| mmcc | is there a way to see the contents of the offloaded queue in some readable fashion? | 00:00 |
|---|---|---|
| mmcc | one twist here is that I added a ton of files, then had to kill syncdaemon then restarted it and removed all those files. When it came back up, it had a lot of AQ_FILE_NEW_ERRORs because the files weren't there anymore as it processed the queue | 00:04 |
| mmcc | I'm wondering if those errors stopped the events in the queue from running, but their presence in the queue is stopping progess? | 00:05 |
| mmcc | progress | 00:05 |
| mmcc | ok, done for the day. I'm going to leave it running for a while and see if anything kicks in to restart progress | 00:08 |
| JamesTait | Good morning all! :) | 08:21 |
| mandel | morning | 08:25 |
| * mandel reboots to test new unity | 09:47 | |
| rye | mandel, ralsina: what happens when i run u1sdtool on windows w/o control panel running? I see syncdaemon starting but u1sdtool does not release the terminal. Then I Ctrl+C the process, run u1sdtool again and get the second set of ubuntuone-syncdaemon process | 09:47 |
| rye | ah, looks like metadata is corrupted | 10:04 |
| mandel | rye, uh, no idea, I have never used u1sdtool on windows | 10:12 |
| mandel | rye, sorry for the later response, I'm constantly logging in and out to test the new unity I compile.. looks like the standalone apps do not work :( | 10:12 |
| ralsina | rye: I don't think I ever tried it without u1cp running | 10:38 |
| mandel | ralsina, morning! | 10:47 |
| * mandel logs out to test the new unity | 10:48 | |
| gatox | good morning! | 11:19 |
| mandel | ralsina, I'm back! | 12:04 |
| mandel | ralsina, maldito movistar.. | 12:04 |
| mandel | ralsina, did you get my messages? | 12:04 |
| mandel | gatox, do you copy me? | 12:07 |
| gatox | mandel, yap..... i saw your twit | 12:07 |
| mandel | gatox, ok, so I'm back :P | 12:08 |
| gatox | :P | 12:08 |
| mandel | alecu, I have seen several times the scope crashing when testing the preview, please let me know if you have anything I can use to do a better try | 12:19 |
| ralsina | mandel: now I did! | 12:24 |
| ralsina | and good morning! | 12:24 |
| mandel | ralsina, he, I though you were up, fixed the network already | 12:25 |
| dpm | hi everyone. I was wondering if someone could help me with a problem with u1 | 12:30 |
| dpm | on quantal | 12:30 |
| mandel | dpm, you should be more demanding ;-) | 12:31 |
| dpm | ok | 12:31 |
| dpm | mandel, you MUST help me! | 12:31 |
| dpm | I know where you live | 12:32 |
| mandel | hehehe | 12:32 |
| mandel | dpm, I can try :) | 12:32 |
| dpm | cool :) | 12:32 |
| ralsina | mandel: I am usually up at 7AM, catchup on mail, see if anyone needs me on IRC then I leave for an hour or so | 12:33 |
| dpm | mandel, so every time I start the UI, the status remains as either "Loading..." or "Starting sync..." (or something similar). All my files are synced already, so there is no need to sync. In any case, I don't seem to get u1 to start. Any tips on debugging this, or getting it to work? | 12:33 |
| ralsina | mandel: so, up, but not here :-) | 12:33 |
| dpm | also 'u1sdtool --status' hangs up forever | 12:34 |
| mandel | dpm, well, if you wanted to take a look on what is going on you can kill the u1 daemon and start it again with --debug so that we see what is going on in the stdout | 12:34 |
| mandel | dpm, can you try that? | 12:35 |
| dpm | mandel, yeah, what's the name of the daemon? | 12:35 |
| dpm | askubuntu to the rescue: http://askubuntu.com/questions/40518/how-to-kill-the-ubuntuone-daemon | 12:36 |
| mandel | dpm, ubuntuone-syncdaemon | 12:36 |
| dpm | ok, I didn't see any errors on the debug output, and also 'u1sdtool --status' returns some output now. Let me see what the UI does | 12:38 |
| gatox | ralsina, mandel dobey what are we using now to replace expanduser?? it used to be expand_user..... but i can't find it.... | 12:39 |
| mandel | gatox, why do you want to expand user? | 12:39 |
