=== gpc is now known as Gulfstream === Gulfstream is now known as gpc [01:05] Good morning friends :) [01:06] morning TrickyJ [01:06] karni: hey!! morning :) howzz u doing !! [01:07] TrickyJ: heh. busy as hell :/ [01:07] Hmmm.. Yaaa !! i can understand same out here :( waiting for ma weekends !! :( [01:07] Desperately need one... [01:08] I'm waiting for like 2-3 weeks in the future. Regular (busy!) weekends are no good.. [01:08] gotta get back to my stuff.. [01:08] TrickyJ: have a gread day [01:09] karni: thanks buddy :) [01:33] how you start the u1 gui to register a new box, cant seem to find the icon? [01:35] espen77: me menu -> ubuntu one [01:35] upper right [01:36] karni: fresh dayly natty install, not there [01:36] espen77: ah, dunno then.. [01:36] espen77: try https://wiki.ubuntu.com/UbuntuOne/FAQ [01:37] karni: yeah, thanx [01:47] karni: for some reason the u1 control panel wasnt installed "sudo apt-get install ubuntuone-control-panel-gtk" fixed it [01:47] espen77: oh.. glad you fixed it! === mr_ian_ is now known as mr_ian [08:10] rye, are you there? [08:10] otto___, hi, yes, i am here [08:13] rye, last week I spent time debugging why I can't use U1 to sync my photo archive and the conclusion was, that a few of my video files are too large (bug #702309) [08:13] Launchpad bug 702309 in ubuntuone-client "Ubuntuone-process makes hunders of files like /tmp/tmpXXXX, fills hard dirve (affects: 1) (heat: 6)" [Undecided,Confirmed] https://launchpad.net/bugs/702309 [08:14] rye, now I'm debugging why I can't use U1 to sync my music archive and the reason seems to be that some file names are too long..? (bug #706816) [08:14] Launchpad bug 706816 in ubuntuone-servers "Ubuntu One deletes files on the server side if file names are too long (affects: 1) (heat: 258)" [Undecided,New] https://launchpad.net/bugs/706816 [08:16] otto___, what is the full path to 'Antonio Vivaldi - Concerto No. 1 in Mi maggiore, RV 269 "La primavera": III. Danza pastorale allegro.oga' ? [08:19] rye: /home/otto/Ubuntu One/U1 bug test Antonio Vivaldi/Le quattro stagioni, Op. 8 No. 1-4 (Musici di San Marco feat. conductor: Alberto Lizzio)/Antonio Vivaldi - Concerto No. 1 in Mi maggiore, RV 269 "La primavera": III. Danza pastorale allegro.oga [08:23] otto___, ok, that looks like bug #657195 which I will now request to be backported to lucid, but I need to test the behavior now [08:23] Launchpad bug 657195 in ubuntuone-client (Ubuntu Maverick) (and 3 other projects) "syncdaemon deletes files when local machine errors out on creation of partial (affects: 1) (heat: 33)" [Undecided,New] https://launchpad.net/bugs/657195 [08:35] rye, I don't think so, since the files initially actually do transfer to U1 so the partial phase filename shouldn't be an issue.. [08:42] rye: I can see all the files at https://one.ubuntu.com/files/, even the aaaaa......txt with 255 chars. [08:42] otto___, yes, that looks like that bug (which is really a bug and the fix should be backported) but I am trying to find out why the files get deleted [09:05] otto___, is your /home filesystem encrypted? [09:07] otto___, on any of your machines that are in the same Ubuntu One account? [09:11] rye: this computer is not encrypted, but another computer (my laptop) on this account is [09:11] otto___, bug 344878 [09:11] Launchpad bug 344878 in linux (Ubuntu) (and 2 other projects) "file name too long when creating new file (ecryptfs_lookup: lookup_one_len() returned [-36] on lower_dentry) (affects: 51) (dups: 5) (heat: 317)" [Undecided,In progress] https://launchpad.net/bugs/344878 [09:15] so, the initial assumption is - the file gets uploaded by machine w/o that filename restriction, then the file gets downloaded to another machine that firbids the creation of a filename of that length. So the file is not there. Upon restart the rescan notices that file is not available there (since they are all Queried()) and removes the file on the server [09:17] rye: I'll go and turn on my laptop. If that is the cause, then the files should be deleted when the laptop tries to sync. [09:18] otto___, thanks, performing similar test here - both laptops are running lucid, right? [09:20] rye: yes [09:26] aquarius: you around? [09:28] jderose, I am, but I'd like to not have a call, for two reasons [09:28] 1. I don't feel very well, so concentrating on what you're saying is not gonna be easy :( [09:28] (1) is already enough reason, np. :) [09:28] 2. we need to get ralsina in on the conversation, because he runs the desktop team and so would be making a decision on whether we have time to do what you're aiming for in this cycle [09:29] aquarius: cool, just let me know when is a good time. feel better! [09:30] jderose, I could probably concentrate on a text chat so we can work out the details, though. I'm not quite