=== rmcbride_ is now known as rmcbride [02:31] ehlo [02:37] schneikies [02:37] you could do ad-hoc Group Policy stuff with UbuntuOne [02:38] fyi, UbuntuOne is on Ars http://arstechnica.com/open-source/guides/2009/12/code-tutorial-make-your-application-sync-with-ubuntu-one.ars/1 === wgrant_ is now known as wgrant === GreySim_ is now known as GreySim [13:03] urbanape, ping [13:09] aquarius: hey, got your review notes. [13:48] thisfred: You should re think writing spam blogs for a living, I don't think it's sustainable. [13:49] jblount: but apparently I'm very good at it ;) [13:51] blogger is weird. I have almost no links to external sites, none to dubious sites, so I don't know what their algorhithm looks at. Also, they don't contact you to tell you your blog has been taken down. They block access to the edit interface as well as the public blog. Then when they finally deign to review it, they send me a message that tells me my blog was deleted. That got me awake better than coffee. [13:52] I'm totally going to posterous: if they take me down, I'll still have all the posts in my sent box. [13:52] jblount: well... spam blogs use a renewable source (gullible people); as long as the hosting is green, it should be sustainable [13:56] Chipaca: Good point. === joshuahoover1 changed the topic of #ubuntuone to: Have a question? Post here to joshuahoover | https://one.ubuntu.com | https://launchpad.net/ubuntuone | Please honk if you want a music store [14:16] aquarius: CardinalXiminez_: dobey: jblount: urbanape: teknico: vds: dunno if you noticed, but the 'u1 bug day' in the calendar was off by a couple of hours. bug day is _now_ [14:16] Chipaca, ah. I had not noticed! [14:16] Eek. [14:19] Chipaca: not noticed... [14:20] Chipaca: we use this channel to discuss the bugs? [14:20] Chipaca, joshuahoover1, we need the bug list, don't we? [14:20] https://wiki.canonical.com/UbuntuOne/BugDays [14:21] teknico: please see the other channel's topic :) [14:21] aquarius, joshuahoover1, thanks === Chex_ is now known as ChEx [14:22] teknico, can I trade you a bug? https://bugs.edge.launchpad.net/ubuntu/+source/ubuntuone-client/+bug/488815 is in Italian and I can't read it ;) So choose one random bug off your list and I'll do it instead. === ChEx is now known as Chex [14:22] Ubuntu bug 488815 in ubuntuone-client "Ubuntu One non si connette" [Undecided,Confirmed] [14:23] aquarius, sure [14:23] aquarius, https://bugs.launchpad.net/ubuntu/+source/ubuntuone-client/+bug/480343 [14:23] Ubuntu bug 480343 in ubuntuone-client "Ubuntu One client fails to connect to service" [Undecided,New] [14:29] hmm, why can't I mark bugs as Triaged? [14:30] urbanape: is it a package bug by any chance? [14:31] Yeah, I guess it is [14:32] urbanape: ok, we need to get you access to triage package bugs...let me see if i can get someone to do that for us [14:36] urbanape: i'm sending pedro a list of names that need access to triage package bugs and he'll set that up for us [14:37] hurrah for pedro [14:37] joshuahoover1, thanks for organizing this [14:38] statik: np :) [14:39] meh [14:40] no ennui today, today is a good day to die [14:40] bah, why can't I set a bug as wishlist importance on evolution-couchdb? [14:40] this is why i don't plan things in advance. they always change :) [14:41] aquarius: on the package or on the project? [14:41] package, I think [14:41] "Affects: evolution-couchdb (Ubuntu)" [14:41] that's the package, right? [14:42] dobey, oh, you plan them after the fact? ;-) === mbarnett` is now known as mbarnett [14:44] aquarius: yes. you probably aren't in ~ubuntu-bugs or whatever the team is called [14:44] teknico: no, i just try to do stuff as it needs done. :) [14:44] someone want to trade a bug in Spanish with me? [14:45] and Chinese? [14:45] also, I've been given this bug #495453, but don't have access to it. [14:45] Bug 495453 on http://launchpad.net/bugs/495453 is private [14:45] joshuahoover1, do I need to be in an ~ubuntu-bugs team to mark bugs in evolution-couchdb (Ubuntu) as wishlist? How do I get into that team? [14:46] Also, German and Svenska. I'm feeling all multicultural and stuff. [14:46] urbanape, off to babelfish for you then :P [14:47] whee. Can I use babelfish for my responses as well? [14:47] "MY HOVERCRAFT IS FULL OF EELS" [14:47] that would be Hungarian [14:47] and? [14:48] Like no one has ever translated English to Hungarian to Bantu to Urdu to Chinese before. [14:48] no, but Yiddish might help [14:49] reminds me of the site that made the rounds a short while back, trying to find a stable translation. [14:51] urbanape: #495453 is public now. there wasn't any vital info in it [14:51] I'm guessing that lots of us are getting lots of semi-related bugs. For the client stuff, are there any tips or things we should be looking for in the accompanying logs that would point us towards a