| ralsina | gatox: ubuntuone.platform.expand_user | 12:40 |
| mandel | gatox, should be using the dirspec stuff, right? | 12:40 |
| gatox | ralsina, i thought that was on sso | 12:40 |
| ralsina | mandel: dirspec has dirspec.utils.user_home but not a expand_user | 12:40 |
| ralsina | gatox: nop | 12:40 |
| gatox | ralsina, ack...... just get confuse..... thanks | 12:41 |
| ralsina | gatox: the one that was on sso is now in dirspec | 12:41 |
| mandel | ralsina, we should be using drispec unless we are really really sure | 12:41 |
| gatox | mandel, but dirspec doesn't have expand_user | 12:41 |
| ralsina | mandel: dirspec has just a constant | 12:41 |
| ralsina | mandel: and expnd_home uses that constant :-) | 12:41 |
| dpm | mandel, it seems to stay in INIT status forever: http://pastebin.ubuntu.com/1230328/ | 12:41 |
| mandel | dpm, what is the ouput of the daemon? | 12:42 |
| mandel | ralsina, ah, ok | 12:42 |
| dpm | mandel, https://pastebin.canonical.com/75452/ - sorry, the initial output got lost, not enough scrollback. Let me kill & restart the daemon to get full output | 12:47 |
| mandel | dpm, movistar is being annoying this morning and I can't get to that pastebin.. can you send me an email with it? | 12:49 |
| dpm | sure | 12:50 |
| mandel | dpm, thx, I'm off to a quick lunch and I'll ping you back asap, ok? | 12:51 |
| === mandel is now known as mandel|lunch | ||
| dpm | mandel|lunch, sure thanks, and enjoy! | 12:52 |
| dobey | hmm | 12:57 |
| dobey | dpm: what's wrong exactly? do you have a ~/.cache/ubuntuone/log/syncdaemon-exceptions.log file that is non-zero in size? | 12:58 |
| dpm | dobey, what's wrong is that the UI stays in "File sync starting..." forever. I've got several log files. Do you need to see the content of any of these? -> http://pastebin.ubuntu.com/1230377/ | 13:09 |
| dobey | dpm: mostly i'm interested in what syncdaemon-exceptions.log says right now | 13:10 |
| dpm | dobey, here it is: http://paste.ubuntu.com/1230380/ | 13:11 |
| dobey | dpm: apt-get install gir1.2-syncmenu-0.1 | 13:12 |
| dobey | dpm: install that, and restart the syncdaemon, and you should be fine again | 13:17 |
| dpm | great, thanks dobey, that sorted it | 13:24 |
| dobey | sure. :) | 13:25 |
| dpm | mandel|lunch, it seems it's fixed already, thanks for your help too ^ | 13:25 |
| dobey | it's a Recommends, so if you have your system configured to not install them by defaulted, it might not have gotten installed on upgrade. the resulting bug you hit is fixed in trunk and will be fixed in 4.0.0 in quantal, but workaround for now is to install that gir package :) | 13:26 |
| gatox | :S the change to expand_user broke my tests | 13:30 |
| dobey | gatox: broke how? | 13:31 |
| gatox | dobey, problems that expand_user is expecting bytes....... broke several tests, but the fix should be only in one place....... | 13:32 |
| dobey | ah | 13:32 |
| === m_conley_away is now known as m_conley | ||
| dobey | [>....................] resync = 0.4% (19335616/3906763776) finish=319.0 | 13:59 |
| dobey | min speed=203040K/sec | 13:59 |
| dobey | that's going to take a while... | 13:59 |
| ralsina | dobey: there is a /proc file to raise maximum speed | 14:03 |
| ralsina | dobey: but yes, it's going to take a day or so :-) | 14:03 |
| ralsina | dobey: because that estimate usually *under*estimates | 14:03 |
| alecu | hello all! (says a late alecu) | 14:03 |
| dobey | that "min" wasn't for the speed | 14:03 |
| dobey | it was "319.0 min" left | 14:03 |
| dobey | yay random line breaks | 14:04 |
| ralsina | 203040K/sec is quite quickie :-) | 14:04 |
| dobey | SATA III is nice :) | 14:05 |
| === mandel|lunch is now known as mandel | ||
| mandel | alecu, ping | 14:11 |
| mmcc | morning folks | 14:16 |
| gatox | mmcc, hi | 14:20 |
| mmcc | hi gatox! | 14:21 |
| alecu | mandel: pong | 14:24 |
| mandel | alecu, do you have a working scope or are still stuck with the vala problem? | 14:24 |