following what you want; you want a database which *anyone* can pull? [09:30] rye: opened laptop, files disappeared. Also did some manual testing: on the laptop a maximum file name length is 143 chars before Nautilus tells file name is too long. [09:31] the files are still visible at https://one.ubuntu.com/files/ [09:31] aquarius: yeah, so the idea is that we store the metadata for the entire novacut catalog in a database that everyone pulls down... it's just an easy way to let people know when new episodes are available and whatnot. [09:32] aquarius: then we use a 2nd database that works the normal-per-user desktopcouch way... this stores recently watched, user ratings, stuff like that [09:32] jderose, ok, and...do you want to put something on their machines which knows how to pull that database, or do you want desktopcouch to somehow magically know about that DB? [09:33] rye: I guess it takes some time before they are deleted, but they are not downloaded to the laptop client on ecryptfs. [09:34] it doesn't have to be in desktopcouch that handles it, as long as desktopcouch dosen't try to sync back the other way. but i think this is a pretty cool use case... sort of like a better rss feed [09:34] rye: hmmm. although I was able to make a file called "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaAAAAAA.aaaaaaatxtaaaaaaaaaaaaaaaaaaaaa" on the laptop which uploaded nicely. [09:35] jderose, ok. I am 81.6% sure that you'll be able to do this without assistance from us. In 11.04, we'll have the notion of "limited access OAuth tokens" for couchdbs at U1, where you can create an OAuth token which can read one of your databases but not write it. [09:35] ah, very cool.... so i this is way to do a one-way share of a database with someone else, in other words? [09:36] jderose, so, what you do is, create one of those tokens for this database. Then you write a desktopcouch replication provider, with that token hardcoded into it, and make the replication provider part of novacut. The replication provider pulls down that DB, with the oauth token, and when asked to push that DB it just does nothing. And you add the db to excluded_names so that it doesn't get pushed to my U1 storage. [09:36] the reason I'm only 81.6% sure that this will work is that I'm not sure if excluded_names databases are excluded from the U1 replication provider or from *all* replication providers. [09:37] need to talk to CardinalFang or thisfred about that. [09:37] gotcha [09:37] in the limit case, if excluded_names excludes your replicatin provider too, you can run your own replication daemon rather than having DC do it for you with the replication providers, but that's a bit lame :) [09:38] otto___, ok, i have already received filename too long error with your filename, waiting for it to finish sync my test data and then will restart the syncdaemon to see whether next local rescan will trigger the removal [09:39] aquarius: this is some design work James Raymond has been doing for player catalog (best in chrome ATM) - http://dev.novacut.tv/ [09:39] jderose, so, we need to confirm with CardinalFang and thisfred and possibly vds and mandel, but does the above make sense? [09:40] aquarius: so the idea is that this webapp is save in couchdb, talks to couchdb to traverse through catalog, search, and so on. when running as native app, we shove this is webkit, use desktopcouch and all the ayatana goodness... but we can also deploy as vanilla web page [09:40] cancel_upload_and_delete_on_server, awesome [09:40] aquarius: yeah, i think it makes sense [09:41] jderose, nice. You'll need extra cleverness to put a couchapp in desktopcouch, mind, because requests back need to be oauthed, but it's doable; I've got a little webkit around somewhere which oauths all requests [09:41] otto___, yes, encrypted filenames bring unexpected behavior for syncdaemon... testing the backport branch... [09:42] aquarius: i'm working on the oauth issue right now, it's a bit of a pain, but i'm making progress [09:42] jderose, what you do is hook navigation-decision-required (or something like that) signal in the webkit object, and then oauth-sign the associated networkRequest URL [09:44] aquarius: have you actually got that to work? man, i tried so much, but i can't get the webkit handler to work right... if you have sample, i'll love to see it [09:44] jderose, I'm sure I did this for a demo at some point, but I can't find it. [09:44] ok, with generations-enabled client it is a bit better - the files are not removed, though I am not sure whether that is a feature or a corner-case [09:45] jderose, do you have what you've done so far? and I can look at fxing it? [09:47] aquarius: hmmm, looks like i don't have it.... test dir that i blew away without committing, but if you've made it