should show GBytes not bytes [17:26] hrw: if you wanna fix it, tap the current plan [17:26] hrw: this will be fixed in next version. [17:28] karni: 'files in could\n0 file(s)XXX server bug' does not look nice [17:28] s/could/cloud/ [17:28] hrw: because it was a server bug. and it's a debug build ;) [17:28] haha [17:29] *is a server bug [17:29] we're changing to REST, which fixes it. thanks for noticing that, hrw :) [17:29] * karni is getting some heat from hrw [17:29] facundobatista: you're getting an erRor! [17:30] karni: 'sync also..' should be 'sync also...' [17:30] karni: and all entries there are disabled [17:30] facundobatista: in the bug report you say that no other replication service show up :-) [17:30] hrw: yes there are. it's a stub for feature that has not been merged. [17:31] karni: I have 2+20 not 20 which app reports [17:31] facundobatista: your problem is specific to us (u1-hackers). Please follow pedronis' instructions on email with subject "hackers PPA CouchDB and desktopcouch" [17:32] hrw: good! found a server bug [17:32] will file against U1F before I confirm it's not just me. [17:32] karni: and had to reenter settings to get proper email/username/currentplan infos [17:33] karni: would be nice to have option to mark which directories have to be synced as now it is unknown for me as a user [17:34] karni: using 'sync' gives "Not syncing: sync in progress" but no u1 icon in notification [17:34] ralsina_: ping [17:34] * hrw ends a day [17:34] hrw: thanks [17:34] have a nice rest of day [17:34] hrw: no sync for next release, comment invalid [17:34] hrw: thanks a bundle !! [17:35] hrw: great feedback. have a great evening [17:38] ralsina_: whenever you are back, a windows review for https://code.launchpad.net/~mandel/ubuntuone-client/close_files_in_tests/+merge/58343 would be very welcome [17:39] Take care everyone. [17:39] alecu: ping [17:39] hi joshuahoover [17:40] alecu: do you know of any workaround for bug #745540 ? [17:40] Launchpad bug 745540 in ubuntuone-client (Ubuntu) (and 5 other projects) "Method "CreateItem" with signature "a{sv}(oayay)b" on interface "org.freedesktop.Secret.Collection" doesn't exist (affects: 7) (heat: 36)" [Undecided,Invalid] https://launchpad.net/bugs/745540 [17:40] * mandel walks dog, might be back later [17:43] joshuahoover, I have a branch to fix it, and it will probably be available on nightlies soon, but we won't include that in natty [17:43] joshuahoover, perhaps on an sru [17:43] alecu: yeah, an sru [17:44] alecu: ok, just wanted to make sure there wasn't a workaround i could give a user being affected by this...thanks! :) [17:44] joshuahoover, anyway: this is only happening to people installing the gnome3 ppa, so as a workaround perhaps uninstalling that ppa and installing the gnome-keyring from natty [17:44] alecu: yeah, ok...makes sense [17:45] ok, I'm off to lunch! === alecu is now known as alecu-lunch === czajkows1i is now known as czajkowski [18:44] mandel: got it! [18:48] nessita: when I click the u1 launcher now, the control panel shows the sign up page briefly, and then goes to the regular account tab, it didn't do that before I think? [18:49] thisfred: it always did that. Since beta 2, I notice dbus being much slower than before, so you get to see that splash screen [18:49] I'm looking at bug #762722, and I may be grasping at straws [18:49] Launchpad bug 762722 in ubuntuone-control-panel "Opening the Control Panel during sync removes progress indicator (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/762722 [18:49] thisfred: this behavior is present (showing the splash for a fraction of time) since release 0 [18:49] nessita: ah, so before it was gone faster, probably [18:50] thx [18:50] thisfred: exactly, you're welcome [18:56] rye, ralsina_: could you please? https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/services-improve/+merge/58362 [18:56] nessita: sure! [18:56] rye, ralsina_: I know is black magic but is the inly thing that made the damn thingy be shown [18:57] nessita: at least it's *short* ;-) [18:58] ralsina_: since you can reproduce the bug, could you please confirm by field testing? [18:58] nessita: trying right now! [18:59] awesome [19:00] nessita: got Value could not be retrieved. (ServerError: (400, ('invalid_consumer_token_pair', 'Invalid consumer and t [19:00] in red. [19:01] ralsina_: you need to apply the patch that pedronis emailed to us [19:01] and a desktopcouch crash [19:01] urgh [19:01] ralsina_: email