| mandel | alecu, I have been using a very old version and I'm seeing some crashes, is that the case what you have? | 14:25 |
| alecu | mandel: I got past that, and now have a nice solution for testing async code in the standard synchronous tester. | 14:25 |
| mandel | alecu, awesome! | 14:25 |
| alecu | mandel: but the current branch seems to have some issues when running IRL | 14:25 |
| alecu | mandel: I'm taking a look at it right now, and will propose it after the team meeting. | 14:26 |
| mandel | alecu, ok, I just wanted to see the entire thing working since the ui is nearly done, I'm focusing on automated testing atm | 14:26 |
| mmcc | So, does anyone have tips on debugging a stuck syncdaemon queue? | 14:31 |
| mmcc | I have a syncdaemon doing nothing since yesterday afternoon, but it has 200 items in its queue and many more in the offload queue | 14:32 |
| mmcc | something is blocking progress, and I'm wondering if there's a way to look at what's in the offloaded queues | 14:32 |
| mandel | mmcc, the expert you are looking for is called facundobatista :) | 14:34 |
| chaselivingston | mmcc: installed today's build and i'm stuck at getting information, please wait | 14:35 |
| facundobatista | mmcc, do you have tracebacks in the log? | 14:35 |
| chaselivingston | mmcc: and i have 7 sd's running | 14:36 |
| mmcc | facundobatista: I don't think so, let me check again | 14:38 |
| mmcc | chaselivingston: definitely want to kill all those sd's and start again | 14:38 |
| chaselivingston | mmcc: trying that now | 14:38 |
| dobey | how the heck did you have *7* syncdaemons running? | 14:39 |
| chaselivingston | dobey: wonderful question | 14:39 |
| mmcc | chaselivingston: when you quit the app, how are you quitting it? only the quit item in the u1 icon on the right side of the menubar actually quits syncdaemon too | 14:39 |
| mmcc | if you do cmd-q, it leaves sd up | 14:39 |
| chaselivingston | mmcc: ah, i've been using cmd-q | 14:39 |
| mmcc | now, why it doesn't just reconnect to the same sd… that's a bug in IPC somewhere I guess | 14:39 |
| ralsina | mmcc: we need to find how to catch cmd-q :-/ | 14:40 |
| mmcc | ralsina: yeah, it's next on my list | 14:41 |
| chaselivingston | mmcc: using Ubuntu One > Quit from the menubar seems to have it locked up now | 14:41 |
| mmcc | chaselivingston: another end of the same bug. Note that using quit from the "Ubuntu One" menu that's just to the right of the apple menu is the same as cmd-q | 14:42 |
| chaselivingston | mmcc: ah ok | 14:42 |
| mmcc | chaselivingston: the menu item with the red U that's always there in every app, over to the right, that's the one that'll also kill syncdaemon | 14:43 |
| chaselivingston | mmcc: clicking the red u does nothing for me | 14:46 |
| mmcc | facundobatista: my syncdaemon-exceptions.log has about 10 exceptions from about the time when I noticed it getting wedged: https://pastebin.canonical.com/75471/ | 14:46 |
| mmcc | chaselivingston: if it's already not responsive, yeah the red u won't show either. time to kill it with activity monitor | 14:46 |
| chaselivingston | mmcc: piled up a bunch of sds again | 14:47 |
| mmcc | chaselivingston: that's confusing - did you kill the 7 SDs, then restart and get more than one new one? | 14:48 |
| chaselivingston | mmcc: apparenlty | 14:48 |
| mmcc | er, by restart I mean restart control panel | 14:48 |
| mmcc | chaselivingston - yikes. well, I guess I'd appreciate seeing your syncdaemon.log, then | 14:50 |
| chaselivingston | mmcc: https://pastebin.canonical.com/75474/ | 14:51 |
| mmcc | chaselivingston: well I can fix that exception for you, but I'm not sure why it keeps spawning new ones… can you also send controlpanel.log ? :) | 14:52 |
| mmcc | chaselivingston: to fix it, delete ~/Library/Preferences/ubuntuone/syncdaemon.conf | 14:53 |
| chaselivingston | mmcc: https://pastebin.canonical.com/75476/ | 14:53 |
| mmcc | that is, first kill everything, delete that file, then start the app again | 14:53 |
