mandel | morning all! | 07:06 |
---|---|---|
=== zyga is now known as zyga-afk | ||
JamesTait | Good morning all! :) | 08:31 |
=== zyga-afk is now known as zyga | ||
gatox | good morning | 11:11 |
gatox | restart.....brb | 11:18 |
gatox | back | 11:23 |
mandel | gatox, buenas :) | 11:35 |
gatox | mandel, buenas | 11:35 |
mandel | gatox, I have started with the integration of u1 with the fsevents I might have some progress (nearly working) tom :) | 11:38 |
gatox | mandel, i added some modifications to macfsevents..... and now i'm fixing the tests that were already broken, and adding a few more..... and i hope to finish later fixing test_darwin for filesystem_notifications..... and that should be ready....... | 11:38 |
gatox | i want to finish with this NOW! jeje | 11:39 |
gatox | mandel, in other news! i started a Unity Lens yesterday which allows you to publish files from u1...... directly from the dash :D..... it's almost done | 11:40 |
mandel | gatox, hehe well I think that if we get to the point where we can execute the daemon on mac by wend and it kind of works we are ok :) | 11:40 |
mandel | gatox, vala? | 11:40 |
gatox | mandel, nop...... i'm doing it on python.... then i'm going to do it on vala..... in python i can have it working in a couple of minutes as a proof of concept | 11:41 |
gatox | because i can reuse some code that i already have | 11:41 |
mandel | oh, ok, sounds cool | 11:42 |
ralsina | good morning! | 12:20 |
mandel | ralsina, morning! | 12:20 |
gatox | ralsina, hi | 12:21 |
ralsina | Hola mandel, gatox | 12:21 |
mandel | ha! got integration tests running between python and objective-c! | 12:28 |
gatox | mandel, yeyyyyyy | 12:30 |
gatox | i have macfsevents tests working!!! | 12:30 |
gatox | a successful day for tests! | 12:30 |
gatox | jejeej | 12:30 |
mandel | hahaha | 12:30 |
ralsina | so, is the port done already? ;-) | 12:32 |
mandel | ralsina, I wish! :) | 12:32 |
ralsina | mandel: what's stopping you from finishing? ;-) | 12:33 |
gatox | ralsina, emotional pain | 12:33 |
gatox | jejeje | 12:33 |
mandel | the universe! | 12:33 |
mandel | :P | 12:34 |
mandel | ok, lunch | 12:41 |
alecu | hello, all! | 13:24 |
gatox | alecu, hi! | 13:24 |
ralsina | thisfred: hats off to your pedantry sir. Well played. | 13:24 |
thisfred | hehe | 13:27 |
thisfred | I had to look up the greek plural, I admit. I had some Greek in school, but I sucked at it and dropped it at the earliest opportunity | 13:27 |
=== m_conley_away is now known as m_conley | ||
mandel | ralsina, we have or 1-1 today, what time is good for you? | 13:49 |
ralsina | mandel: I am in somewhat of a problem, let's do it in ~1 hour? | 13:49 |
mandel | ralsina, sure, after the standup sounds good? | 13:50 |
ralsina | mandel: awesome | 13:50 |
mandel | ralsina, superb :) | 13:50 |
gatox | alecu, i have submit the macfsevents branch with the tests fixed, let me know if you want to review it.... i'm moving to fix test_darwin for filesystem notifications now | 14:18 |
alecu | gatox, looking. | 14:36 |
alecu | gatox, is this the branch? https://code.launchpad.net/~diegosarmentero/+junk/python-macfsevents | 14:37 |
gatox | alecu, yes | 14:37 |
ralsina | dobey, alecu, gatox, thisfred, briancurtin, mmcc, mandel: standup in 3' | 14:57 |
thisfred | yipyip | 14:57 |
thisfred | me | 15:00 |
briancurtin | me | 15:00 |
gatox | me | 15:00 |
mandel | me | 15:01 |
ralsina | me (no notes) | 15:02 |
gatox | dobey, alecu mmcc ? | 15:03 |
ralsina | alecu, dobey, mmcc pingingining | 15:03 |
dobey | meh | 15:04 |
mandel | ralsina, AFAIK mmcc is not here, or so he told me on friday | 15:04 |
ralsina | mandel: right | 15:04 |
gatox | ah right | 15:04 |
ralsina | his nick is here though ;-) | 15:04 |
ralsina | so alecu is last, thisfred go! | 15:04 |
thisfred | DONE:Bug #1009505 TODO: Bug #999574 (and sub bugs) BLOCKED: no NEXT: briancurtin | 15:04 |
mandel | ralsina, does that mean that you work at my 8 am, because ralsina was here ;) | 15:04 |
ubot5 | Launchpad bug 1009505 in U1DB "get_keys_from_index is useless for multicolumn indexes" [High,In progress] https://launchpad.net/bugs/1009505 | 15:05 |