subject is "hackers PPA CouchDB and desktopcouch" [19:01] nessita: I'll look for it [19:01] nessita: can you forwards? I may have been too wuick with delete [19:03] ralsina_: sent! [19:03] nessita: cool, thx === ivanka is now known as ivanka-train [19:07] nessita: good news, I can see the whole thing [19:07] nice [19:07] nessita: not *so* god news, it's wider than the window and has horizontal scrollbar [19:07] ralsina_: that's "by design" [19:07] nessita: but it's much better [19:07] sorry :-/ [19:08] nessita: no problem [19:08] so, then +1 fieldtest! [19:08] great [19:09] nessita: good work making gtk behave [19:09] ralsina_: I'm very "unhappy" in the sense that this should not be needed [19:09] but I prefer this to having users confused :-) [19:10] nessita: if I were you I would just file it with other things I would rather pretend never happened :-) [19:10] jejeje [19:10] ponelo con los jeans nevados y listo [19:10] oops, sos demasiado joven :-) [19:10] I did wear jean nevados [19:11] hahahaha... at least you are not now wearing jean cutoffs with the pockets hanging below the legs. Everyone was using that on London. It looks as nice as it sounds. [19:16] 0.0 [19:18] alecu: could you please add instructions to https://code.launchpad.net/~alecu/ubuntu-sso-client/changed-secrets-spec/+merge/57947 to test IRL? [19:19] nessita, sure [19:19] thanks! [19:20] nessita: just in case you don't believe me, http://www.squidoo.com/cutoff-jeans-for-spring look at the bottom gallery. [19:20] * nessita looks [19:21] * ralsina_ notices he's procrastinating his windows review. [19:21] * fagan hates his isp [19:21] ralsina_: im here :) [19:21] Hi there fagan [19:21] any reviews or anything? [19:21] ralsina_: I like them :-) I would cut off some of my jeans as well! :-D [19:21] actually yes! [19:21] nice [19:21] nessita: me too, but don't do the pockets thing :-) [19:22] ok, I won't [19:22] * ralsina_ likes them, not going to do it to his jeans ;-) [19:22] fagan: it's a bit late for you isn't it? [19:22] ralsina_: 7:22 [19:22] its cool [19:22] ok, run the tests on this one: https://code.launchpad.net/~mandel/ubuntuone-client/close_files_in_tests/+merge/58343 [19:23] fagan: according to the instructions in the merge request [19:23] oki doke so is that just windows or natty? [19:23] fagan: just windows [19:23] cool [19:23] nessita: could you do a natty review of that one? [19:23] nessita: if you can't I will [19:23] ralsina_: I can! [19:23] nessita: cool! [19:30] is it a known problem that desktopcouch-service crashes on login? [19:30] thisfred: what kind of crash? 400? [19:31] nessita: haven't looked yet [19:31] thisfred: if it's 400 invalid consumer, please apply what pedronis said in the email [19:31] thisfred: subject "hackers PPA CouchDB and desktopcouch" [19:32] ralsina_ or mandel http://paste.ubuntu.com/596184 what am I missing [19:32] I havent built the client on windows yet so I dont know what im missing [19:32] fagan: install configglue using easy_install [19:32] ralsina_: ah ok [19:33] nessita: thx [19:33] thisfred: was that the issue? [19:33] nessita: it's a 400 BAD REQUEST, so not sure [19:33] I'll look at the email in a bit [19:34] ralsina_: one more http://paste.ubuntu.com/596188 [19:35] mandel: can you please explain to me what the "self.timeout - 0.2" is for? [19:35] mandel: what if that is a negative value? [19:35] nessita: alecu: https://code.launchpad.net/~thisfred/ubuntuone-client/lp-764057/+merge/58373 solves bug #764057 [19:35] Launchpad bug 764057 in ubuntuone-client (Ubuntu) (and 1 other project) ""program is still running: ubuntu one file synchronization in progress" at logout (affects: 2) (heat: 3261)" [High,Triaged] https://launchpad.net/bugs/764057 [19:35] thisfred: ack [19:35] fagan: ohhh that's trickier, I don't have the URL handy, ask mandel tomorrow, I'll do the test [19:35] ralsina_: ok cool then [19:36] * fagan gets dinner [19:36] thisfred: be sure to provide the fix also for stable-1-6 [19:36] nessita: how do I do that? [19:37] thisfred: branch lp:ubuntuone-client/stable-1-6, apply the analog changes, push under same branch name adding a -stable-1-6 suffix (or similar), then propose for merging against the lp:ubuntuone-client/stable-1-6 target branch [19:37] nessita: thx, will do [19:38] letmeknow [19:44] thisfred, I don't wholly agree with hiding the progressbar on disconnection [19:45] alecu: it was the simplest possible fix, though [19:45] yup, I understand that [19:45] thisfred, but