| chaselivingston | mmcc: ah, now the control panel loads all the way | 14:54 |
| chaselivingston | mmcc: and file sync is up to date | 14:55 |
| mmcc | chaselivingston: good. you were the victim of an incompatible default conf change I made a week or so ago and since fixed. Sorry… | 14:55 |
| chaselivingston | mmcc: haha, thanks for fixing that | 14:56 |
| mmcc | chaselivingston: you're welcome, glad it was simple | 14:57 |
| briancurtin | mumble in a few minutes? | 14:58 |
| dobey | briancurtin: afaik | 14:58 |
| mandel | that is the plan, yes | 14:58 |
| briancurtin | hopefully my mumble works on this new setup | 14:59 |
| ralsina | oops team, to the mumble! | 15:00 |
| ralsina | briancurtin, joshuahoover mumble? | 15:02 |
| briancurtin | just joined | 15:02 |
| ralsina | DMN | 15:03 |
| ralsina | This worked 20 minutes ago!!!! | 15:03 |
| ralsina | let's do this tomorrow :-( | 15:04 |
| ralsina | Sorry team | 15:04 |
| briancurtin | does that mean you have to twitter shame yourself? | 15:04 |
| ralsina | yes it does | 15:04 |
| dobey | ok | 15:04 |
| ralsina | I don't understand what's wrong with my audio, it worked every time until last week :-( | 15:05 |
| ralsina | it even worked in my previous call | 15:06 |
| ralsina | sigh | 15:06 |
| dobey | heh | 15:06 |
| ralsina | https://twitter.com/ralsina/status/251337090955087872 | 15:06 |
| gatox | dobey, i'm having some problems with some tests when they tried to use the user_home var from dirspec...... maybe you can help me | 15:07 |
| mandel | oh | 15:08 |
| dobey | gatox: sure. are you getting unicode objects instead of str objects, in those cases? | 15:08 |
| mandel | ralsina, normal stand up then? | 15:08 |
| ralsina | mandel: yes | 15:08 |
| mandel | me | 15:09 |
| mandel | :P | 15:09 |
| gatox | dobey, i'm trying to run some tests for u1-cp...... but when the expand_user function is executed in u1-client (which has an assert to ensure that receives and return bytes), after doing the operation of replace('~', user_home)..... with user_home being: from dirspec.utils import user_home ...... it fails because the result is an unicode obj..... because user_home is unicode..... and i'm not touching that in any of my tests..... do you know | 15:09 |
| gatox | who might be modifying that? | 15:09 |
| dobey | me | 15:09 |
| briancurtin | i need to type | 15:09 |
| ralsina | alecu, dobey, mandel, gatox, briancurtin, mmcc: standup at will | 15:09 |
| facundobatista | mmcc, most surely a bug I'm in the process of fixing (one branch already in, need to send the second), to unlock it you need to restart SD | 15:09 |
| mmcc | facundobatista: ok, cool. thanks for looking. let me know if there's anything I can do to help | 15:09 |
| dobey | gatox: it's a bug in contrib.testing.testcase.BaseTwistedTestCase i think | 15:10 |
| ralsina | alecu, dobey, mandel, gatox, briancurtin, mmcc: forget it no standup wither | 15:10 |
| ralsina | I know what you guys are doing | 15:10 |
| ralsina | and the missing call is my fault, no need to make syou suffer for it | 15:10 |
| facundobatista | mmcc, are you using nightlies? | 15:10 |
| dobey | ok | 15:10 |
| mmcc | facundobatista: I am using trunk, let me check if I'm out of date | 15:10 |
| gatox | dobey, thx! i'll take a look at that... i was trying to find who is creating the path to _trial_temp but i couldn't remember/find it | 15:10 |
| mandel | we standup or not? | 15:11 |
| mandel | ok.. stupid lag.. | 15:12 |
| mmcc | facundobatista: I'm slightly out of date, but I'm not missing any revisions from you | 15:12 |
| * mandel goes back to c++ | 15:13 | |
| dobey | gatox: there is also a related issue to that in dev-tools, but the main issue here is the testcase in u1client | 15:13 |
| mmcc | facundobatista: also, this is on mac os x, just in case that is important to know :) | 15:13 |
| briancurtin | mandel: you could have also typed "goes back to nightmare" | 15:13 |