ubot5 | Launchpad bug 999574 in U1DB "deal with database files copied between machines and backup restoration" [High,Triaged] https://launchpad.net/bugs/999574 | 15:05 |
briancurtin | DONE: I had no internet service on friday so I could only do work with the info I already knew and the tabs I had open in chrome. i reconfigured, rebuilt, etc and still nothing. need to try a few more path combinations, but testability examples still do not even work... | 15:05 |
briancurtin | TODO: give this a last shot then document it all and see what happens... | 15:05 |
briancurtin | NEXT: gatox | 15:05 |
ralsina | mandel: it was "away" ;-) | 15:05 |
gatox | DONE: | 15:05 |
gatox | Reviews on friday, Improves in macfsevents and fixed the tests, small u1-freaky friday project during the weekend to publish files from the dash. | 15:05 |
gatox | TODO: | 15:05 |
gatox | Finish fixing test_darwin for filesystem_notifications. | 15:05 |
gatox | BLOCKED: | 15:05 |
gatox | No | 15:05 |
gatox | mandel, go | 15:05 |
mandel | DONE: Fixed lp:~mandel/ubuntuone-fsevents-daemon/add-dispatcher-tests according to reviews. Added integration tests for the inclusion of the daemon code in u1-client. Bug 1010511 | 15:05 |
mandel | TODO: Bug 1011659. Finish integration tests. Reviews. 1-1 ralsina. | 15:05 |
mandel | BLOCKED: no | 15:05 |
ubot5 | Launchpad bug 1010511 in Ubuntu One FsEvents daemon "Json messages are too long" [Medium,Triaged] https://launchpad.net/bugs/1010511 | 15:05 |
mandel | ralsina, go! | 15:05 |
ubot5 | Launchpad bug 1011659 in Ubuntu One FsEvents daemon "FSE_ARG_MODE args are not correctly set to the client" [Medium,In progress] https://launchpad.net/bugs/1011659 | 15:05 |
ralsina | DONE: some reviews, boting mgmt stuff, py3 thinking, discussions | 15:05 |
ralsina | TODO: 1-1s, tech leads call, think really really hard | 15:05 |
ralsina | BLOCKED: no | 15:05 |
dobey | λ DONE: reviews, triage, broken release of protocol | 15:05 |
dobey | λ TODO: fix protocol, finish releases/uploads, stable release, tarmac tweakery | 15:05 |
dobey | λ BLCK: None. | 15:05 |
dobey | alecu: go | 15:05 |
ralsina | dobey: I need to mumble with you a tiny bit further | 15:07 |
dobey | ralsina: ok. now, or later (after lunch)? | 15:08 |
ralsina | dobey: now if at all possible | 15:08 |
ralsina | since alecu seems to be MIA, EON | 15:08 |
ralsina | EOM | 15:08 |
mandel | ralsina, che, and my 1-1 :( | 15:08 |
ralsina | mandel: in 5' | 15:08 |
dobey | ok. let me hop back on | 15:08 |
ralsina | mandel: it's an emergency, trust me, you want me to talk to dobey 1st | 15:08 |
mandel | sure | 15:08 |
thisfred | mandel: did you get any time to look at u1db jenkins? | 15:11 |
alecu | sorry guys, I got a call. | 15:12 |
mandel | thisfred, no sorry.. I need to finish the fsevents asap and I'll focus on that | 15:12 |
mandel | thisfred, I have a calendar issues, this must be done before I go on holidays and I'm the only one that has worked on it atm :( | 15:12 |
mandel | thisfred, will do my best during the night while I cry watching the news | 15:12 |
thisfred | why cry? Did spain get eliminated from the championship? Or from the EU? :P | 15:13 |
thisfred | mandel: don't kill yourself | 15:13 |
mandel | thisfred, we got a bailout for 100.000 million eurs | 15:14 |
thisfred | mandel: is there anyone else with windows jenkins knowledge on the team? | 15:14 |
mandel | thisfred, briancurtin can certainly help you | 15:14 |
briancurtin | thisfred: im about to get some coffee, but i can help with windows jenkins stuff | 15:15 |
thisfred | mandel: ok, cool. briancurtin, how strapped for time are you, on a scale of 1 to mandel? | 15:15 |
thisfred | awesome | 15:15 |
ralsina | mandel: mumble? | 15:15 |
mandel | ralsina, mumble! | 15:15 |
alecu | DONE: ssl branch for SD, reviews, mumbled with gatox re macfsevents | 15:15 |
alecu | TODO: finish sd branch, use patched twisted in jenkins builds with briancurtin | 15:15 |
alecu | BLOCKED: no | 15:15 |
dobey | only a hundred million? was it a bailout offer from facebook? | 15:16 |
briancurtin | thisfred: i should have some time to look at it today. it would be a nice break from pounding my head against the desk | 15:17 |