the thing is that both the progressbar and the logout inhibit will be disabled [19:46] thisfred, and they won't be re-enabled when the connection is restored, but instead they will be re-enabled when a new file is completed transferring or a new file is added to the transfer queue [19:46] alecu: right, so we kill two bugs with one line. I know the other bug was 'wishlist' but I figured we might as well [19:47] alecu: right, that's not necessarily a problem though? When we reconnect we're not necessarily interested in any progress made before we disconnected [19:49] thisfred, that's a good point. But anyway I would like to see at least a some tests in order for this to land [19:50] and most importantly, some tests to know that this does not break stable [19:51] yeah, I'll look at that. [19:53] you can manually verify that it works by running syncdaemon from the branch, adding a bunch of files, disconnecting, and then trying to log out [19:53] thisfred, also: perhaps we should change the message that's shown when trying to log out [19:54] "file synchronization is in progress" -> "files pending synchronization" [19:54] perhaps that would make rye happy [19:55] alecu: but with this change, it will only inhibit when files synchronization *is* in progress right? [19:56] right, but if ubuntuone disconnects, and we disable the inhibition we are acting just as if synchronization was completed, meaning the user will think his files have finished transferring. [19:57] that's why I'm opposed to hiding the progressbar and disable logout inhibition on disconnects. [19:58] I'm adding this to the bug. [19:59] nessita, rye, ralsina_: what are your opinions on the above? bug #764057 [19:59] Launchpad bug 764057 in ubuntuone-client (Ubuntu) (and 1 other project) ""program is still running: ubuntu one file synchronization in progress" at logout (affects: 2) (heat: 10)" [High,Triaged] https://launchpad.net/bugs/764057 [19:59] * ralsina_ looks [20:00] alecu: on mumble ATM [20:00] alecu: I hate both options? ;-) [20:00] alecu: can we set a message on inhibition? [20:00] or is it just a flag? [20:02] ralsina_, we are currently setting only one message... let me find it. [20:02] FILE_SYNC_IN_PROGRESS = Q_("File synchronization in progress") [20:03] ralsina_, I propose we should change that to "files pending synchronization", or something like that. [20:03] but I'm not sure if we are able to change that at this point [20:04] grmbl [20:04] we can't [20:04] according to https://wiki.ubuntu.com/NattyReleaseSchedule we can't [20:04] pity [20:04] ok, maybe on the first SRU ;-) [20:04] right now, I say disable on disconnect [20:05] ralsina_, but "we are acting just as if synchronization was completed, meaning the user will think his files have finished transferring" [20:06] nessita: uhm. why are you running tarmac by hand? [20:06] alecu: well, since the user can't do anything about the disconnect... I think it falls under "it just (doesn't) work" [20:06] dobey: I ran it for stable-1-6 since it was failing for a branch for roman [20:06] ok [20:06] grr [20:07] alecu: in any case, nothing we can really fix NOW, but I would like if you could spend a couple of hours thinking the best possible solution. [20:07] alecu: and then write it down in that bug and mark it accordingly. Ok? [20:08] please don't [20:08] dobey: ack [20:08] ralsina_, ok. That means we should try to push thisfred's branch to just disable it on disconnections to stable? [20:10] alecu: yes [20:11] ralsina_, thanks [20:11] ralsina_, also: I'm waiting for a re-review from rye on this branch I resubmitted, but he doesn't seem to be around. Are you able to take a look at it? https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58324 [20:12] alecu: in 20' or so, sure [20:12] alecu: I'm off mumble now. What were you needing? [20:13] nessita, I wanted to ask more opinions on a bug above, but we've already settled. [20:13] nessita, but maybe you can advise on this: [20:13] can we have a modicum of patience with shoving stuff in stable now, please? [20:14] dobey: we need to land stuff to try to release before natty final release [20:14] nessita, I'm waiting for a re-review from rye on this branch I resubmitted because I forgot to set the target to stable, but he doesn't seem to be around. Should I just set it to approved? https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58324 [20:15] dobey: we have at least one critical, that Chipaca wants in (and not waiting to