duplicate? [14:52] especially the capabilities mismatch? The two I've seen are either fresh installs or recently upgraded with nothing showing as a candidate for upgrade. [14:54] not sure. i suspect most client bugs are going to be dupes [14:54] nothing showing as a candidate for upgrade is likely a problem where they had the jaunty PPA, then upgraded to karmic, and the last version in jaunty is versioned poorly so it shows up as newer than the version in the official karmic archive. re-enabling the PPA or removing/adding ubuntuone-client should fix it [14:58] upgrading to karmic comments out all your PPAs, doesn't it? [14:58] yes [14:59] but in my experience it also does a s/jaunty/karmic/ on the sources lists [14:59] I have the same question about dupes. We could do with a way of identifying the same bug showing up a lot. Not sure what the best way of doing that is (apport? a script which looks in attached logfiles?) [14:59] so you can just open the software sources prefs, and re-enable and should be good [15:00] aquarius: comparing syncdaemon-exceptions.log is probably going to be useful for that, yes [15:00] outside of that, probably not a whole lot to compare [15:00] dobey, http://launchpadlibrarian.net/35501410/.home.gerald..cache.ubuntuone.log.oauth.login.log.txt suggests that the punter doesn't have networkmanager -- is my best response to that "the PPA version doesn't require networkmanager; here are PPA install instructions"? [15:00] it would be really useful if we could extend the apport retracer on the server, to compare log files [15:01] aquarius: yes you do, i'm taking care of that now for everyone (well, putting together the list so pedro can do it for us :) ) [15:01] aquarius: dup of 357395 [15:02] dobey, ooh good lad yourself. :) [15:03] aquarius: and the beginning of the description for 357395 has a pointer on installing the beta ppa [15:03] it does indeed [15:03] are there known 64-bit problems? [15:03] dobey, yeah; if we could autodupe "no networkmanager" and "noaccesstoken" bugs it'd be way helpful. [15:04] aquarius: well, as soon as the world gets the SRU installed, new versions of those will be gone anyway [15:04] but yes [15:04] noaccesstoken is fixed in the sru? sweet :) [15:05] do we do the standup thing? [15:05] not while we're on a bug day, I don't think [15:05] oh, ok [15:05] I'll run off to the bank then :) [15:07] dobey, what's the canonical NoAccessToken bug? #453825 is marked as a dupe of #451670 but #451670 seems to be the httplib 2.6.3 (HTTPSConnection._tunnel_host) bug [15:09] https://code.edge.launchpad.net/bugs/488413 [15:09] Ubuntu bug 488413 in ubuntuone-client "SD crashes when no token found" [High,In progress] [15:09] excellent., cheers [15:12] * aquarius is amazed at the number of people who don't have networkmanager [15:13] * dobey is amazed that wicd/connman don't provide the same dbus interface [15:13] grmbl grmbl [15:13] yeah [15:13] ok, that's all my easy duplicate-of-a-common-problem bugs done :) [15:14] i'm not really sure how to analyze the log file that was attached to this bug report about files not syncing: https://bugs.edge.launchpad.net/ubuntu/+source/ubuntuone-client/+bug/448093 any syncdaemon experts want to teach me what to look for? [15:14] Ubuntu bug 448093 in ubuntuone-client "ubuntuone is not updating all files" [Undecided,Incomplete] [15:20] Yeah, bugs are hard. Especially when they all seem very similar, but not necessarily duplicate. [15:21] urbanape: ++ [15:21] joshuahoover1, for a bug which is marked incomplete, and is still awaiting a response from someone, do I need to do anything? or just move on to the next bug in the list? i'm wondering how to track that a human looked at it today and it was still in the incomplete state [15:21] yeah, and then you have to fix them [15:21] what kind of silliness is that! :) [15:22] statik: those are ones that we depend on lp's expire functionality to take care of...but maybe i should incorporate something into the script that generates the list to check for the last person to comment...if it's in the team, then we shouldn't put it in the list [15:23] ooh that would be cool [15:23] For bugs like this: #494191 where "it works, after I restart", what's the proper thing to do, short of, "Yeah, we'll work on making it more robust and reliable in the future." [15:23] i don't think lp's expiration does what people think it does [15:23] if there's no there there. [15:23] dobey: no? what does it do? [15:23] i've seen several bugs in my list that when opened just say "this bug expired N days ago" [15:23] which doesn't really help with anything :) [15:23] yeah i saw that this morning too [15:24] joshuahoover1, should i be doing any edits on this wiki page to show that i processed my bug list, or will it just be overwritten next week? https://wiki.canonical.com/UbuntuOne/BugDays#Elliot