| gatox | dobey, ok..... i'll try to fix that..... i'm blocked until that is working | 15:13 |
| facundobatista | mmcc, not really, thanks :) | 15:13 |
| mandel | briancurtin, lol yes, a little, and is not just the language ;-) | 15:14 |
| rhys_ | Can anyone tell me some reasons for why Ubuntu One is not working for me? I replace files on one machine, they don't get updated on the others. I delete files on one machine, I then replace them. They sync up. I boot another machine, Ubuntu One syncs, the files are deleted. So I go back to the machine that I replaced them on, only to find they've been deleted from there since I booted the new machine. | 15:17 |
| rhys_ | I'm a sysadmin and have tried completely resetting ubuntu one per the wiki twice now. | 15:17 |
| rhys_ | I've tried a few different things, but mostly it seems when I boot my laptop or separate machine, changes just get borked. | 15:18 |
| rhys_ | This is across multiple new installs of 12.04.1 | 15:18 |
| mandel | rhys_, I think rye can help you, can you rye? | 15:22 |
| dobey | gatox: i need to get lunch, but i can help more after if needed. | 15:25 |
| gatox | dobey, thx...... luunch here too i think...... | 15:26 |
| gatox | dobey, i'm looking at that code right now..... i was going to do some tests | 15:26 |
| dobey | ok | 15:26 |
| gatox | dobey, i'll let you know, don't worry | 15:26 |
| dobey | bbiab :) | 15:26 |
| rye | rhys_: may I ask you to collect and send the logs for both machines using the script - http://people.canonical.com/~roman.yepishev/us/ubuntuone-ubuntu-syncdaemon-diag.sh | 15:37 |
| rhys_ | rye: ok. The problem is this is kinda an ongoing thing. I'm afraid if I just send you both the logs they will be too filled with noise for the bug to show up | 15:38 |
| rhys_ | because the truth is it *mostly* works. | 15:38 |
| rye | rhys_: ok, are the files removed edited in any specific application? | 15:39 |
| rhys_ | but like i put .xchat2 into the sync. Initially it works, all machines get my settings, transparent background and whatnot. Yesterday I added a few new channels to autojoin. I saw the gnome3 notification that a handful of files in that directory were updated and syncing. Today my new machine tells me Ubuntu One is in sync, but the autojoin is not there. | 15:40 |
| rhys_ | rye: KeePass2 happens occasionally, which I've already seen a bug for. Its a .NET application, which I can understand. .purple (pidgin) does not work at all. | 15:41 |
| rhys_ | the removed files are done via nautilus or bash. | 15:41 |
| rhys_ | well. the rm command. | 15:41 |
| === gatox is now known as gatox_lunch | ||
| rhys_ | rye: if you want a reproducible error, make two machines and try to sync a .purple directory. Very quickly Ubuntu One will hang in "Syncing Files" forever | 15:50 |
| rye | rhys_: do you have any "Traceback" in ~/.cache/ubuntuone/log/syncdaemon.log ? | 15:52 |
| mmcc | hi facundobatista, I've got an empty queue now but the web UI is showing directories on the server that no longer exist on my desktop. Is this part of the same bug? I removed the parent directory I was using for my stress test and saw AQ_UNLINK_ERROR in syncdaemon.log: https://pastebin.canonical.com/75484/ | 15:54 |
| mmcc | I'm thinking that our error feedback could be better… | 15:55 |
| rhys_ | rye: there are no traceback | 15:56 |
| rhys_ | no tracebacks on either of the 2 machines I can get to | 15:56 |
| rhys_ | Ok. so I updated the autojoin list. .xchat2/srvlist_.conf was just updated. I can see it on the webGUI. | 15:58 |
| rhys_ | Right now Xchat is running on my home machine, which is set also to be in sync. | 15:59 |
| rhys_ | I will go home tonight, record exactly what I do, and see if it works. If it doesn't, I'll send you the logs per that script. | 15:59 |
| rhys_ | for both machines | 15:59 |
| mmcc | rhys_: I have a dumb question, is this only happening when your u1 updating the conf files of a running program? | 16:02 |