gatox | ok...... lunch for me! | 15:17 |
=== gatox is now known as gatox_lunch | ||
dobey | briancurtin: standing desks are better for pounding your head against, i hear | 15:18 |
thisfred | briancurtin: ok, cool. I will share some files with you that jam gave me, that should include all or most of the needed dependencies. | 15:18 |
dobey | bbiab, off to get lunch | 15:32 |
elopio | mandel: ping. I have questions about the proxy. Can you let me know when you have some time, please? | 16:15 |
mandel | elopio, shoot! | 16:15 |
elopio | mandel: do we support socks? | 16:15 |
mandel | elopio, not yet | 16:15 |
elopio | mandel: what's the port number that u1 uses? | 16:16 |
ralsina | mandel: I thought we did on windows if configured system wide? | 16:16 |
=== gatox_lunch is now known as gatox | ||
mandel | elopio, any port, you just have to configure the proxy and set it in the gnome control panel | 16:17 |
elopio | mandel: but I mean, without proxy. | 16:18 |
mandel | ralsina, in win can be the case, I might be outdated | 16:18 |
ralsina | mandel: ok, maybe we don't :-) | 16:19 |
ralsina | elopio: 443 | 16:19 |
elopio | so, that means that squid thinks that we are just talking https? | 16:19 |
mandel | elopio, yes, squid should not care, do you have problems? | 16:20 |
elopio | mandel: nop. Just trying to understand. | 16:20 |
ralsina | elopio: yes, we pretend to be https | 16:20 |
ralsina | thisfred: you are going to love that book. | 16:21 |
thisfred | ralsina: yeah, thanks for the recommendation, it looks interesting | 16:22 |
ralsina | thisfred: or hate me in the process. I need someone else to read it so I can tell him how awesome it is and have him say "yes, I know". | 16:22 |
thisfred | It may be a while, though, I have a huge pile of to read right here at home, and I'm not allowed to buy any more before we move | 16:22 |
ralsina | hahaha | 16:23 |
thisfred | unless I get the ebook... ;) | 16:23 |
ralsina | Kindle FTW | 16:23 |
thisfred | kobo actually | 16:23 |
ralsina | same thing | 16:23 |
thisfred | is almost exactly the same, but many more non-drm titles | 16:23 |
* ralsina kinda wants the kobo with the light, but .... I have all those books on kindle :-( | 16:23 | |
thisfred | yeah, that was one of the reasons for buying it: no lock in either | 16:24 |
elopio | mandel: If the proxy only allows http, u1 will not work? | 16:28 |
mandel | elopio, I don't know I've never tried.. | 16:28 |
ralsina | elopio: a proxy that forbids https would be very strange | 16:28 |
elopio | ralsina: according to the survey, it seems that some people can do http but not https. | 16:29 |
elopio | at least, they think that. | 16:29 |
ralsina | elopio: I think it's more "they think that" | 16:29 |
dobey | elopio: that would be entirely useless to pretty much any corporation that even has a proxy. it's almost certainly because the ui is confusing | 16:30 |
elopio | sounds reasonable. | 16:30 |
ralsina | alecu: mumble? | 16:31 |
alecu | ralsina, sure | 16:31 |
elopio | now, last question. | 16:31 |
dobey | i really wish launchpad would stop using users' e-mail addresses for From: in e-mails | 16:31 |
elopio | how do I know what servers, authentication shemas and protocols we should support? | 16:32 |
mandel | elopio, maybe they use the same for http and https and that confuses them | 16:32 |
elopio | mandel: from your email I get that we support squid, with no auth, ncsa and kerberos. | 16:33 |
elopio | I'll start with that. | 16:34 |
mandel | elopio, yes, we should.. | 16:34 |
elopio | we should test sometime the other configs. On moztrap I'm defining them as environments, so at least we will know how bad is our coverage. | 16:36 |
mandel | alecu, ralsina, gatox: I know you will enjoy the following: http://www.pointerpointer.com/ | 16:55 |
dobey | gah | 16:56 |
dobey | fixing this test is harder than i thought | 16:56 |
gatox | mandel, sometimes i'm afraid to ask how you get to those pages jejejeje | 16:57 |
mandel | gatox, from thisfred this time :) | 16:57 |
gatox | jajajaa | 16:57 |
dobey | ok wtf | 16:59 |
dobey | how did this test ever not hang then | 16:59 |
dobey | grr | 16:59 |