work, at least i know it's possible [09:49] aquarius: ah, but one thing... will it work with ajax requests? i really want to avoid rebuilding pages, which get's slow. but i've done some tests with ajax talking directly to couchdb... very responsive [09:50] aquarius: anway, i'll let you get to felling better... thanks. just ping me when your up for a skype! :) === IAmNotThatGuy is now known as M0hi [09:57] jderose, the request is still made by webkit, whether it's initiated from javascript or from a link [09:57] I'll try and gen something up [09:58] aquarius: ah, okay [10:22] rye: marking bug #706816 as duplicate of #657195 [10:22] Launchpad bug 706816 in ubuntuone-client "Ubuntu One deletes files on the server side if file names are too long (affects: 1) (heat: 258)" [High,New] https://launchpad.net/bugs/706816 [10:23] otto___, ok, i have cherrypicked a patch for this and will talk to dobey today regarding the terms of lucid-backport-work [10:24] jderose, http://paste.ubuntu.com/557580/ [10:24] jderose, found the thing I built before [10:24] jderose, that's by no means perfect, and it's broken under natty, but the principle is sound I think [10:24] aquarius: awesome, thank you! woot! [10:25] aquarius: cool. well, i think i'm gonna wonder off to bed... thanks again. [10:25] rye: thanks! [10:25] jderose, no problem [10:25] way I found that file: search irc logs until I found myself talking about it; note the date I was talking about it; use gnome-activity-journal to see all files I edited during that day. This zeitgeist stuff might have some legs. :) [10:26] hehe :) [10:26] otto___, thanks for pointing this out, I tried to confirm this issue under various filesystems and failed to get the files removed, all I needed was encryptfs and non-generations-enabled syncdaemon version [10:26] well, non-generations-enabled version was enough [11:30] hello all [11:33] duanedesign, hi, just a thing to know - ecryptfs imposes its own limit over the filename length, bug 344878, but with ubuntuone (<= Lucid Lynx) it gets more interesting - bug 657195 [11:33] Launchpad bug 344878 in linux (Ubuntu) (and 2 other projects) "file name too long when creating new file (ecryptfs_lookup: lookup_one_len() returned [-36] on lower_dentry) (affects: 51) (dups: 5) (heat: 317)" [Undecided,In progress] https://launchpad.net/bugs/344878 [11:33] Launchpad bug 657195 in ubuntuone-client (Ubuntu Maverick) (and 4 other projects) "syncdaemon deletes files when local machine errors out on creation of partial (affects: 2) (dups: 1) (heat: 18)" [Undecided,New] https://launchpad.net/bugs/657195 [11:37] ahh [11:37] and i finally got around disabling filename encryption on my files [12:04] good morning! [12:05] o/ === teknico is now known as teknico_away [13:25] * CardinalFang wonders what madness aquarius has been saying. [13:34] CardinalFang, if I add a db to excluded_names, does that exclude it from *all* replication providers, or just the ubuntuone replication provider? [13:47] alecu CardinalFang dobey mandel nessita thisfred vds: standup in 13 minutes! [13:47] ack [13:47] 10/4 [13:48] ralsina: ok [13:51] aquarius, I don't remember. Looking. [13:52] aquarius, all. [13:52] CardinalFang, damn. [13:54] aquarius, it is kind of silly, now that I look at it. [13:54] CardinalFang, so...did you read in the scrollback what jderose was trying to do? My proposed solution (add new replication provider which only replipulls, and hardcodes the token to do so, plus adding the pulled DB to excluded_names so that it doesn't get pushed up to U1) won't work if excluded_names will exclude it from the custom replication provider as well [13:54] CardinalFang, I can see why it happens, though; the management DB should not be pushed to other replication providers. [13:55] aquarius, I did read it, yes. [13:55] CardinalFang, so, what are alternative ways of making it work? I mean, one way is for jderose to run his own replication daemon, but that's a bit hacky and inefficient. [13:56] alternative approach: have novacut post a *continuous* replipull replication request to that DB when it starts up (and list that DB in excluded_names, of course) [13:57] aquarius, that would do it. I fear "continuous" still. As long as his servers can scale, it might be okay. [13:58] But, sockets are a scarce resource. [13:58] they're not his servers; they're our servers. [13:59] * CardinalFang reads again. [13:59] Oh, standup. [14:00] me [14:00] Standup [14:00] me [14:00] me [14:01] alecu, dobey, mandel, thisfred, vds? [14:01] me [14:01] me [14:01] me [14:01] me [14:02] * ralsina is going to get dobey's phone and wake him up long distance 5 minutes before standup starting tomorrow [14:02] nessita, go ahead! [14:02] DONE: more beautification branches. Filled tons of bug report to keep track of design