| rhys_ | mmcc: it very well could be. | 16:02 |
| mmcc | In that case, I wonder if the running app is prepared for changed conf files, you know? | 16:03 |
| rhys_ | simple updates always work as far as my memory stretches. this is always concerning the variations of running programs. I understand such concurrency is hard to do. what does ubuntu one look for? mtime? | 16:03 |
| rhys_ | if mtime is the only value, a lot of programs which constantly touch files could be the source of all my woes. | 16:04 |
| mmcc | disclaimer that I'm not really a linux guy (yet) so maybe that's not a problem due to some common conf file design pattern that I don't know of... | 16:04 |
| rhys_ | design pattern. lol. a lot of software is written with "whatever sounds good." such considerations of external programs updating files as they're being used is probably not a high consideration. | 16:05 |
| mmcc | rhys_: what I was wondering is if you start e.g. xchat then change its conf file, does it have any way of knowing that you've changed it, or does it just read it once at startup? | 16:05 |
| mandel | mmcc, if it uses pyinotify yes, similar as in mac, you have to pay attention | 16:06 |
| mandel | mmcc, same as on windows, you place a watcher that tells you the app changed and there you go, asp.net pages do that for example | 16:06 |
| mmcc | mandel: of course it *could* watch, but does it? | 16:06 |
| rhys_ | mmcc: i'm worried about a worse problem. when i go home tonight and shutdown xchat, it will write that file with its current settings, making it have the newest version, which then overrides the changes I made today. I don't think it does that, but if it does that would explain it. | 16:06 |
| mandel | mmcc, AFAIK xchat does not | 16:07 |
| mandel | mmcc, depends on the app | 16:07 |
| mmcc | mandel: right. I bet most don't expect their conf files to be written to while they're running | 16:07 |
| mandel | mmcc, yes, most don't expect it, you sometimes can tell them to reload them, of if the use gconf they might get a signal, but that is about it | 16:08 |
| mmcc | rhys_: yep, that could be a problem, and as mandel points out, it'll depend on what each app does | 16:08 |
| mandel | rhys_, oneconf had the idea of letting you sync some of the configs, but I don't know how far it got | 16:10 |
| mandel | rhys_, and ofcourse some settings cannot be synced, as in, it does not make sense to sync them | 16:10 |
| mmcc | in the future, maybe we can all have settings synced in u1db… | 16:11 |
| mmcc | brb, coffee | 16:11 |
| rhys_ | true. But I would love to have application profiles for things like Thunderbird, Firefox, Pidgin, Xchat, Terminator. | 16:12 |
| rhys_ | I'd write them if given good documentation | 16:13 |
| mandel | ok, EOD here | 16:22 |
| mandel | see you all tom! | 16:22 |
| ralsina | lunchtime + banks | 16:32 |
| ralsina | will be back asap | 16:32 |
| === gatox_lunch is now known as gatox | ||
| facundobatista | mmcc, probably yes | 16:46 |
| facundobatista | mmcc, try doing a rescan from scratch | 16:46 |
| mmcc | facundobatista: ok. using u1sdtool, right? dumb question, what is the volume id to pass to --rescan-from-scratch, is it the node_id that I get from u1sdtool --info=$HOME/Ubuntu\ One | 16:49 |
| mmcc | ? | 16:49 |
| facundobatista | mmcc, --rescan-from-scratch="" | 16:49 |
| mmcc | facundobatista: ack, thanks! | 16:49 |
| dobey | only ∞ more bugs to go | 17:08 |
| dobey | whee, and 2 more hours until the raid is initialized | 17:16 |
| === trijntje_ is now known as trijntje | ||
| mmcc | in other news, changing the fsevents daemon from a concurrent queue to a serial queue seems to be working fine, but I lost the logs to check because they got rotated out :) | 17:39 |
| mmcc | looking at logging.conf, it looks like I can set the filesize really high, but maybe can't tell it to just never rotate… anyone know different? | 17:39 |
| dobey | you don't have both copies when they get rotated on osx? | 17:41 |