alecu | ralsina, http://packages.python.org/defer/defer.html#defer.inline_callbacks | 17:01 |
=== zyga is now known as zyga-afk | ||
dobey | ah | 17:03 |
dobey | fixed | 17:03 |
dobey | i think | 17:03 |
ralsina | alecu: http://readthedocs.org/docs/nose/en/latest/api/twistedtools.html | 17:04 |
mandel | ok, eod for me, see you all tom! | 17:13 |
=== yofel_ is now known as yofel | ||
gatox | mandel, bye! | 17:13 |
ralsina | I'll have lunch now. | 17:21 |
* ralsina signs up on the do not ping list | 17:21 | |
dobey | we have a machine that goes ping! | 17:22 |
dobey | i guess i should wait to ping ralsina then :) | 17:31 |
briancurtin | thisfred: do you have any documentation on how you got that share setup? like what versions of liboauth and openssl and such? also, what exactly should i do with that? i think it may be better to just know all of the deps and i can write some script to set that stuff up so its repeatable | 17:38 |
thisfred | briancurtin: unfortunately, no. I never built u1db on windows myself, this was what was passed on to me. | 17:39 |
briancurtin | do you know what bzr branches are required or what i would actually be executing? im not sure what to do with that share in relation to jenkins | 17:40 |
thisfred | We will probably need the MS express compiler, and see how far it gets with make. | 17:40 |
thisfred | briancurtin: so we want to test lp:u1db | 17:40 |
thisfred | trunk | 17:40 |
thisfred | what we want to run is make check, which builds the C code and runs all tests | 17:41 |
thisfred | let me see if I can dig up jam's mail in which he suggested how to start | 17:41 |
thisfred | mandel: do you have jam's response? I remember it not being super elaborate, but my thunderbird has lost the mail. | 17:51 |
thisfred | ah he's eod | 17:57 |
thisfred | found it, briancurtin sent you the mail, though I fear it's not gonna be a huge help | 18:01 |
briancurtin | thisfred: i'll take a look and see what happens | 18:01 |
thisfred | yeah, I can probably translate failures/errors into missing dependencies for you | 18:02 |
briancurtin | thisfred: cool. i'll just start building and running and piecing it together | 18:02 |
thisfred | thanks so much! | 18:02 |
dobey | alecu, ralsina: care to review https://code.launchpad.net/~dobey/ubuntuone-storage-protocol/fix-context-test/+merge/109687 real quick? | 18:10 |
briancurtin | thisfred: i think this will need a bit more work than i'm able to give at this point. the makefile is gcc or whatever specific, won't work with microsoft's nmake | 18:37 |
thisfred | briancurtin: ok, I'll ask jam tomorrow. He had it set up on his own windows machine, but he must have used a different way to build it then | 18:38 |
dobey | he might have used mingw | 18:44 |
briancurtin | that also didnt work, so there's probably something else in play that i dont have setup or know about | 18:45 |
dobey | oh, hmm | 18:45 |
dobey | briancurtin: if it requires gnu make, maybe you need to run "gmake" instead of just "make" ? | 18:46 |
briancurtin | i dont seem to have that setup, i guess ill just wait until jam gets back with exactly what before i go on another wild goose chase like qt testability | 18:51 |
dobey | sure. just suggesting what seems obvious at the moment based on what you said. :) | 18:54 |
thisfred | dobey: pretty sure he used visual studio express | 19:01 |
thisfred | but no idea how | 19:01 |
briancurtin | the makefile references commands like export that don't exist on windows in general. i am going to guess it may have been built via cygwin | 19:02 |
thisfred | I think he may not have used the makefile for building in that case | 19:03 |
alecu | dobey, +1 | 19:09 |
dobey | thanks alecu | 19:10 |
dobey | thisfred: i'm guessing visual studio wasn't used as i don't see any project files in the tree | 19:11 |
thisfred | dobey: and yet that is what I remember him telling me. Perhaps I didn't get all the needed files | 19:13 |
thisfred | I sent a mail asking for more info. | 19:13 |
ralsina | alecu: mailed you a draft, feedback eagerly expected :-) | 19:16 |
ralsina | briancurtin: assigning https://bugs.launchpad.net/ubuntuone-windows-installer/+bug/1011774 to you | 19:18 |