issues. Chased reviewers a bit. Missed the CoP meeting since my laptop died for 30 miutes approx, apparently due to heating issues. [14:02] TODO: bug #706894, bug #706906, maybe bug #704434 [14:02] BLOCKED: nopes [14:02] LOVE: the weekend with the family next to the pool [14:02] HATE: this freaking and unbearable heat :-) [14:02] NEXT: CardinalFang [14:02] Launchpad bug 706894 in ubuntuone-control-panel (Ubuntu) (and 1 other project) "Improve 'Loading' messages, replace with a stacked spinner (affects: 1) (heat: 6)" [Medium,Triaged] https://launchpad.net/bugs/706894 [14:02] Launchpad bug 706906 in ubuntuone-control-panel "Token info is logged (affects: 1) (heat: 6)" [Critical,Fix committed] https://launchpad.net/bugs/706906 [14:02] Launchpad bug 704434 in ubuntuone-control-panel "Stop the backend service when UI is done (affects: 1) (heat: 6)" [Medium,Triaged] https://launchpad.net/bugs/704434 [14:03] DONE: Fixed replication bug, #705448, in desktopcouch. Testing my fix for another bug WRT dbus-client calls to service from within service. [14:03] TODO: Test. Wrap up playlist API, finally, damnit. Get reviews of all three. [14:03] BLOCKED: None [14:03] ralsina, please. [14:03] DONE: reviews, talked to lots of people, did my canonicaladmin work, read code. [14:03] TODO: reviews, management stuff, have HR fix things for me, someday actual coding ;-) [14:03] BLOCKED: no [14:03] thisfred! [14:03] DONE: messaging / refactoring branch TODO wrap up and land | progress bars BLOCKED : no [14:03] vds!@ [14:03] DONE: first branch for #701029 landed second on it's way [14:03] TODO: starting the views [14:03] BLOCKED: nonope [14:03] mandel: please [14:03] DONE: Added extra param to u1trial so that paths can be ignored. Updated windows branches of u1client to use u1trial. found a bug in Filesystem notification on Windows. Fixed Windows branch following alecu review [14:03] TODO: Fix fs notifications. finish IPC. [14:03] BLOCKED: no [14:03] /me passes alecu a windows machie [14:03] DONE: branch for new events on SD (command queued, command unqueued) needs review. Initial work on the code to aggregate file sync events. [14:03] TODO: think about that a little more, do more code. [14:03] BLOCKED: no [14:03] HATE: radeon driver bugs that crash natty laptop [14:03] LOVE: working from the maverick desktop [14:03] EOM? [14:04] comments first [14:04] no comments here, other than I may melt myself any time soon [14:04] I want a u1cliente release tomorrow :-) [14:04] ralsina: +1! [14:04] nessita: go to a bar :-) [14:05] nessita, it's hot? [14:05] CardinalFang: 31C and raising [14:05] ralsina: I might, later today. Thanks! [14:05] mandel: I've been told we have a windows build box? [14:05] ralsina: really?!?! when, where? [14:05] mandel: and that canonical will buy us windows dev. tools licenses if we need them :-D [14:05] mandel: chipaca said you knew about it and would tell me ;-0 [14:06] mandel: so the answer is "it's probably mythical" [14:06] nessita, Nice. 11C here. It's the only time the weather is nice in Florida -- when all the plants are brown. [14:06] ralsina: oh, I think he is talking about the one used for wubi :) [14:06] no, no [14:06] I never said we had one [14:07] I said we needed to get one, or something like that. This is about the ticket to get one. [14:07] Chipaca: "one thing I forgot: windows buld box, mandel knows about it". Sorry I misunderstood you ;-) [14:07] Chipaca: ok then [14:08] CardinalFang: 33.8C right now! we're expecting 37C as per the weather forecast [14:08] So, about tomorrow's release... [14:08] ralsina, Chipaca: I can ask elmo, we must have a windows machine since we have signed the msi and we do distribute wubi, shall I? [14:08] anyone has a very good reason why not do it? As in "if we do it, we will kill our users puppies". [14:08] oh, one more DONE/TODO: some android-music coverart++ for beuno. [14:09] ralsina: I've been using SD all wekeend and except some minor bug reports, all seems pretty good [14:09] mandel: no, don't do that [14:09] nessita: same here [14:09] mandel: remember we were going to write a ticket asking for it to be set up, with all the details? [14:09] beuno, I don't know I have any cover art in tunes. I haven't seen any yet. Do you have a test tune for me? [14:09] mandel: and when I say "we" I mean "you" [14:10] Chipaca: yes I remmeber I was going to do it, I've got the email somewhere [14:10] Ok, so I think we are all now talking about stuff that was not on the standup. eom :-) [14:10] CardinalFang, I think I do, yes! btw, part of the reason I wanted to steal you away from ralsina was to see if you could look into the itunes-not-working-on-android thingie [14:10] Chipaca: but who do I send the mail to, rt? [14:10] beuno, itunes=mp4? [14:11] beuno: we forgot our "steal CardinalFang" chat