| mmcc | dobey: yes, until it fills all 5 that it's configured to keep around | 17:43 |
| briancurtin | i hate pkg_resources | 17:47 |
| dobey | mmcc: you can increase that number as well as the log file size itself; but outright stopping the rotation is a bit non-trivial i think, as the logger is using a rotating log class | 17:49 |
| mmcc | dobey: yeah, I ended up looking at the code. much easier to fake it with a big filesize or large #files. thanks | 17:49 |
| dobey | what was that python module to use for parsing time strings and getting the difference between them in seconds or days or whatever? | 17:50 |
| briancurtin | dobey: in the stdlib? datetime.datetime, and then i believe adding/subtracting between datetime's produces a datetime.timedelta | 17:54 |
| dobey | ah, timedelta, yes | 17:55 |
| * alecu misses python | 17:58 | |
| dobey | hrmm, but how do i get a datetime() from an ISO string? | 17:59 |
| dobey | '2012-09-25 18:07:09.697594+00:00' for example | 18:00 |
| briancurtin | dobey: i'll give it a shot | 18:00 |
| dobey | sad thing is that i remember doing this before, but i have no idea where exactly :) | 18:02 |
| dobey | ah, nevermind | 18:03 |
| dobey | i don't need to parse the string actually | 18:04 |
| dobey | launchpadlib gives me a datetime.datetime already! :) | 18:04 |
| briancurtin | ah nice | 18:04 |
| alecu | I was about to suggest http://wiki.python.org/moin/WorkingWithTime | 18:05 |
| alecu | but, that sounds much better | 18:05 |
| gatox | i need to take a break....... i'll be back in 20min | 18:06 |
| * gatox kicks the door | 18:07 | |
| === gatox is now known as gatox_brb | ||
| dobey | crap | 18:20 |
| dobey | TypeError: can't subtract offset-naive and offset-aware datetimes | 18:20 |
| dobey | :( | 18:20 |
| dobey | and datetime.utcnow() or datetime.now() only give offset-naive times | 18:21 |
| ralsina | Hello again | 18:27 |
| === gatox_brb is now known as gatox | ||
| dobey | how do i get a datetime object *with* offset for datetime.utcnow()? :( | 18:31 |
| dobey | brb | 18:37 |
| mmcc | dobey, when you get back, should there be an updated systray icon in the client-data project? | 18:51 |
| mmcc | oh never mind, I was only looking at the big resolutions | 18:52 |
| dobey | mmcc: i think we need a different icon for the mac menu perhaps. | 19:13 |
| mmcc | dobey: yes, now that I look through them, it looks like we don't have anything in there that fits | 19:15 |
| mmcc | style-wise, that is | 19:15 |
| dobey | mmcc: do those need to be in .icns also? | 19:17 |
| mmcc | dobey: I think so. I just tried a hacky test to set the systray QIcon to a .icns, and it didn't show up, but didn't give me any errors to work with either | 19:19 |
| mmcc | I think they'll need to be .icns so that they'll look right on a retina screen | 19:19 |
| dobey | is there a reference for menubar icons with relation to 'retina' as well? | 19:20 |
| mmcc | dobey: looking | 19:23 |
| dobey | man, the datetime api is really annoying :( | 19:27 |
| gatox | dobey, just to let you know..... i'm still trying to find who is modifying the user_home var..... but i'm not being able to find that yet | 19:27 |
| ralsina | gatox: saw your tweet ... querés un tilo? ;-) | 19:32 |
| dobey | gatox: the BaseTwistedTestCase is setting HOME | 19:32 |
| ralsina | dobey, gatox: just in case, we are supposed to ignore $HOME | 19:32 |
| dobey | or well, changing HOME | 19:32 |
| dobey | or was it the temp dir | 19:33 |
| gatox | ralsina, i know...... | 19:33 |
| mmcc | dobey: if we were using the apple API directly, we could send it a PDF for the status item image. I don't know if Qt will be happy with that though, I'm going to test now | 19:33 |
| gatox | dobey, yes...... it seems..... but it's really weird..... that part of the code seems to not being executed.... i'll keep testing that | 19:34 |
| dobey | gatox: testcase.py:396 | 19:34 |
| dobey | err | 19:34 |