ubot5 | Ubuntu bug 1011774 in Ubuntu One Windows Installer "Set the right version number to the nightly windows installer" [Undecided,Confirmed] | 19:18 |
briancurtin | ralsina: cool, thanks | 19:18 |
elopio | mandel: when I set the squid proxy with NCSA, shouldn't ubuntu ask for my password? | 19:22 |
elopio | or is it just read from the file? | 19:22 |
ralsina | mandel: if you want to see how pointerpointer cheats, put your pointer near the top-right corner and then move it bit by bit | 19:32 |
mandel | elopio, no, it will read from the file and firefox will ask for it (or the u1control panel) | 19:39 |
mandel | thisfred, I do, let me foward it to you | 19:39 |
mandel | oh, I'm not here, I'm a ghost ;) | 19:39 |
dobey | ralsina: does https://code.launchpad.net/~dobey/ubuntuone-storage-protocol/fix-context-test/+merge/109687 look ok to you? | 19:39 |
ralsina | dobey: looking | 19:39 |
thisfred | mandel: thx, found it already ;) | 19:39 |
ralsina | dobey: sure | 19:40 |
ralsina | dobey: right paths and everything | 19:40 |
ralsina | dobey: global approve | 19:40 |
elopio | mandel: yes, I'm not sure what I'm doing wront. But you are not here, I'll bother you tomorrow. | 19:41 |
elopio | *wrong | 19:41 |
mandel | thisfred, replied to the email, there is no need to use the makefile, if you look at it is only python calls | 19:42 |
mandel | thisfred, or you can use the linux term emulator on windows whose name I cannot remember | 19:42 |
mandel | elopio, did you create the password file? | 19:42 |
thisfred | cygwin? | 19:42 |
mandel | thisfred, that one :) | 19:42 |
elopio | mandel: yes, on the squid server. | 19:43 |
briancurtin | so is this a cygwin application or a windows application? | 19:43 |
elopio | but now when I set it as the proxy on my ubuntu client, everything is denied. | 19:43 |
mandel | briancurtin, is a windows app but cygwin make the compilation life easier, also I think he used the easy_install from cygwin which might explain the reason of why the latests version of cython would not compile the thing | 19:44 |
briancurtin | that's too many moving parts in different systems, im just going to wait until there are directions | 19:44 |
mandel | elopio, hm.. and do you get the password dialog, lets say from firefox? | 19:44 |
mandel | briancurtin, that is the main reason was not able to finish the setup, there is no documentation to be found | 19:45 |
mandel | elopio, let me go home (I'm at the office) and connect from there to give you a hand, ok? | 19:45 |
elopio | mandel: no. Just access denied | 19:45 |
mandel | elopio, vuelvo en 10 min max :) | 19:46 |
elopio | mandel: dale. | 19:49 |
dobey | why couldn't unicode() be in python 3, but just deprecated until 3.6 or something, when everyone is actually only using python3 | 19:53 |
ralsina | dobey: because unicode() is an error | 19:53 |
ralsina | dobey: noone should be using it in python 2 either | 19:54 |
dobey | ralsina: which is why everyone uses it everywhere | 19:54 |
ralsina | dobey: yes | 19:54 |
ralsina | dobey: it's tempting and evil | 19:54 |
dobey | and it makes trying to support both 2 and 3 a pain | 19:54 |
ralsina | dobey: you can implement it trivially | 19:54 |
dobey | particularly when doing things like isinstance(blah, unicode) | 19:54 |
ralsina | unicode = str | 19:55 |
ralsina | in that case unicode = basestring (or however it's called) | 19:55 |
mandel_mac | elopio, I'm back! :) | 19:55 |
mandel_mac | elopio, so, lets start from the beginning, did you edit the config or copied the settings file I sent? | 19:55 |
dobey | can't i just solve my problems with violence? | 19:55 |
elopio | mandel_mac: from the beginning, I should say that I have squid3. | 19:56 |
elopio | I edited the config file adding your info. | 19:56 |
mandel_mac | elopio, oh! so you are in 12.04 | 19:56 |
elopio | mandel_mac: yes. I've just found that I thought it was working because there's a directive that allows localhost :) | 19:57 |
mandel_mac | elopio, the diffs are tiny and we can be a smart ass and generate a config file | 19:57 |