friday :-) [14:11] ralsina, beuno, mind if I get these two bugs fixed first? [14:11] s/fixed/landed/ [14:11] CardinalFang: of course not [14:11] CardinalFang, yes [14:11] CardinalFang, not at all [14:12] ralsina, we did. Fridays are forgetful days [14:12] * CardinalFang hurries. [14:12] beuno, CardinalFang then let's talk after you are done with those bugs, ok? [14:14] * beuno nods [14:14] mandel: work with ralsina until it is done, then rt [14:15] CardinalFang: are "those two bugs" the ones preventing me from couching my replicator? [14:15] Chipaca: as in send it to ralsina, dicus it and then rt? [14:15] Chipaca, really only one of them does, I think. [14:16] clairemitchell: say hi to beuno and chipaca and everyone else [14:16] hello everybody! [14:16] hey hey clairemitchell! Welcome [14:16] thank you [14:17] clairemitchell: ooh! hi :) [14:17] I'm going to be working with Ivanka...just in at the deep end so when I surface with questions...you will know [14:17] ivanka: is this the clairemitchell for whom you asked permission to the roadmap? [14:18] ivanka has just stepped away from her desk... [14:18] :) [14:18] :) ok [14:18] hey clairemitchell! Welcome to the show that never ends. :) [14:18] nessita: alecu: does ubuntuone-client/control panel have its own .desktop file? I can't seem to find it on my system [14:19] thanks [14:19] welcome clairemitchell, you will probably hear a lot from people in this channel :-) [14:19] More than you want, even. [14:19] I'm all ears [14:19] thisfred: it does, are you running nightlies? [14:20] hi guys [14:20] beuno: received my e-mail :)? [14:20] nessita: I'm pretty sure I am, maybe it's just not installed atm [14:21] karni, I did! have not caught up yet [14:21] beuno: ack [14:21] thisfred: right, nigthlies packaing had a bug that was resolved last Friday [14:21] beuno: you seen the latest tweaks to google maps? [14:22] Chipaca, vectors and live traffic in buenos aures? [14:22] thisfred: let me land a couple of branches and then you can re try with nightlies [14:22] beuno: no, UI changes [14:22] nessita: right, looks like I have client, but not control panel. Installing now [14:22] ralsina: I just remembered that d-obey is on a day off today [14:22] Chipaca, have not. On the web? [14:22] nessita: ah, still no .desktop [14:22] nessita: oh, right, I gave it to him! (stupid me ;-) [14:22] beuno: yes [14:22] nessita: let me know if something needs review [14:23] thisfred: a new nightlies needs to be built, and for that a new branch needs to be landed :-) [14:23] thisfred: what nightlies revno are you running? [14:23] Chipaca, not sure I can tell what has changed [14:23] the layers stuff, maybe [14:23] beuno: the viewing options [14:23] beuno: layers, satellite vs map view, etc [14:24] Chipaca, right, yes, nice [14:24] nice way of hinting there's more [14:25] beuno: very nice, game-like; the older stuff was more consistent with the zoom&pan controls, so i expect them to change soon :) [14:25] Chipaca, agreed. Anything game-like is always a win [14:26] heh. or maybe we don't remember the bad game interfaces :) [14:26] mind you, I'd love to be able to hit ~ and get an api introspecting tool [14:26] ...because they fail instantly [14:27] nessita: control panel shows 0.5.1+r51 on maverick and natty [14:28] I love how the subsonic api is completely consistent in its inconsistency [14:29] mostly [14:29] see? :) [14:29] * thisfred implodes [14:30] getPlaylist takes an "id" parameter to specify what playlist you want. Guess what parameter "deletePlaylist" takes. [14:30] Chipaca: yes, it is [14:30] Chipaca, id, surely? [14:30] combined length of the songs? :) [14:30] ivanka: it is done [14:31] thisfred: 53 should have it, if nightlies are being built (can't confirm that) [14:31] aquarius: playlistId [14:31] thisfred: :) [14:31] Chipaca: thank you [14:31] Chipaca, not according to the spec it doesn't. [14:31] Chipaca, delete takes id, but create takes playlistid, IIRC [14:31] nessita: ok thx, let me know if there is any way I can help speed that up [14:31] hah! you're right [14:31] and our view is wrong :) [14:31] but create takes playlistId, yeah [14:31] create takes playlistId, but that's because you pass two sets of IDs; playlistID and songID(s) [14:32] OUR VIEWS ARE NEVER WRONG [14:32] right [14:32] so, bug in our view [14:32] I forgot. CardinalFang, beuno, merge proposal is in place. If it should stay where it is, then ok :) [14:32] at some point, we may want to propose a 2.0 api [14:32] beuno: no, no, it's not *wrong*. It's merely taking a sabbatical on being right. [14:32] the U1M create playlist, that is. [14:33] beuno: should I propose a branch to fix this? [14:33] karni, awesome, the ui is +1 from me, so it's up to CardinalFang to spare a minute or two to look at the code [14:33] beuno: do any of our clients depend on the broken view? [14:33] Chipaca, that would be most awesome [14:33] Chipaca, they do not [14:33] beuno: ok [14:33] Chipaca, OTOH, I can't see how to delete or edit playlists in teh android client [14:33] which is why this has not popped up sooner [14:34] tests are an awesome, awesome thing [14:34] shame I found this writing tests for something else :) [14:34] beuno: CardinalFang gave a written ACK IIRC, but I'll he happy for him to give it a +1 ;) [14:34] beuno, hahaha edit playlists, how naive you are [14:34] Right. [14:34] mandel: do you use the sd SYS_QUEUE_CHANGED message? [14:34] karni, land! [14:35] karni, or do I need to? [14:35] Chipaca, well, there are tests for this [14:35] beuno: /me panics as I've never landed/merged ;) [14:35] Chipaca, but they probably make the wrong assumptions [14:35] Chipaca, do you want to merge? [14:37] thisfred: seems like nightlies are not being built [14:37] beuno: merge who? [14:37] ralsina: have you got any build error for nightlies? [14:37] nessita: sad smiley [14:37] nessita: nope [14:37] * ralsina checks anyway [14:38] https://code.launchpad.net/~mkarnicki/ubuntuone-android-music/u1f-create-new-playlist/+merge/46963 [14:38] Chipaca, I meant CardinalFang :) [14:38] beuno: ah. No, I don't want to merge CardinalFang. [14:38] beuno: but thanks for asking [14:38] nessita: I have a green marck on a build on jan. 21 [14:39] * beuno is reminded of monkey island and randomly trying to merge objects [14:39] nessita: for controlpanel that is [14:40] karni, +1 [14:40] CardinalFang: thank you :) [14:40] ralsina: right, 2 new branches landed, so a new nightlies should be built. Maybe we need to wait a bit longer. [14:41] karni, for Android code, we don't have any good way to automate testing and conditional landing. So, push directly to lp:ubuntuone-android-music . [14:41] CardinalFang: roger, will do! [14:42] nessita: maybe dobey's server is taking the day off too [14:43] ralsina: I'll wait a bit longer, maybe the build takes more time than we think [14:43] nessita: could be. Let's give it 30 minutes. [14:43] sure [14:44] ivanka: if you're feeling like it, I'd appreciate a comment on bug #706920. If you're not feeling like it I won't weep, promise. [14:44] Launchpad bug 706920 in ubuntuone-client "Interface "a la Mac OSX" ? (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/706920 [14:45] ralsina, ping [14:45] aquarius: pong [14:46] ralsina, who's in charge of making sure that Evolution's couchdb addressbook works in natty? I've had a couple of reports that it doesn't [14:46] that would be rodrigo, I think [14:46] * ivanka looks [14:46] rodrigo_: ping? [14:47] beuno: can't merge, lp:ubuntuone-android-music belongs to 'Ubuntu One hackers' to which, sadly, I don't belong. [14:47] Ah. i'll do it. [14:47] I guess I will have to switch to evolution to eat that dogfood too. [14:47] But it's really really not my kind of dog food :-) [14:47] ralsina, nessita: installed the missing package [14:47] CardinalFang: Thank you. [14:48] hi dobey! what would the missing package be? [14:48] dobey: was that preventing nightlies builds (or they are pretty slow)? [14:48] ralsina, rodrigo_? really? he doesn't work for us :) [14:49] aquarius: I must be confusing him with another rodrigo :-) [14:49] aquarius: let me check :-) [14:50] aquarius: according to directory.canonical.com he does ;-) [14:50] ralsina, yeah, but he's rotated, right? [14:50] ralsina, us == U1, not us == canonical :) [14:50] aquarius: yes, he's rotated but promised to keep looking at bugs related to evolution plugin [14:51] aquarius: heh, I was scared because quassel shows opensuse in his personal info ;-) [14:51] * ralsina thought he missied important mail over the weekend [14:51] ralsina, ooh, smart. Although I think we should probably have someone test it end-to-end -- there are changes to evolution, to couch, to desktopcouch, lots of stuff has altered in this cycle so finding all the places where it's not working would be good [14:52] aquarius: yeah. Since we don't have rodrigo this cycle, the evolution stuff is somewhat neglected. I'll get on it. [14:53] ralsina, yeah, that's what I thought -- we don't want to let it fall through the cracks [14:53] aquarius: yep [14:53] * ralsina is really, really going to hate switching to evilution [14:55] nessita: nightlies are dependent on how busy the lp buildds are [14:56] karni, merged. Congratulations. [14:56] dobey: is there any place where I can see the build queue for our nightlies? [14:56] CardinalFang: \o/ Thank you! (my first merge to 3rd party!) [15:01] beuno: I'll be back a little later === bac` is now known as bac === mbarnett` is now known as mbarnett [15:16] nessita: not entirely, but you can watch https://launchpad.net/~ubuntuone/+archive/nightlies/+packages [15:19] thanks === alecu is now known as alecu-lunch [15:40] * alecu-lunch has lunch at gcoop [15:48] alecu-lunch: when ever you can, can you take a look at the fix_runner branch again? === teknico_away is now known as teknico [16:04] aquarius: have any bug numbers for the evolution problems you mentioned earlier? [16:05] ralsina, no; I just got asked "is evolution and U1 contacts supposed to be working in natty?" Ted (Gould) is the one who mentioned that to me specifically [16:06] aquarius, let me know the bug #, or just assign them to me [16:06] aquarius, but what problem is he having? [16:06] not syncing? [16:06] rodrigo_, from ted: " Evolution can't contact it. If I click on it Evolution says that the DBus messages are lost" [16:07] you now know as much about ted's problem as I do ;) === beuno is now known as beuno-lunch [16:07] aquarius, ok, that sounds like desktopcouch not running, so tell him to check that, or if not, quit evolution and run /usr/lib/evolution/e-addressbook-factory on a terminal and see the output [16:08] rodrigo_, yeah, could be -- this was why i wanted to be sure we'd tested it all the way through :) [16:09] rodrigo_, but...when you ask for the DC port, DC should get started up, no? [16:09] so it can't not be running [16:10] aquarius, yes, but if it crashes, evo gets a timeout [16:10] if it's complaining about dbus, it means it's getting a timeout when asking for the port [16:24] rodrigo_, http://i.imgur.com/s8bFO.png [16:26] getting the output of /usr/lib/evolution/e-addressbook-factory would me more helpful :-D [16:34] * ralsina luncch [16:36] dobey: thought i filed this on friday, but couldn't find it this morning (doh!) so i filed it (again): bug #706994 [16:36] Launchpad bug 706994 in ubuntuone-client "No ubuntuone-preferences in nightlies on Maverick - 1.5.2+r816~maverick1 (affects: 1) (heat: 6)" [Medium,Confirmed] https://launchpad.net/bugs/706994 [16:39] joshuahoover: because i marked it a dup of a bug i fixed on friday [16:39] dobey: oops, sorry about that [16:40] also i am 'out' today === beuno-lunch is now known as beuno [18:29] nessita: ping [18:35] joshuahoover: pong [18:36] nessita: a user is getting a message about "Cannot create an item in closed collection" when trying to add his computer...any ideas? (guessing keyring) [18:36] nessita: i searched around for this type of error message but came up empty [18:36] joshuahoover: yes, known bug, fixed already committed but not released [18:36] nessita: is there a workaround? [18:36] joshuahoover: he should unlock the keyring as workarounbd [18:36] nessita: ah, ok [18:36] let me look for the bug # [18:36] alecu-lunch: ping [18:37] joshuahoover: bug #703343 [18:37] Launchpad bug 703343 in ubuntu-sso-client "Ubuntuone fails to prompt for keyring password (affects: 2) (heat: 12)" [High,Fix committed] https://launchpad.net/bugs/703343 [18:38] nessita: thank you! === alecu-lunch is now known as alecu [18:59] nessita, pong [18:59] alecu: how can a user unlock the keyring? the same way one can lock it from seahorse? [18:59] nessita, yeah, that's the only way I know about [19:00] nessita, I don't know of any cmd line way [19:00] nessita, (other than using the dbus api) [19:02] alecu: right, thanks! [19:04] joshuahoover, to unlock the keyring: "System->Preferences->Passwords and Encryption Keys", then right click on the "Passwords: login" or "Passwords: default" folder, and then "Unlock". [19:05] joshuahoover, the password is the login password for the "login" keyring, or whatever password the user entered when he was prompt to create the "default" keyring. (this last case is when ubuntu is set to autologin) [19:05] prompt->prompted [19:34] karni, ping [19:38] CardinalFang, so, this playlists change, it would probably be worth pusing upstream [19:38] whatdya think? [19:57] beuno: pong [19:58] karni, heya [19:58] how's it going? [19:58] beuno: hi :) thanks for the mail [19:59] beuno: fine, wrapping up stuff as mentioned, time to move forward :) [19:59] karni, how's it looking to send me an apk? [19:59] we've got some time from the design team [19:59] beuno: will you be around for a while? [19:59] beuno: oh cool! [19:59] so I'm sending them everything I can to get them started [19:59] right, sure [19:59] karni, I will be around for a while, yes [20:00] beuno: I'll send you it asap [20:00] thanks karni [20:00] beuno: yw! [20:02] thisfred: ping [20:02] joshuahoover: pong [20:02] thisfred: i have a rather vague support request: Error mensager: *Error org.freedesktop.DBus.Error.Spawn.ChildExited: Process /usr/lib/desktopcouch/desktopcouch-service exited with status 1 [20:02] thisfred: that's it [20:02] thisfred: thoughts? [20:03] beuno: btw I did 37h Mon-Fri and dedicated the weekend for some university time [20:03] karni, no