| dobey | gatox: contrib/testing/testcase.py:396 | 19:34 |
| gatox | dobey, yep.... i'm there | 19:34 |
| gatox | dobey, ok..... let me keep trying some stuff and i'll let you know | 19:35 |
| dobey | gatox: what do you mean it's not being executed? | 19:35 |
| gatox | dobey, i've modified the python the PYTHONPATH to point to the u1-client that i have, where i have lots of prints....... and some of them i can see it in the console..... but the ones related to the testcase.py file not | 19:37 |
| dobey | gatox: where is your test code? | 19:37 |
| gatox | dobey, in u1-cp | 19:38 |
| dobey | gatox: in trunk? where can i see what you're doing exactly? | 19:38 |
| mmcc | so, launchpad finally let me propose this branch -- fixes the sync_menu dummy for windows and macs, and adds a test_common: https://code.launchpad.net/~mikemc/ubuntuone-client/fix-dummy-sync-menu/+merge/126766 | 19:38 |
| mmcc | btw, not sure I like the convention of naming things "common" that are only common to two platforms. not sure I like "nonlinux" better though | 19:39 |
| * mmcc is not sure about so many things | 19:39 | |
| gatox | dobey, nono..... i didn't push any code..... i'm working here with a modified version of u1-cp and u1-client to try to fix the tests..... everything was working but with expand_user the new tests are broken.... so i'm trying to find out who is touching the user_home var and converting that into unicode | 19:39 |
| gatox | dobey, let me debug a little bit longer | 19:39 |
| mmcc | also, that review is pretty short, might be a nice escape from say, pkg_resources, briancurtin :) | 19:42 |
| briancurtin | mmcc: i'll review but still can't run CP IRL because of pkg_resources. can run tests, though | 19:44 |
| mmcc | briancurtin, ouch. well, this is a client branch, so maybe that's easier? | 19:44 |
| dobey | gatox: what is the MRO of your test case? | 19:45 |
| gatox | dobey, mro? | 19:45 |
| dobey | gatox: class hierarchy | 19:45 |
| briancurtin | mmcc: oh, yeah. looking at it now | 19:45 |
| === yofel_ is now known as yofel | ||
| gatox | dobey, it ends up inheriting from TestCase (Twisted)...... just a sec..... let me copy the complete hierarchy | 19:46 |
| gatox | dobey, from ubuntuone.controlpanel.gui.qt.tests import BaseTestCase -> from ubuntuone.controlpanel.tests import TestCase -> from ubuntuone.devtools.testcases import TestCase -> from twisted.trial.unittest import TestCase | 19:50 |
| dobey | gatox: ok, this is the other issue then, in dev-tools | 19:51 |
| dobey | oh hey, my raid is ready to test/use now | 19:54 |
| dobey | and i need to call a bank, brb | 19:54 |
| dobey | sigh | 20:01 |
| dobey | well, at least my script is working sort of ok | 20:03 |
| dobey | Closed 9 bugs; 52 remain Incomplete. | 20:03 |
| ralsina | dobey: I imagined there was a reason fr the shuffling of low 6-figure bugs | 20:04 |
| dobey | heh | 20:04 |
| dobey | ralsina: i got tired of closing the incompletes by hand, so i wrote a script. needs a little tweaking, but working good so far now :) | 20:05 |
| mmcc | gatox, I just remembered that you asked yesterday to see my sync_menu mac/win fix branch. Here it is: https://code.launchpad.net/~mikemc/ubuntuone-client/fix-dummy-sync-menu/+merge/126766 | 20:24 |
| gatox | mmcc, ack...... reviewing that | 20:24 |
| gatox | mmcc, +1 from me | 20:25 |
| mmcc | great, thanks gatox | 20:25 |
| gatox | mmcc, thx for catching that :P | 20:26 |
| mmcc | gatox: no prob. it was pretty easy to track down :) | 20:26 |
| briancurtin | arghhh, trunk tests pass on jenkins but not on my machine. | 20:41 |
| gatox | 5:40 already..... eod here ...... i need to kill some aliens with the xbox to relax | 20:42 |
| gatox | bye people! see you tomorrow! | 20:42 |
| dobey | sigh. wonder what caused that :( | 20:58 |
| === m_conley is now known as m_conley_away | ||
| mmcc | late lunchtime | 21:13 |
| dobey | later all | 21:25 |
| briancurtin | see ya | 21:25 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!