mandel_mac | elopio, if you look in lp:ubuntuone-dev-tool in the data folder you can find a squid config template that will work in both squid2 and squid3 | 19:57 |
mandel_mac | elopio, if you look at the template you will see that there are lots of paths that are changed, the main reason is because u1-dev-tools points to _trial for the tests | 19:58 |
elopio | looking. | 19:59 |
ralsina | dobey: want to tackle these test errors? They are related to a branch you did https://pastebin.canonical.com/67818/ | 20:00 |
mandel_mac | elopio, you can even be lazier and import the SquidRunner from ubuntuone/devtools/services/squid.py and tell it to start squid that will generate a conf and will run squid in your system using your user | 20:00 |
dobey | ralsina: they are? | 20:02 |
elopio | mandel_mac: I suppose this is important | 20:03 |
elopio | http_access allow nonauth_port_connected | 20:03 |
elopio | http_access allow password | 20:03 |
ralsina | dobey: I am guessing yes because of the difference between dbus.dict and dict | 20:03 |
mandel_mac | elopio, use, very, did I miss that in the email | 20:03 |
mandel_mac | ?? | 20:03 |
elopio | mandel_mac: yes. | 20:04 |
elopio | I'll try again. | 20:04 |
mandel_mac | elopio, ouch! that is the reason I added the config that worked, just in case I forgot something, sorry | 20:04 |
dobey | ralsina: hrmm, odd | 20:04 |
gatox | mandel_mac, hey! you are copying my nick! jeje | 20:04 |
elopio | :) don't worry. I didn't looked at it until now. | 20:05 |
gatox | ok...... eod for me!! i'm pretty sure that i'm going to finish with all the tests tomorrow!! \o/ | 20:05 |
mandel_mac | gatox, I'm just trying to screw with the autocomplete ;) | 20:05 |
* gatox is excited | 20:05 | |
mandel_mac | gatox, do not touch yourself thinking about tests... | 20:05 |
ralsina | bye gatox! | 20:05 |
mandel_mac | gatox, nos vemos mañana ;) | 20:05 |
gatox | i think tomorrow i'm going to have a u1-client fsevents branch to propose!!!!!! | 20:05 |
gatox | :D | 20:05 |
mandel_mac | gatox, we should do a pylon ar talk about fs integration hehehe | 20:06 |
gatox | mandel_mac, no, i'm hurting myself with the tests jeje | 20:06 |
dobey | ralsina: is there a bug about those tests failing? | 20:07 |
ralsina | dobey: ask facundobatista, he gave me the trace | 20:07 |
ralsina | dobey: I am guessing no | 20:08 |
dobey | facundobatista: ^^ :0 | 20:09 |
elopio | mandel_mac: great. Finally I could see the password dialog. | 20:09 |
dobey | err :) | 20:09 |
elopio | thanks! | 20:09 |
mandel_mac | elopio, no problem, sorry I got back so late :) | 20:09 |
mandel_mac | elopio, wonder.. is there a wiki page in which we could add this? | 20:10 |
elopio | mandel_mac: I'm thinking to add it to wiki.ubuntu.com/UbuntuOne/Testing/Proxy. | 20:10 |
facundobatista | dobey, I don't think so, do you want me to open one? | 20:10 |
dobey | facundobatista: please | 20:10 |
elopio | mandel_mac: but for now, I'm using the pad: http://pad.ubuntu.com/u1-proxy | 20:11 |
mandel_mac | elopio, ok, I think it would be nice to some how script this, doing something similar to the u1-dev-tools plus programatically setting it in gsettings would be really nice | 20:12 |
facundobatista | dobey, 1011822 | 20:12 |
dobey | bug #1011822 | 20:12 |
ubot5 | Launchpad bug 1011822 in Ubuntu One Client "Failing tests in tests.platform.credentials.test_linux" [Undecided,New] https://launchpad.net/bugs/1011822 | 20:12 |
elopio | mandel_mac: yes. | 20:13 |
elopio | mandel_mac: what does that config file from u1-dev-tools do for the localhost? | 20:14 |
elopio | http_access allow localhost | 20:14 |
elopio | shouldn't that be removed? | 20:14 |
mandel_mac | elopio, the runner exposes the settings and the tests force u1 to use them | 20:15 |
dobey | mandel_mac, elopio: there's no reason you couldn't write a test suite on top of the devtools apis, which does what you need for integration testing that stuff | 20:16 |
elopio | mandel_mac, dobey, yes, we could call the devtools from qt-testability, or whatever framework we have to start the proxy. | 20:17 |
elopio | but, the tests and the squid will be in the same machine. So, by having http_access allow localhost we are letting them pass without authentication. | 20:18 |