be SRU'd) [20:15] alecu: yes to that last bit. [20:15] alecu, thisfred: I have a third opinion regarding the branch [20:15] nessita: i understand that, but we need to not introduce changes that may cause more criticals as soon as we upload a new version [20:16] dobey: right, we were hoping to be able to land all of the stable-1-6 proposed branches (as far as i know). But maybe we should be more selective [20:17] alecu, thisfred, ralsina_: so, we need to distinguish the case where *the user* decided to disconnect, from the case where syncdaemon was disconnected from the server [20:17] becasue a rollout, for instance [20:17] nessita, I agree [20:17] nessita: well facundobatista's branch to bump the simultaneous transfer count concerns me. it doesn't have a lot of testing [20:17] alecu: ralsina_: in retrospect I think using the inhibitor was a bad idea: the logout inhibitor is meant to alert the user to things that may break when you log out, which is not the case for sd. We sort of abuse it, since we have no other way to notify the user of the fact that there's a sync in progress and they may want to wait a bit with logging out [20:17] dobey: right, __lucio__ wanted that in, we should ping him [20:17] nessita: why? [20:18] thisfred: because if syncdaemon was disconnected, we want to inhibit shutdown [20:18] thisfred: if the user explicitely disconnected, we don't [20:18] nessita: really? what if it stays down for hours [20:18] thisfred: well, the user can always click on "shutdown the same" [20:19] thisfred: is not like we force the user to wait no matter what [20:19] i think we should always inform the user at shutdown if there pending operations [20:19] nessita: sure, but we tell them they should wait, or things may break [20:19] if the user then chooses to explicitly shutdown, that's the user's choice [20:19] nessita: "Interrupting the program may cause you to lose work" [20:20] thisfred: we don't say that thing may break. Thunderbird will prevent shutdown if you have an ongoing compose email [20:20] nessita: so users won't disconnect. I wouldn'y [20:20] thisfred: and shutting down syncdaemon when it has still pending tasks, may cause work lose [20:21] really? [20:21] so the message is correct [20:21] thisfred: yeah, suppose this scenario: [20:22] you're travelling to argentina, you will use your laptop there. In order to work, you mandatory need a document that you're syncing with U1. The latest version of the doc is in your desktop, which will be inaccessible from Argentina [20:22] thisfred: you modify the doc, trusting that U1 will have the latest version [20:22] nessita: that's not the same as losing work [20:22] well you will appear to lose work, you won't actually lose the data [20:22] nessita: it just won't be in ubuntu one [20:22] thisfred: it is. A week without the document is lost work [20:23] nessita: losing work, I read as data loss, as in thunderbird losing an unsaved composed message [20:24] thisfred: losing work != data loss [20:24] nessita: I'm just telling you how I read it [20:24] thisfred: if you have to redo all you work in Argentina becasue you couldn't access your doc, that's lost work [20:24] and how other users may as well\ [20:24] thisfred: so, I agree we should not prevent the system to shutdown if the user requested the disconnection [20:25] is exactly the case that Martin Pool reported [20:25] thisfred: but if syncdaemon was disconnected becasue a rollout, we should alert the user [20:25] alecu, ralsina_: ^ [20:26] i think we should always alert [20:26] nessita: I don't think that's what he reported: he was not connected because syncdaemon lost his credentials right? [20:26] dobey: I agree is the easiest solution, but please read the report in bug #764057 [20:26] Launchpad bug 764057 in ubuntuone-client (Ubuntu) (and 1 other project) ""program is still running: ubuntu one file synchronization in progress" at logout (affects: 2) (heat: 10)" [High,Triaged] https://launchpad.net/bugs/764057 [20:26] thisfred: syncdaemon never looses credentials :-) the credentials were not in the system [20:27] thisfred: the key, from my POV, is: connection 'Not User With Network' [20:27] nessita: well, still: it's emphatically not the case that he disconnected intentionally [20:27] "Not User", that means there is no user intention to keep working with SD [20:27] nessita: reading that, there are multiple complaints [20:27] thisfred: I think we should alert if the connection is "With User" [20:28] dobey: re-reading to see what you mean [20:28] nessita: and i agree with poolie that the UI is not the best there, but I don't think the answer is "don't alert the user" [20:28] dobey: is syncdaemon status is "Not User", I don't think we should alert [20:29] nessita: i don't agree [20:29] dobey: why not? [20:30] I think we should just change the message, from "file synchronization is in progress" to "files pending synchronization" [20:30] well the user may have disconnected temporarily but forgotten to reconnect [20:30] nessita: because the UI is the problem, not the feature [20:30] alecu: I'm agreeing with that more and more [20:30] we should fix the UI, not break the feature [20:30] dobey: what UI are you referrring to? and how is broken (I'm not following)? [20:31] nessita: see poolie's 2nd/3rd bullet points in that bug description [20:32] alecu: nessita: I'm rejecting the branch, as I don't think anyone likes it as is. discerning between with user and not user will be a much bigger change, and alecu's proposal I like, but it's too late for that IIUC [20:32] thisfred: it's too late for ANY change, really [20:33] "There's no obvious way to see what if anything is happening or how long it will take." is a very good point [20:34] and i just added a comment on the bug itself [20:34] dobey: right, but there is nothing we can do at this point regarding that part of providing more visibility for file sync [20:34] alecu: maybe the control panel should have a progress tab... [20:34] nessita: really, there is nothing we can do at this point for anything regarding that UX. it is not critical. [20:35] yes, this bug is not a blocker [20:35] we really should not be changing any code at this point in 11.04, bug alas [20:37] alecu: do you have any ideas on what could cause bug #762722 ? See my last comment. [20:37] Launchpad bug 762722 in ubuntuone-control-panel "Opening the Control Panel during sync removes progress indicator (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/762722 [20:39] thisfred: https://bugs.launchpad.net/ubuntuone-control-panel/+bug/762722/comments/4 [20:39] Launchpad bug 762722 in ubuntuone-control-panel "Opening the Control Panel during sync removes progress indicator (affects: 1) (heat: 6)" [High,Confirmed] [20:40] so let's keep it as it is and improve it on the first SRU or whenever possible [20:41] dobey: hmm, yeah that may be it. Once again having two different applications here bites us. [20:45] let's not lose perspective this is a rare corner case. It hits people who are disconnected without knowing, have non-synced files, try to logout, and look at the actual inhibition warning ;-) [20:46] holy moly xscane is a horrible UI [20:46] ralsina_: yeah that one's not so serious. 762722 *is*, though, in my opinion [20:47] dobey: need to scan stuff? there's supposed to be easier ones out there. [20:47] * ralsina_ looks at 762722 [20:47] ralsina_: since the progress bar may actually tempt people to click the launcher [20:47] yeah [20:47] ralsina_: i just got a scanner. i need to write an app to fulfill the reason for buying it though. was just trying it out in gimp to see if it worked [20:47] dobey: I use simplescan [20:47] and while just plugging it in, is all i really needed to do, scanning in GIMP gives you some horrible UI [20:48] ralsina_, thisfred: I don't find that bug serious, but confusing for users [20:48] thisfred, I guess it has to do with the "application that is updating it falls off dbus" [20:48] yeah, not a blocker, just annoying [20:48] does the bar reappear eventually? [20:48] thisfred: were you able to reproduce? [20:48] nessita: losing the progress bar is not serious? [20:48] thisfred: i don't think it's a blocker. [20:48] ralsina_: it only reappears when logging out and in a again [20:49] thisfred: and i fear it may mean a much larger and coordinated change to fix [20:49] thisfred: I don't think we can fix it now [20:49] thisfred: not really, is confusing and annoying [20:49] dobey: I agree, it's not, but it's more serious than the other one [20:49] thisfred: but you could start a branch so we have a fix ready eventually [20:49] thisfred: i do agree with that :) [20:49] thisfred: we can land into trunk all that we need, that will go to O [20:49] ralsina_: well, if dobey's right, we can't fix it, it needs changes to unit [20:49] in any case, compared to the bugs we had a month ago? We are looking good ;-) [20:49] y [20:50] thisfred, so, it's not that it disappears when