Murphy [15:24] you could check last comment vs. when it was marked incomplete, or such, with a script, though [15:24] statik: i was planning on just overwriting...i'm up for suggestions :) [15:38] Would a no access token cause the machine to shut down after 10 minutes? [15:38] webm0nk3y, rawk. I thought so. Nice [15:42] dobey, if ssl.c is throwing EOF in violation of protocol (bug #489277), is that the HTTPSConnection bug, or something weirder? [15:42] Launchpad bug 489277 in ubuntuone-client "Ubuntu One produces a connection error when launching." [Undecided,New] https://launchpad.net/bugs/489277 === solarion_ is now known as Solarion [15:44] aquarius: sounds weirder [15:44] k. I shall assign it in your direction then :) [15:46] jblount is about to get interviewed live on http://www.ustream.tv/channel/statik-talk [15:56] heh. bonus points for launchpad finding lucio even though I spelled his name wrong ;) [15:57] aquarius: i often assign the user "lucio" vs. the user "lucio.torre" to bugs...lp reminds me that "lucio" doesn't normally work on bugs...pretty handy too! [15:58] dobey, http://launchpadlibrarian.net/36199469/.home.omri..cache.ubuntuone.log.oauth.login.log.txt looks to me like the dude's clock is wrong? [15:59] hmm, valid apport stuff, but the title of the bug (translated) is: "Your QQ version does not support this super-group, may not be able to obtain a complete list of group members. Please download the latest QQ2008 to http://im.qq.com experience." Spam? [15:59] bug #495402 [15:59] Launchpad bug 495402 in ubuntuone-client "您的QQ版本不支持本超级群，可能无法获取完整群成员名单。请到http://im.qq.com下载最新QQ2008进行体验。" [Undecided,New] https://launchpad.net/bugs/495402 [15:59] aquarius: it would have had to become way wrong during the oauth token getting phase. and that doesn't seem generally plausible, unless it was during an automatic tz change or something, which is generally hard to try to do [16:00] aquarius: probably started the applet and went to watch an episode of seinfeld or something, and then came back and clicked "OK, add this pc" [16:00] dobey, heh. I'll ask them to try again then :) [16:01] aquarius: granted the error handling of that case could be better [16:07] do we have a FAQ for using U1 with firewalls enabled? [16:07] any particular port settings or services that need to be explicitly enabled/activated? [16:08] aquarius: http://twitter.com/djco/statuses/6699873284 [16:09] djc: I asked moz developers about that when I was starting on Bindwood. [16:10] No plans yet for making the storage pluggable, which is about all it would take. [16:10] But that's okay, since now we can target other browsers as well (hello, Chromium) [16:15] urbanape: if you can connect to https://one.ubuntu.com/ in your browser, you should be good [16:15] urbanape: we don't need any special ports really [16:15] dobey, do local firewalls block local firefox access to localhost:234782 for the callback? [16:16] aquarius: only if you're insane and you configured iptables on the lo interface or something [16:16] dobey, yeah, good, just checking :P [16:16] aquarius: though i think there are firefox extensions that break connection to localhost === CardinalXiminez_ is now known as CardinalFang [16:17] dobey, yeah, I got that impression when the callback failed for some people [16:21] anyway, time to get some lunch [16:21] bbiab :) [16:24] this is a whole lot of dbus not responding: http://launchpadlibrarian.net/36705554/.home.jostein..cache.ubuntuone.log.oauth.login.log.txt thoughts? [16:51] right, off to get my daughter. Back later, all. [18:12] poopy day for freenode, what? [18:14] urbanape_: apparently fnode is getting DoSed the F out [20:17] djc, ping? [20:22] aquarius: pong [20:24] djc, is it possible to apply ?group= or ?descending with couchdb-python? [20:25] with views [20:25] ? [20:25] also, why ask here rather than #couchdb? :) [20:25] djc, 'cos I knew you were here and I know you better than cmlenz ;) [20:26] yep, with views [20:27] but, yeah, you can: db.view('blah/blah', descending=True) [20:28] although it often takes the form of opts = {'startkey': 'a', 'endkey': 'b'}; db.view('foo/bar', **opts) [20:29] aha, sweet, you can pass arbitrary extra options to the view. quality. [20:29] I think we might have inadvertently stomped on that in desktopcouch.records. Must fix that ;) [20:29] why yes, couchdb-python is pure quality [20:29] Question. I heard on the FLOSS Weekly interview that any desktop developer can hook into One by hooking into CouchDB and getting all the neat replication stuff, but is it at all or will it ever be possible to develop for the website side? Like, say I wanted to get Getting Things GNOME (todo list program) hooked into One. Would I be limited to replication, or would it be possible to have it set up to edit tasks through the web interface? [20:30] GreySim, we've been thinking