elopio | or am I missing something? | 20:18 |
dobey | only if connecting over the lo interface | 20:18 |
mandel_mac | elopio, no, they wont | 20:18 |
dobey | if i understand your concern correctly, and that's how squid works | 20:19 |
mandel_mac | what dopey said | 20:19 |
elopio | mandel_mac: I'm trying to figure out why u1 worked on my first test. I had squid and u1 on the same machine, but didn't add the rule to allow the password group. | 20:19 |
dobey | hmm | 20:20 |
mandel_mac | elopio, hm.. we would need to look at the logs | 20:20 |
mandel_mac | elopio, if you did not restart it (u1) I would not be surprised | 20:20 |
elopio | mandel_mac: I'll start from scratch, because now my config is a mess. | 20:21 |
mandel_mac | elopio, ok, perfect, we can catch up tom (I'll be back in 10 hours ;) ) | 20:21 |
dobey | ralsina: actually, i think this is a unicode issue | 20:21 |
ralsina | dobey: really? | 20:21 |
elopio | with the things I changed, now on both machines I get asked for proxy credentials. So that's success. | 20:21 |
dobey | ralsina: \xe1 vs. \xc3\xa1 and \xfa vs. \xc3\xba | 20:22 |
dobey | ralsina: when the tests are run under a spanish locale | 20:22 |
elopio | mandel_mac: go, be fruitful and multiply. I'll ping you tomorrow. | 20:23 |
ralsina | dobey: oh, interesting | 20:23 |
ralsina | dobey: feel free to toss it at gatox if you want ;-) | 20:23 |
dobey | the dbus.Dictionary() change i made only changed a few cases that were passing empty dicts, which is not the case in the tests :) | 20:23 |
mandel_mac | elopio, ok, see you tom :) | 20:24 |
mandel_mac | all, laters! | 20:24 |
dobey | not sure if the issue directly is in sso or client though | 20:25 |
ralsina | dobey: in the meantime, maybe suggest the tests will work using "C" locale and let's make it medium priority | 20:27 |
dobey | the tests do work with C locale. i'm a bit surprised they didn't fail for me though. | 20:31 |
dobey | maybe that string isn't translated to shqip, or doesn't have accents | 20:32 |
ralsina | shqip? | 20:41 |
dobey | albanian | 20:42 |
ralsina | dobey: oh, ok | 20:43 |
ralsina | We should have a translation that translates every letter as ®® | 20:44 |
ralsina | so we get unicode and longer strings everywhere | 20:44 |
dobey | we shouldn't have tests that rely on translations being correct | 20:45 |
ralsina | dobey: well, for some tests it's a requirement, for example when testing that a string is diplayed correctly. That's always going to be locale-dependent. | 20:46 |
ralsina | And here it doesn't depend on that, it depends on the data types being correct | 20:46 |
ralsina | In other words, the words are right, the encoding is wrong | 20:47 |
ralsina | I'm done for the day | 20:48 |
ralsina | will look at mail late tonight, so mail me if you need reviews or anything | 20:48 |
dobey | well, translations shouldn't get loaded at all | 20:49 |
dobey | but yay, isolation | 20:49 |
dobey | oh, interesting | 21:08 |
dobey | the tests are indeed failing for me now | 21:08 |
=== m_conley is now known as m_conley_away | ||
theseb | is ubuntu client buggy? | 21:19 |
theseb | it won't start on 12.04 and won't finish syncing | 21:19 |
theseb | what should i do? | 21:19 |
dobey | why won't it start? | 21:22 |
theseb | dobey: i get same error message | 21:26 |
theseb | dobey: as in a 1+ year old bug report | 21:26 |
theseb | i'll do it again | 21:26 |
theseb | CredentialsError | 21:27 |
theseb | DBusException(dbus.String(u'Traceback (most recent call last):\n File "/usr/lib/python2.7/dist-packages/dbus/service.py", line 707, in _message_cb\n retval = candidate_method(self, *args, **keywords)\n File "/usr/lib/python2.7/dist-packages/ubuntuone-client/ubuntuone/platform/credentials/linux.py", line 185, in find_credentials\n reply_handler=reply_handler, error_handler=error_handler)\n File "/usr/lib/python2.7/dist-packages/dbus/proxies.py", line | 21:27 |
theseb | 137, in __call__\n **keywords)\n File "/usr/lib/python2.7/dist-packages/dbus/connection.py", line 584, in call_async\n message.append(signature=signature, *args)\nValueError: Unable to guess signature from an empty dict\n'),) | 21:27 |
