[00:25] Hello. I would like to have my entire contacts database erased/reset/deleted. Is there anyone who can do this? [00:29] My contacts section on the ubuntu one website has been in an error state for 6 months or more, and I have 200+ contacts that can not be synced to any device and are not viewable/editable on the ubuntu one website. I can resync these contact from my phone to my ubuntu one account, but only if the database backend is reset or something. [00:31] fishscene: http://voices.canonical.com/ubuntuone/?p=943 [00:34] bummer [00:34] So no hope of having my contacts database reset even though it doesn't work when they say it should =\ [00:35] dobey: Thanks for the info though. I suppose I'll wait and bug them after they're done with the upgrades. :) [01:01] Night o/ === eu is now known as Guest49641 [07:26] hi [07:27] will there be an option to set folder as 'do not synchronize, just copy from machine xy to cloud'? [07:28] hrw@lumpek:Ubuntu One$ find . -name *u1conflict|wc -l [07:28] 85 [07:28] I use those data only on one machine and u1 daemon is running on few machines [07:30] and it is not fun when firefox lost extensions/options/bookmarks/passwords just because I keep profile on u1 [08:16] hrw, him what version of U1 are you running? [08:37] * rye adds the firefox profile dir to U1, sans Cache directory [08:45] Thoroughly happy Thursday, everyone! :) [08:45] JamesTait, hello :) [08:45] mandel: o/ [08:46] rye: precise one [08:47] rye: but problems were present even with oneiric one in past [08:47] hrw, I have started syncing firefox profile now, I suppose I should experience this shortly too [08:47] hrw, i suppose sqlite dbs were moved to conflict files? [08:51] rye: I use this laptop rarely - mostly on conferences [08:51] rye: http://paste.ubuntu.com/720497/ [08:52] rye: was planning to use U1 to share some of .dotfiles between machines but looking at mess which my firefox profile is I will rather skip it and keep them in git [08:56] [08:57] sorry - network problems at conference [10:06] hi! [10:06] gatox, buenas! [10:07] mandel, como va? [10:17] mandel, +1 to the branch [10:23] gatox, sweet, lets land it :) [10:24] gatox, you need to update your ubuntuone-dev-tools or you wont be able to runt tests today [10:24] gatox, the fix landed [10:25] mandel, ok..... [10:28] gatox, once you have the new tools, can you run the tests in all the other projects but ubuntuone-client and let me know if they pass? [10:29] mandel, ok, let me finish the upgrade..... i'm upgrading several packages [10:34] mandel, brb, have to restart [10:55] gatox, ok, I'm in 12 failing tests.. is good compared to 200 :D [10:56] mandel, jeje i'm looking for the deb..... upgrade didn't do the trick it seems [10:56] gatox, just do a python setup.py install from lp:ubuntuone-dev-tools [10:56] gatox, maybe the build takes a little too long [10:56] mandel, ahhhhok [10:59] gatox, I fear that we also have broken tests in the other projects which means that we will be block when trying to land anything :( [11:00] the sooner we fix this.. [11:00] mandel, i'm running the tests [11:00] mandel, woww....... something is really wrong [11:01] mandel, wait, i need to test it with this branch too: https://code.launchpad.net/~tcole/ubuntuone-dev-tools/qname-in-errors/+merge/80549 ? [11:01] because i'm getting 400 failures [11:02] gatox, fkndasasd [11:02] yep...... ¬¬ [11:02] gatox, a cagarla! where are you getting those? [11:03] mandel, it seems that everything is related to this: ubuntuone.devtools.testing.txcheck.TXCheckTest.runTest [11:03] are you sure the branch landed? [11:04] [FAIL] [11:04] Traceback (most recent call last): [11:04] File "/usr/local/lib/python2.7/dist-packages/ubuntuone-dev-tools/ubuntuone/devtools/testing/txcheck.py", line 355, in run [11:04] raise problem [11:04] ubuntuone.devtools.testing.txcheck.MissingReturnValue: MissingReturnValue for ActivationDetectorTestCase.setUp [11:04] ubuntuone.devtools.testing.txcheck.TXCheckTest.runTest [11:04] one of the cases [11:05] gatox, yeah, I know.. the way I solve it is by using a @defer.inlineCallbacks in the setUp because the setUp in theory returns a deferred [11:06] yeah, the txcheck stuff checks for common problems when chaining deferreds in setUp/tearDown [11:06] you need to make sure that setUp and tearDown return deferreds [11:06] and that those deferreds are chained with the deferred obtained from calling super().