the application falls off dbus, but instead when the application shows up on dbus again. [20:50] BTW, since almost everyone is here, we seem to be getting good feedback from users, so pat yourselves on the back once or twice in my behalf. [20:51] ralsina_: thanks! [20:51] * nessita is happy [20:51] can't we "poll" to make the bar appear? From syncdaemon, or something. [20:51] ralsina_: it's all for the good of science [20:51] to celebrate, let's make some mate! [20:51] thisfred, and it looks like a launcher bug with non-windowed applications. [20:51] (yes, i picked up portal 2 already) [20:52] dobey: looks like a very fun game. If I were a gamer, I'd buy it. [20:52] alecu: I think the problem is that the launcher opens u1cp, and from then on, the launcher stops listening to syncdaemon. We are probably the only ones that have two applications talk to the same launcher [20:52] ralsina_: even if you're not a gamer, you should buy it and play it. it's worth it just for the humor [20:52] dobey: you're not working today right? because I would need a review from you (branch that fixes the services layout issue) [20:52] thisfred: we are the only ones using the "non-windowed app talks to launcher" too [20:52] right [20:53] nessita: no. but if there are problems with tarmac, i would consider it critically urgent and i should be pinged about it [20:53] thisfred, the thing is that the launcher is resetting the icon when the control panel claims the dbus name [20:53] dobey: we all try to respect your day off :-) [20:53] ralsina_: well the thing is we use the launcher for a windowed app, *and* for syncdaemon [20:53] thisfred: yeah. So we are a corner case of a corner case. [20:53] thisfred, and ignoring whatever syncdaemon set there. [20:53] dobey: ok, sorry for not pinging you, from my POV it was workaroundeable by running tarmac myself [20:54] alecu: right [20:54] alecu: that's not fixable by us though right? [20:54] nessita: unfortunately, running tarmac as you presents other problems, as it doesn't hit some cases that the normal tarmac setup does, when running the tests [20:54] thisfred, a workaround would be to have syncdaemon re-enable the progressbar every time it updates it [20:54] and whenever the cp starts. [20:54] dobey: such as? [20:54] alecu: hmmmm yes, but that's working around a bug [20:55] ralsina_, absolutely. And non-blocker [20:55] so, there are NO blockers? [20:55] alecu: so do you think I should file a bug against unity? [20:55] nessita: well i guess you have a full checkout of the branch for merging with tarmac, which can cause problems, and which masked the broken test which was using bzrlib.WorkingTree to find the config file [20:55] ralsina_: there is a blocker already landed in stable-1-6 [20:56] nessita: ok, ack [20:56] dobey: right. You tweaked your tarmac to cover that case? [20:56] nessita: there are also occasionally some other issues that crop up when it is being run under cron [20:57] nessita: no, tarmac by default uses lightweight checkouts, so the test failed because of the way bzrlib.WorkingTree was being used. so i had to fix that test when i discovered it was broken [20:58] nessita: there were some other gio issues though, that didn't crop up when you ran tarmac, and did on my tarmac instance, which i did have to work around; because fixing gio is a much harder problem to deal with there. [20:58] nessita, ralsina_: also, my server is on newer hardware now; so the going to sleep problem should be resolved [20:58] dobey: awesome! [20:59] dobey: go on a trip somewhere this weekend to test it ;-) [20:59] dobey: ok, good to hear [21:00] alecu: did you add the IRL test instructions to https://code.launchpad.net/~alecu/ubuntu-sso-client/changed-secrets-spec/+merge/57947 ? [21:00] and, can I have a second review for https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/services-improve/+merge/58362 ? [21:00] nessita: on it [21:01] thisfred: thanks! [21:01] ralsina_: i have enough trips lined up for the next few months already :) [21:01] nessita, sorry, no. The IRL test instructions would be "install the gnome3 ppa" [21:02] alecu: you need to get rye to re-approve your resubmitted branch. [21:02] alecu: how did you test it? using the gnome3 ppa? [21:02] dobey: rye is not around, and that's the critical we're trying to land [21:02] dobey: and, as you can see in the MP, he already proposed before the branch was superseeded to target stable [21:03] nessita, I only tested it with the