theseb | dobey: there you go | 21:27 |
dobey | theseb: have you installed a custom python in /usr/local? | 21:27 |
theseb | no....recently installed ubuntu 12.04 | 21:27 |
theseb | no fancy mods | 21:27 |
dobey | installed cleanly? | 21:27 |
theseb | dobey: yup | 21:28 |
dobey | what does running "/usr/bin/env python" say? | 21:28 |
theseb | unless some app quietly installed a custom python w/o telling me | 21:28 |
theseb | good question | 21:28 |
theseb | (laptop /home/cs) % /usr/bin/env python | 21:28 |
theseb | Python 2.7.3 (default, | 21:28 |
theseb | ...etc. | 21:28 |
theseb | % which python | 21:28 |
theseb | /usr/bin/python | 21:28 |
dobey | hrmm | 21:29 |
daker-cloud | dobey: bug 711162 ? no ? | 21:29 |
ubot5 | Launchpad bug 711162 in Ubuntu One Client stable-3-0 "ubuntuone-login crashed with ValueError in call_async(): Unable to guess signature from an empty dict" [Undecided,Triaged] https://launchpad.net/bugs/711162 | 21:29 |
dobey | daker-cloud: yes i know what bug # it is. it's open in my browser already :) | 21:29 |
daker-cloud | ok | 21:29 |
dobey | rye: ^^ | 21:30 |
dobey | theseb: what does running "/usr/lib/ubuntu-sso-client/ubuntu-sso-login-qt --app_name Test" do? | 21:31 |
theseb | dobey: it asks me to sign up to test | 21:31 |
theseb | dobey: there was a form that popped up | 21:31 |
theseb | is that good? | 21:31 |
dobey | odd. yes it's supposed to do that | 21:35 |
dobey | theseb: do you have any processes currently running with ubuntu-sso or ubuntuone in the process name? | 21:35 |
theseb | dobey: yes... % ps -ef | grep ubun | 21:36 |
theseb | cs 1872 1293 0 00:22 ? 00:00:00 gnome-session --session=ubuntu | 21:36 |
theseb | cs 1907 1872 0 00:22 ? 00:00:00 /usr/bin/ssh-agent /usr/bin/dbus-launch --exit-with-session gnome-session --session=ubuntu | 21:36 |
theseb | cs 1910 1 0 00:22 ? 00:00:00 /usr/bin/dbus-launch --exit-with-session gnome-session --session=ubuntu | 21:36 |
theseb | cs 2084 1 0 00:22 ? 00:00:00 /usr/lib/ubuntu-geoip/ubuntu-geoip-provider | 21:36 |
theseb | cs 2339 1 1 00:23 ? 00:15:29 /usr/bin/python /usr/lib/ubuntuone-client/ubuntuone-syncdaemon | 21:36 |
theseb | cs 3996 1 0 00:43 ? 00:00:00 /usr/bin/python /usr/lib/ubuntuone-client/ubuntuone-login | 21:36 |
theseb | cs 17131 1 0 16:31 ? 00:00:00 python /usr/lib/ubuntu-sso-client/ubuntu-sso-login | 21:36 |
dobey | theseb: please don't paste them all | 21:36 |
dobey | theseb: use paste.ubuntu.com if you want to paste the list :) | 21:37 |
theseb | dobey: sorry...perhaps all these crashed ubuntuone instances are creating lots of zombie processes? | 21:37 |
theseb | dobey: should i kill -9 all that stuff you see? | 21:37 |
dobey | it doesn't seem like they are zombies | 21:37 |
dobey | theseb: what does u1sdtool -q do? | 21:37 |
theseb | dobey: i SHOULD see syndaemon and login? | 21:37 |
theseb | % u1sdtool -q | 21:38 |
theseb | ubuntuone-syncdaemon stopped. | 21:38 |
theseb | stopped syncdaemon i guess | 21:38 |
dobey | theseb: ok, are the processes all still there, ore gone? | 21:38 |
dobey | or even | 21:38 |
theseb | i still see syncdaemon and other stuff | 21:38 |
theseb | login and sso-login | 21:38 |
dobey | theseb: ok, then yes, i think kill -9 them | 21:39 |
theseb | YEAH! | 21:40 |
theseb | when i kill everything associated w/ ubuntuone and restart the client it works! | 21:40 |
theseb | dobey: thanks so much | 21:40 |
theseb | dobey: of course i wish it never died at all but this is a doable workaround for a while i guess | 21:41 |
dobey | cool | 21:41 |
dobey | glad it's working now at least | 21:41 |
dobey | seems something got stuck and dbus wasn't working right for it | 21:41 |
theseb | dobey: great! now i know what to check for....leftover ubuntuone crud...awesome | 21:42 |
dobey | theseb: not sure why it is getting stuck like that. that is weird :) | 21:43 |
dobey | well, past time for me to go | 21:48 |
dobey | later all | 21:48 |
=== salgado is now known as salgado-afk | ||
=== eu is now known as Guest70325 | ||
=== Guest70325 is now known as adorilson |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!