{setUp,tearDown} [11:07] tcole, yep, so @defer.inlineCallbacks is the way I have been fixing those :) [11:07] that tends to be the easiest approach yeah [11:08] just add the decorator and yield the return value of super().etc... and you're good [11:09] tcole, on it, problem is that there are lots :) [11:15] let me know if I can help [11:25] hrw, i am not sure you would want to have the whole firefox profile under ubuntu one control. the performance impact on the disk is substantial. Zeitgeist is logging the changes and it is being fed a constant stream of changes for sessionstore.js and other files [11:42] gatox, tcole please take a look: https://code.launchpad.net/~mandel/ubuntuone-client/fix-tests/+merge/80554 [11:42] * mandel walks dog [11:44] ack [11:46] mandel: +1 [12:40] mandel, +q [12:40] +1 === eu is now known as Guest15678 === Guest15678 is now known as adorilson [13:05] gatox, tcole cool :) [13:06] gatox, I'll move to the control-panel to get that fixed :) [13:06] mandel, ok!! [13:13] rye: zeitgeist is one of things which never gets installed on my machines. [13:20] hello everyone [13:21] nessita, hi [13:21] ok, I have to say this here: La puta!!, 662 fail in control panel!! [13:21] nessita, hello, we have had better days :P [13:21] mandel: why? [13:22] hi gatox [13:22] nessita, i couldn't update my branches, because we are working in the broken tests with mandel [13:23] nessita: we finally landed/enabled the correctness checks for setup/teardown in twisted tests [13:23] it's been ... enlightening [13:23] heh [13:23] tcole: hi there! did you check the rest of the suite before landing the fix? :-) [13:24] nessita: ? [13:24] maybe we should add an option to toggle that check on [13:24] tcole: I mean, did you confirm the u1client and u1controlpanel and etc tets suite kept passing? [13:24] nope [13:24] if they don't pass, they're actually really seriously broken [13:25] maybe not in a fail-all-the-time way, but in a pass-only-by-luck way [13:25] nessita, u1-client is fixed, I'm working on control panel atm [13:25] tcole: I know, but, we can't just broke everything in a minute... in a future, it wuld be a good idea [13:25] tcole: to give some heads up before landing [13:25] don't shoot the messenger [13:25] but yeah, it could have been better coordinated [13:25] well, fwiw, i don't think tarmac has the new version installed [13:25] tcole: well, that branch was yours, right? :-) [13:26] I meant don't blame the branch for breaking anything :) [13:26] it just made existing breakage manifest [13:27] tcole: I understand how your branch is "correct", but we can't break all the passing suites all the sudden. I agree on fixing things, but we need some heads up [13:27] tcole: so you could propose the branch and let me and the ubunet-discuss list know about it [13:28] well, the one thing I didn't expect was that it would go into use right away [13:28] so each one can commit to have their suite passing with that branch within some time frame [13:28] I thought we were either using versioned sourceeps, or packages [13:28] tcole: nightlies are being built on every branch land, and we're using that in the desktop [13:28] tcole: btw, your qname-in-errors branch has 2 approve votes [13:29] ah, that I did not know [13:30] I thought we could do it like in the past when we update the branch and then increment sourcedeps for each dependent project as they were updated to pass [13:30] tcole: nah, that's just server :-) [13:30] nightlies will also get installed in tarmac if someone runs apt-get upgrade in the vm, or if it breaks and a new puppet deployment happens [13:30] yeah, I've been away from client for a long time [13:30] mm [13:31] tcole: it would be trivial to add a command line option to enable the check [13:31] tcole: is ok, I was just trying to plan for a future situation ;-) [13:32] dobey: if we end up blocked for more than a few hours, let's consider it -- but given how frequently people make these mistakes, it needs to be the default asap [13:32] (more than a few hours, counting from an hour or two ago) [13:32] tcole: sure. i'm just making the suggestion. personally, i think if correct things cause broken things to manifest their breakage, then fix the broken stuff, not the correct one :) [13:32] Well, I can do a branch in the meantime [13:33] but yeah [13:33] I'm happy, at least, that we've not had trouble with false positives so far [13:33] although "correct" things that break your own API are just f'n broken (grrr, gtk) [13:35] heck, if u1client was that easy to fix, it shouldn't be a problem :) [13:35] what I would have liked, actually, would have been a good option to make these things warnings at first, rather than hard errors [13:36] but it wasn't really immediately obvious that there was a nice way to do that with unittest [13:36] yeah, most of the mistakes are trivial [13:36] yeah i'm not sure how to do that with unittest. [13:36] just easy to make [13:37] I hate that twisted makes the obvious things wrong, more than asynchrony needs to [13:37] tcole: having a warning.warn in setup? [13:38] it's tricky [13:38] we want to warn about the specific set of test classes which are getting run [13:39] there doesn't seem to be a good way to accurately capture that except via the actual test loader we use [13:39] oh neat [13:39] client dailies got requested to build successfully it seems [13:41] it would be nice if unittest had a WARNING type of result for a test, rather than just pass/fail/error/skip [13:41] yeah, that's exactly the problem I ran up against [13:41] i guess we could perhaps write one [13:41] but meh [13:42] I want it to be orange! [13:43] it will be burnt orange just like the 70s were [13:43] mandel, ¬¬ === m_conley_away is now known as m_conley [13:43] :) [13:43] jeje [13:44] speaking of the 70s [13:44] cheap trick just came up on my playlist [13:47] hello all! [13:48] hola alecu [13:48] hi alecu [13:48] alecu, hi! [13:50] I guess we have no standup, right? [13:50] because we are 20 min late :P [13:53] mandel, i guess not [13:55] mandel, gatox, alecu: you can do it now! I love to read it [13:56] me [13:56] me [13:58] mandel, alecu ping [13:59] nessita, should i start? :P [13:59] yes! [13:59] gatox: go [13:59] DONE: [13:59] Woke up and found out that mandel was going crazy with the broken tests. [13:59] TODO: [13:59] Keep fixing the broken test in sso and the installer, fix some branches in order to land them, keep working in the unicode huge issue! [13:59] BLOCKED: [13:59] yay [13:59] No [13:59] nessita, go :P [14:00] DONE: sprinting, talking a lot [14:00] TODO: more of the same! finish branch for I/O failure [14:00] BLOCKED: no [14:00] NEXT: dobey? alecu? mandel? [14:00] goin, writing notes, one sec [14:00] * alecu is writing notes [14:00] λ DONE: survey, fixed SRU review issues [14:00] λ TODO: tarmac puppet conf updates [14:00] λ BLCK: No. [14:01] goin, writing notes, one sec [14:01] ? [14:01] DONE: Found bug 882440 and fixed it. Also found 882446 and fixed it. Rereview nessitas branch. [14:01] TODO: Look at control panel issues relqted with the above. Propose auto-update branch. [14:01] BLOCKED: no [14:01] Launchpad bug 882440 in ubuntuone-dev-tools "The testing package is not installed by setup.py (affects: 1) (heat: 6)" [High,Fix committed] https://launchpad.net/bugs/882440 [14:03] nessita, you have a +1 [14:03] DONE: team mumble, got pyqt4 in linux crossing an ntlm isa server! both http and CONNECT versions, devised storage protocol "ping"-like check [14:03] TODO: work on tunneling using qt4 [14:03] BLOCKED: no [14:04] mandel: yey! [14:04] alecu: any news on the review? [14:05] nessita, I didn't get much further yesterday, sorry. [14:06] alecu: could you please allocate some time today? is so big that almost any new branch on trunk will cause conflicts :-/ [14:06] nessita, I'll work on it today, I'll promise :-) [14:06] alecu: thanks!!! [14:06] Todo: relearn how to read sheet music [14:06] dobey: glad it was with two e's! ;-) [14:13] ok, I'll brb [14:13] (later) [14:16] that is a bbl, right? [14:32] heh [14:34] bbl myself [14:35] alecu, ralsina is going to have 1-1 today with us?? or he is not working et al? i don't remember that [14:36] gatox, his is