integration tests in the branch [21:03] dobey: can you please approve? [21:03] or we can have alecu's approving [21:04] alecu: fair enough, the thing I care the most is that the code keeps working with non-gnome3-ppa code [21:04] nessita: it should wait for rye to review it. it's not going anywhere before tomorrow anyway. so unless rye is on holiday the rest of the week, i think we should wait for him [21:04] nessita, the tests in the branch simulate both a gnome2 and a gnome3 keyring [21:04] the integration tests, that is. [21:05] dobey: I would like nightlies built from that branch before tomorrow for testing [21:05] dobey: so please, work with me on this one :-) [21:05] dobey: I can ask ralsina_ to second-approve [21:06] nessita: nightlies are built off trunk, and the trunk branch already landed with that change. [21:07] dobey: shouldn't we build nightlies with stable-1-6 until natty is out? [21:07] nessita: you already have my second approve on that one [21:07] on srvices-improve, I mean. [21:07] nessita: no [21:07] nessita: anything in stable-1-6 should have been landed in trunk first. [21:08] dobey: like you said, is already on trunk [21:08] ralsina_: can you review, please, https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58324 ? [21:09] nessita: yes, and can test from trunk. nightlies are built from there. and are in need of packaging updating already [21:11] nessita: on it! [21:13] nessita: can't say I understand what that branch is doing. Passes the tests, though. [21:13] alecu: can you please explain to ralsina_ what the branch does? [21:14] nessita, you don't want to land that on stable, do you? [21:14] alecu: why not? [21:15] nessita, I think we should not be risking unstability for all users in order to fix a bug that only affects users of a ppa [21:15] nessita, and that applies to every branch we are reviewing today. [21:16] nessita, oh, it's the "other signal" branch. [21:16] sorry7 [21:16] alecu: what do you mean "to fix a bug that only affects users of a ppa"? [21:16] right [21:17] I though we were discussing the gnome3 ppa branch [21:17] alecu: this is not the ussoc one :-) [21:17] right, right. I'm sorry [21:17] * ralsina_ is getting branch-dizzy [21:18] ralsina_, ok, let's go back to this branch. [21:18] https://code.launchpad.net/~alecu/ubuntuone-client/the-other-signal-stable/+merge/58324 [21:18] alecu: yes. What's it doing? [21:19] this branch creates a new dbus proxy for the credentials management interface, and listens to the CredentialsFound signal [21:20] ralsina_, this is so the gnome-settings-daemon plugin can know when syncdaemon or couch has successfully requested authentication tokens [21:20] oh, this is in the gsd plugin? [21:20] and install the "Ubuntu One" bookmark. [21:20] forget that, I get it now [21:20] ok, anyone field tested it? [21:20] ralsina_, well, this is for the gsd-plugin, but it's not in the gsd plugin itself. [21:21] ralsina_, I did, nessita did, roman tested trunk. [21:21] alecu: yes, figured out while I was typing :-) [21:21] * ralsina_ should type slower or think faster [21:21] ok, +1 then [21:21] ralsina_, cool [21:25] * alecu heads for kindergarten, to pick up Amelia. === alecu is now known as alecu-away [21:28] thisfred: does this make sense to you? bug #746276 [21:28] Launchpad bug 746276 in desktopcouch (Ubuntu) "desktopcouch-service crashed with RuntimeError in start_couchdb(): Can not start couchdb. (affects: 13) (dups: 5) (heat: 248)" [Undecided,Confirmed] https://launchpad.net/bugs/746276 [21:29] nessita: I've not seen that before. Let me look at the code [21:29] thisfred: thanks! [21:34] nessita: this error is raised when couchdb does not start within 10 seconds or so. It's not clear why it wouldn't, though [21:35] oops [21:35] thisfred: you ran away! [21:35] :-) [21:35] if only it were that easy ;) [21:35] thisfred: can you do some follow up on that, maybe requesting logs? it has some duplicates [21:35] yep [21:35] genial! :-P [21:41] alecu-away: approved [21:51] ok, back to portal2 [22:03] dobey: noooooooooo dont play portal2 play heroes of newerth with me :) [22:09] so, my ubuntuone-syncdaemon is still chewing cpu [22:09] is there no way i can make it forget it ever saw any files on this machine? [22:24] see you all tomorrow people. Have a nice whatever it is over there :-) [22:26] ralsina_: hah ill see you tomorrow hopefully :) [22:26] bye! [22:28] bye ralsina_! [22:30] ok, gotta walk the dog, later all [22:44] * nessita -> eods