worries [20:03] Just so you know. [20:04] joshuahoover: we'd like to see the replication logs, located at: [20:05] ~/.cache/desktop-couch/desktop-couchdb.log* [20:05] thisfred: cool, i can request that...just wasn't sure if that rather vague error would cause you to go, "oh yeah, try this..." [20:05] joshuahoover: it seems the service is erroring out, but I have no idea why [20:05] thisfred: if you were a REAL magician, you would no why! [20:07] I know. also mensager, what language is that? Portuguese? [20:23] beuno: what worries me is the predictable slowness of the app at this time of day. makes me wanna display windows-style dialog box "Connecting is taking more than usual...". the worst is, I can't do much about it. I wanted to test the thing before sharing the apk, and if I were a regular user, the app would be long gone from my phone. connect+authenticate+setup stage (which normally is ~6 secs) took 3-4 minutes :(. and I know you guy [20:24] beuno: ouch.. if I was cut off, tell me. I'll paste the part of msg. irssi is not splitting up irc messages :< [20:24] karni, __lucio__ is looking into it [20:24] beuno: thank you [20:40] beuno: I'm strongly uncomfortable sharing an apk I can't test in live environment beforehand. Please let me work on it today more (you testing it today would be waste of time considering the slowdown :< ), and you'll have it by tomorrow. [20:41] karni, things are getting back to normal now [20:41] I keep getting notification my u1 storage is full when in reality it isn't. I had this problem back when Maverick was in beta but people here helped me and a later update solved the problem. It's back. Any suggestions? I've already tried stopping everything, deleting the keyring tokens and starting from scratch, no dice. [20:41] beuno: a, let me check [20:41] karni, remember, you can also develop against staging [20:41] karni, it may take a little bit of time to re-balance all the connections [20:45] beuno: never managed connecting to staging, though I consulted lucio (although I'm fine developing offline before I test major new chunks). I still see servers are loaded, but I should notice any change, if it comes soon. [20:46] karni, why didn't you manage? [20:46] any specific errors? [20:46] 184.72.253.124 [20:46] is the IP [20:46] beuno: client gets stuck on set capabilities. I just change oauth stuff to staging.one.ubuntu.com and the client ip [20:46] ok, will try now [20:47] karni, there may be issues with the ssl certificate you need to ignore [20:47] yes, that's the ip [20:47] it'd throw at me if there were. anyway, I'll re-recheck that. [20:50] I keep getting notification my u1 storage is full when in reality it isn't. I had this problem back when Maverick was in beta but people here helped me and a later update solved the problem. It's back. Any suggestions? I've already tried stopping everything, deleting the keyring tokens and starting from scratch, no dice. [20:50] beuno: ok, I see the certificate issue now (didn't have that before, of what I'm sure o_O) [20:54] aha, didn't have it before because I used the old tokens which lucio said where good [20:54] This is the output of u1sdtool --status http://paste.ubuntu.com/557832/ [20:56] jimlovell777, I think all of our support staff is already gone, mind filing a bug? [21:09] beuno: looks like authenticate request is taking ages on staging, whilst on fs-1 it took a minute or so (but completed). precious time drifting away, I'm getting back to code. [21:09] cool [21:57] __lucio__: verterok: How does AQ_DOWNLOAD_COMMIT differ from AQ_DOWNLOAD_FINISHED ? [22:09] <__lucio__> karni, class DownloadFinishedNanny(object): [22:09] <__lucio__> """Supervises the download finished signals. [22:09] <__lucio__> It listens AQ_DOWNLOAD_COMMIT, and generates AQ_DOWNLOAD_FINISHED when [22:09] <__lucio__> the external world is ready for it. [22:09] <__lucio__> """ [22:10] __lucio__: a the nanny. thanks __lucio__ , I shouldn't have used you like that! :) [22:10] by external world I assume FSM, for instance [22:11] <__lucio__> karni, i am going to get rich with my new site, "let me grep that for you" ;) [22:11] <__lucio__> karni, it just keeps uploads from overwriting the current file if the file is open. [22:11] __lucio__: I didn't want to appear lazy :< (but there's http://lmgtfu.com ;) ) [22:12] __lucio__: a you see, now that's some good explanation :) thanks! [22:12] <__lucio__> karni, just kidding, and yes, it was a reference to that site (i use it a lot! :) ) [22:12] heheh [22:12] cool, thanks! [22:18] __lucio__: how is VM_UDF_CREATED as opposed to AQ_CREATE_UDF_OK ? clarification on VM_ is totally sufficient ;) [22:18] ah! [22:18] volume manager ;d [22:18] __lucio__: forget my last question [22:19] <__lucio__> done. [22:19] * nessita eods [22:19] bye!