travelling, right? [14:36] mandel, i tought that he take the day off today and travel tomorrow..... but maybe i'm wrong [14:43] gatox, I believe he's not coming, so no 1-1 today [14:43] alecu, ok, thanks [14:46] * mandel is half way though the tests in control panel \o/ [14:46] mandel, i'm almost finishing sso [14:46] :S [14:48] gatox, I had 600 fails :( [14:48] yacks [14:48] mandel: still trying to fix the controlpanel suite? [14:48] gatox, most borring work EVER [14:48] mandel, totally [14:50] nessita, I've got only 100 tests left of 600 :D === yofel_ is now known as yofel [14:51] alecu, mandel: can we move the talk one hour earlier? [14:51] that is, in 10 minutes? [14:51] I got released from an activity so I would like to catch up with you re:proxy [14:52] wait what? [14:52] dobey: what where who? [14:53] nessita: there was a meeting re: proxy scheduled for 1 hour from now? [14:53] dobey: only a catch up from me with alecu and mandel [14:53] since they ae starting working on that next week [14:53] nessita, sure, I have no problem [14:53] dobey: do you want to be there? [14:54] we'll have to skype instead of mumbling [14:54] nessita: in the meeting yesterday it was suggested to have another meeting about proxy stuff; and i told alecu i should be in it [14:54] nessita, sure, do I have you? I remember you have a work an non work accounts, right? [14:55] dobey: hum... that was not in the summary email from roberto [14:55] dobey: you can join us in 5 minutes in skype, if you want [14:55] (if alecu can) [14:56] how long will it be? :) [14:57] dobey: hem.... one hour top I would say (45' ideally) [14:57] ugh [14:57] i will need to go get lunch soon [14:58] dobey: what time is for you? I thought that now we both were in the same timezone [14:58] is it 11am for you now? [14:59] nessita: we are in same timezone, but i'm not on UDS schedule until next week. i don't normally eat breakfast, and eat lunch a little earlier [14:59] dobey: ah, I see. At what time would you need to be free? [15:01] well you can have it, and just take notes, or record it [15:01] yeyyyy from 600 tests failing to 10.... fixing those 10 [15:01] dobey, I think is just catch up, there should not me much more info that yesterday [15:02] i just wanted to arg^H^Hdiscuss the technical details [15:02] nessita, shall we wait for alecu then :) [15:02] dobey, hehehe [15:02] dobey: you against something in particular? [15:02] mandel: yeah... [15:03] well it seems like the suggestion is to just use qt to solve the problem everywhere. and that's not really feasible to me [15:05] dobey, its an understandable complain, I think it has to be a valance between time to market and being nice with the platform [15:05] dobey, the idea of using a diff process for sd does not sound good to my ears on linux for example [15:06] dobey: qtnetwork, which is a network library independent from the UI library. So is not strictly *qt*, but I understand your point. I kinda agree with you, but if that really solves all the proxy issues in a clean way, and we have no better choice, we may need to give it some consideration. [15:07] dobey: do you have any other option that solves all the proxy issues? (btw, did you read alecu's report on proxies?) [15:08] mandel: b in english is b, not v :) [15:09] nessita: my point is that it won't solve all the proxy issues in a clean way :) [15:09] yes i read the doc [15:10] dobey, sorry, I make the same mistake in all langs, in Spain we pronounce them the same, and we write movil, so I get confused :P [15:10] mandel: :P [15:11] dobey: I see. Besides the SD proxy issue not being solved by qtnetwor, is there any other issue? [15:12] nessita: well, a separate qt process for the protocol, doesn't solve the problem for things using the REST API in SD, via urllib or whatever. [15:12] if I change my host and hostname will this affect my Ubuntu one? [15:12] nessita, one question about control panel, why do we use defer from python and not the twisted one? [15:12] i think the first thing that should be done, is to reduce the number of HTTP libraries in use, to the base minimum [15:12] dobey: right, the fix is using qtnetwork to do the url opening [15:12] nessita, that would be in test_pacakage_manager.py [15:13] nessita: to do that we would need to use the qt main loop in sd, no? [15:14] spartan2276: likely, yes. you will probably need to log in again, as the machine name is used to distinguish tokens [15:14] well thanks for leaving right as i was typing an answer [15:14] people are so impatient [15:15] dobey: hum, I will have that in mind when talking to the guys [15:15] alecu: you around? [15:15] dobey, they are used to a fast internet connection :) [15:15] mandel: yeah,we don't want the controlpanel depending on twisted [15:16] mandel: what problem is causing you? it should not cause any issue [15:16] nessita, dobey that is why qtnetwork is not to be used in sd but to use a diff process to tunnel it, which is ugly [15:16] nessita, just curious [15:16] mandel: that doesn't work for the urllib calls [15:16] mandel: but what about the rest api calls? [15:16] and also, making a proxy to proxy the calls through a proxy, seems a bit… off :) [15:17] dobey, I agree.. I think the problem is the money and the time.. [15:17] nessita, same problem :( [15:18] mandel: ok, let's keep talking about this with alecu [15:19] nessita, dobey yes, better with alecu since he has been looking into it [15:19] time shouldn't be an issue [15:20] if you make time to implement an issue, we're going to end up in the same boat with trying to develop based on urgency [15:20] and the whole point of doing what we're doing with stable branches/etc, is to avoid that [15:21] so we shouldn't make decisions based on that. we should make decisions based on what's correct [15:21] dobey: yeah, I agree we can't (keep) making decisions based on time hurriness [15:22] we need to evaluate the options, decide what is most correct, rather than most quick, and make time estimates for what the most correct option will take, and develop that. [15:23] dobey: I agree. Not sure which is the more correct though... we definitely don't want to add proxy support to twisted ourselves [15:24] well we probably have to do something like we did with the bandwidth limiting stuff [15:25] surely there is some software out in the wild which uses twisted and works with proxies [15:25] anywya, i need to go get lunch :) [15:25] dobey: well, alecu wa sin charge of researching this... and I trust him he did a good research [15:25] bbiab :) [15:26] tcole, nessita can you please takea look at lp:~mandel/ubuntuone-control-panel/fix-tests all tests pass except the ones that use pacakage_manager.inline_callbacks [15:26] yeah i'm not saying i don't trust him, or he didn't do good research. but i do think he didn't necessarily have all the information, or that all the information he did have, was necessarily correct. [15:26] tcole, nessita I'm temted to use the inlineCallbacks from twisted, but I wanted to know if there is a better way [15:27] mandel: have a link? [15:27] mandel: it should not be the way, no, I'll take a look [15:27] anyway, lunch time. we can argue more later :) [15:27] dobey: ack [15:28] nessita, I have not done an MP, let me get you the branch link [15:34] tcole: ping [15:34] tcole: what does the new check that you landed do? [15:44] gatox, ping [15:44] mandel, pong [15:45] gatox, do you have the bug number about the broken tests in control panel? [15:45] gatox, I'm done with it :D [15:45] mandel, oh sorry..... i didn't submit that one.... i only create the bugs for sso and the installer :P [15:45] my bad [15:47] gatox, puto ;) [15:47] gatox, hehe I'll do one then [15:49] nessita, mandel, dobey: I'm back. [15:49] mandel, pong [15:49] alecu, shall we do the mumble (although I'm running out of time) [15:49] mandel, I thought it was scheduled in 10'! [15:50] mandel: yes, it was scheduled in 10 minutes [15:50] mandel, oh, it's past 18 for you [15:50] alecu: before I took a shoot and saw if we could do it earlier [15:50] alecu, mandel: skype? [15:50] nessita, alecu I can stay a little longer, np :) [15:50] nessita, already there [15:50] is dobey coming too? [15:51] oh, skype. [15:51] alecu: when he gets his lunch done :-D [15:51] alecu: my mumble is not working... [15:51] * alecu has not used skype in a while [15:51] ok [15:51] I will install O today or tomorrow, so hopefully it will work again [15:52] gatox, tcole please give me a review: https://code.launchpad.net/~mandel/ubuntuone-control-panel/fix-tests/+merge/80587 :) [15:53] mandel, on it === beuno is now known as beuno-lunch [16:12] dobey: you back from lunch? [16:25] mandel, +1 to your branch [16:25] i'm uploading mine from sso right now [16:28] mandel, alecu please review (fixed tests in sso): https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/tests-broken/+merge/80596 [16:28] nessita, ^ (or you if you are still around) [16:31] gatox: ack [16:32] gatox: [16:32] ubuntu_sso/qt/tests/test_controllers.py: [16:32] 1680: [C0301] Line too long (80/79) [16:33] crap [16:33] fixing [16:33] gatox: there is no need to add this: [16:33] @defer.inlineCallbacks [16:33] def tearDown(self): [16:33] """Tear down.""" [16:33] yield super(PasswordLabeledEntryTestCase, self).tearDown() [16:33] nessita, sure?...... that's way the test are broken.... [16:33] gatox: no way :-) [16:34] mandel, hey! answer! you told me that [16:34] gatox: remove it, and rmove all the addings for setUp and tearDown [16:34] gatox: you don't have to add a tearDown if it's not deifned [16:34] same foe setUp [16:34] gatox: *if* setUp or tearDown is defined, you have to use inlinecallbacks + yield [16:35] mandel, .......... i'm going to kill you...... [16:35] nessita, ok...... i'll fix that after lunch [16:38] sort of [16:38] nessita: back, but still eating. :) [16:38] dobey: can you join us in skype? I'd like you to be there [16:39] for how long? [16:39] I have lunch at 1pm so... 20 min top [16:39] ok, i'm logged into skype [16:40] dobey: I don't have you! [16:40] you wre dohbee? [16:40] yes [16:48] gatox, why? [16:49] * mandel rugby [16:49] mandel, ok.... i'll tell you later :P [16:49] gatox, I told you if, you workaholic :) [16:49] haha === eu is now known as Guest26044 [17:10] oi === beuno-lunch is now known as beuno [17:17] hi guys, i have trouble with ubuntuone [17:18] ubuntu one client says there is all files sync, bat when i open one folder is empty, on web i can see the files, bat it not sync to computer. is there any function how resync the folder? or download files in zip from web? [17:22] Pooky1: have you disconnected, and reconnected? [17:22] i try it [17:31] dobey: damit, still nothing [17:33] Pooky1: can you open a terminal, and restart syncdaemon by running "u1sdtool -q" waiting a few seconds, and then running "u1sdtool -c" and waiting a few (maybe several) minutes, to see if anything changes [17:34] Pooky1: we /are/ experiencing some slowness issues at the moment, which are being worked on. let me get someone who knows a little more about this to help you :) [17:35] dobey: i already did this, bat there is problem there is only one folder, which is not sync, there is 4 subfolders and 4 files, bat nothing of it is sync, its just empty [17:36] hello Pooky1 [17:36] hello duanedesign [17:36] Pooky1: i understand. duanedesign will try to help you further :) [17:36] dobey thanks [17:37] Pooky1: is thi folder that is not syncing in your Ubuntu One folder? [17:38] duanedesign: yes it is ~/Ubuntu One/www/Problematic Folder/... [17:38] Pooky1: could you open a terminal and run the command - u1sdtool --waiting | wc -l [17:39] duanedesign: this command return "0" [17:39] hmm, ok [17:39] * rye will be back in 40 minutes, past eod thoug [17:40] duanedesign, u1filestatus.py? [17:40] http://people.canonical.com/~roman.yepishev/us/u1filestatus.py [17:41] Pooky1: can you check this file to see it contains anything. ~/.cache/ubuntuone/log/syncdaemon-exceptions.log [17:42] duanedesign: this file is empty [17:44] Pooky1: could you pastebin(http://paste.ubuntu.com/) your ~/.cache/ubuntuone/log/syncdaemon.log so I can look at it for problems. [17:44] Pooky1: this file has filenames in it. If you do not want to pastebin it you can email it to me [17:46] duanedesign: there is http://paste.ubuntu.com/720930/ [17:47] * duanedesign looking [17:55] do you see somethink which can couse this problem? duanedesign [17:56] Pooky1: could you try... [17:57] runnning the command: wget http://people.canonical.com/~roman.yepishev/us/u1filestatus.py [17:58] then: python u1filestatus.py [18:00] there is result : Ubuntu One folders: [18:00]