about that, quite a lot. It's not currently possible, but I want it to be. [20:30] aquarius: so, why didn't Ubuntu One start by integrating with Weave in the first place? [20:30] Good to hear. [20:30] Thanks. [20:30] djc, weave doesn't have a swappable backend [20:30] one could rewrite to deal with a CouchDB backend? [20:30] s/rewrite/enhance/ [20:31] weave is also pretty specific to firefox as i understand [20:31] djc, yeah, but you were right the first time -- it'd be a relatively substantial rewrite, rather than an enhancement [20:31] djc, which we didn't have time for [20:31] meh [20:32] just being a backend to weave would make it that much harder to have other browsers work with it [20:32] dobey: you don't have to be "just" a backend for weave [20:32] I think the CouchDB interface is generic enough to serve all kinds of stuff [20:32] and being schema-less should make it fairly easy [20:33] and you also get a bunch of other stuff for free; e.g. Weave doesn't do just bookmarks, it does bunch of other stuff I want synced as well (e.g. Places, prefs) [20:33] yeah; we could have built bindwood as a weave backend, indeed. If weave had a concept of swappable backends then that's almost certainly what we would have done [20:34] what the heck is "Places"? [20:34] new name for bookmarks? [20:34] the Firefox subsystem that stores bookmarks and history [20:34] with the frecency counts that make the awesomebar do its thing [20:35] and we want to sync prefs for things other than firefox as well [20:36] but it's a very hard problem [20:37] dobey: that's not a problem, it would be nigh-trivial to put different kinds of data in Couch (e.g. some things for Weave, some things not) [20:37] djc: you misinterpreted my statement to imply that the problem was in fact couch [20:37] it is not :) [20:37] the problem is applications [20:38] what is the problem? [20:38] most applications store state as configuration [20:38] and state is a horrible thing to sync across machines [20:39] okay, I just think that has nothing to do with the fact that the weave backend should be CouchDB-like [20:39] i was also not disagreeing on that [20:39] ok, good [21:03] dobey: ping [21:03] verterok: hi [21:03] dobey: hi [21:04] dobey: I'm working in a script to generate a simple text file with the syncdaemon dbus API [21:04] dobey: already wired it to the Makefile.am, etc [21:04] dobey: wondering if you have any ideas/recomendations about this :) [21:05] no ideas [21:05] dobey: e.g: my idea is to add the generated file to .bzrignore [21:05] my python isn't that good [21:05] dobey: oh, sorry.. [21:05] and we're not describing the API in XML like we would be doing if it was in C [21:05] dobey: I'm thinking in the packaging side [21:06] dobey: the text file is generated from the xml [21:06] dobey: the xml is generated in runtime [21:06] ok [21:06] but we aren't writing XML ourselves i mean [21:06] dobey: no, right. no handmade xml [21:07] i'm sure you can do funky stuff with python to poke about internally, since it's dynamic and all :) [21:07] i'm not sure what you're trying to ask of me though :) [21:07] dobey: actually I'm getting the api description via dbus ;) [21:07] dobey: so, I added a 'doc' target to the Makefile [21:07] ok [21:07] dobey: but is just sitting there, isn't wired with anything else [21:08] dobey: do I need to do something else to get the generated file included in the package? [21:09] verterok: probably easier for me to look at what you have now, and then suggest what to do :) [21:10] dobey: :) [21:12] dobey: what I have is: a new make target: 'doc', that generates a file: docs/syncdaemon_dbus_api.txt [21:12] yes i understood as much [21:12] is it only generating one file? [21:13] dobey: yes [21:13] dobey: at least ATM [21:13] verterok: we'll probably want a similar doc for oauthdesktop API [21:13] dobey: with a bit more of work, we cna generate another file for oauthdesktop dbus api [21:13] heh [21:13] heh, right :) [21:14] dobey: I need this in order to discuss a new dbus api (for UDFs) with desktop+ [21:14] dobey: but we can file a bug for the oauthdesktop bits ;) [21:15] ok, great! [21:16] I'll propose the branch as it's now, and we can add more later [21:16] dobey: thanks! [21:16] verterok: hrmm. my brain is a bit astray at the moment, but if you go ahead and propose the branch, i can look at it later and make any fixes, or comment in review to make them [21:16] dobey: cool, thanks! [22:08] Does Ubuntu One sync files that are symlinked to ~/Ubuntu One? [22:09] jamalta: we don't support symlinks, no [22:11] dobey: hm.. alright [22:11] dobey: I'll just copy the files for now then, thanks! [22:11] jamalta, in Ubuntu 10.04 you'll be able to designate any folder as synced to Ubuntu One, not just your Ubuntu One folder [22:12] Wow, the client picked that up quick... [22:12] aquarius: That's sweet!