[00:00] rakshasa: You are behind a proxy? [00:00] Yes but never had an issue with it [00:01] unfortunately Ubuntu One currently does not work behind a proxy [00:01] I'll try some stuff myself and If it works I will close ticket [00:02] But gotta go now [00:02] Tnx for the support [00:02] ok, cheers [00:02] anytime === eu is now known as Guest76874 === Guest76874 is now known as adorilson [03:08] I am logging into a remote machine to see why my ubuntu one does not seem to be syncing up properly on it. It did a u1sdtool -s and I get State Ready, not online and connection: Not User With Network. What is wrong? [03:49] I am having a hell of a time getting my one machine to sync up properly. Anyone around to offer help? [03:49] ls [03:51] It just won't seem to sync up [03:52] It is running 11.10, is this a problem? [03:54] Anyone here? [03:58] duanedesign, are you around? [04:00] Is there a naming convention that will not be shared? [04:02] Hello? [04:04] This is a paid service and this is the best help available? [04:07] Is there a secret door for paid support???? [04:08] Wow, asking a question about Ubuntu One gets completely ignored on #ubuntu and also here....strange [04:09] keithclark, I'm sorry, we don't have 24hs support [04:09] the core hours are specified in our support page [04:10] from what I can see, U1 thinks you don't have a network [04:10] I am not sure what it could be, the service is working fine [04:10] beuno, so if paid for, only limited hours support is offered? [04:10] keithclark, yes, like 99% of all services, they don't have 24/7 support [04:11] we cover a huge range of time, but there's an 8 hour gap [04:11] there should be someone around in about 5 hours [04:11] in the mean time, I'd suggest filing a bug with your logs [04:11] they will pick that up then [04:11] I don't have 5 hours [04:11] or email the support address [04:13] keithclark, this is a stock Ubuntu 11.10? [04:14] beuno, yes, being accessed remotely via ssh [04:14] keithclark, and "u1sdtools -c" doesn't connect it? [04:15] it gets the connection info from Network Manager [04:15] connects, doesn't sync [04:15] maybe that it isn't enabled if you're accessing remotely? [04:15] 21:08 < keithclark> I am logging into a remote machine to see why my ubuntu one does not seem to be syncing up properly on it. It did a u1sdtool -s [04:15] and I get State Ready, not online and connection: Not User With Network. What is wrong? [04:15] Yes, same directory synced fine earlier [04:15] I guess that's old [04:16] beuno, yes, that cleared itself up [04:16] so, what do the logs say? [04:16] ~/.cache/ubuntuone/log/syncdaemon.log [04:17] alternatively, you can try disconnecting and reconnecting with -q and then -c [04:17] beuno, done that 5 times [04:17] I see this over and over: [04:17] 2011-11-08 23:16:58,699 - ubuntuone.SyncDaemon.Main - NOTE - ---- MARK (state: ; queue: 0; hash: 0) ---- [04:18] keithclark, that means everything is up to date [04:18] as far as Ubuntu One can tell [04:18] Well it is not [04:19] keithclark: when you say it is not, what is not synced? [04:19] jdobrien, a file in a directory that says on the originating machine that is synced [04:19] It does not show up on the 'to be synced machine' [04:20] keithclark, does it show up on the web ui? [04:20] keithclark: so so I understand what you mean, you have two computers right? [04:20] beuno, oh god, I hate that interface [04:20] jdobrien, yes [04:20] keithclark: what platform are these? [04:20] keithclark, it's ok, you don't have to stay there, just check so we can confirm it's on the server [04:20] windows or ubuntu [04:21] jdobrien, one is 11.04, the sender, the other is 11.10, the receiver [04:21] keithclark: ok. Is the directory under ~/Ubuntu One? [04:21] jdobrien, no [04:22] keithclark: so you have folder on one machine you want to sync [04:22] keithclark: on the machine that is not syncing the folder, can you run u1sdtool --list-folders [04:24] yes [04:24] keithclark: we're looking for the folder you want to sync [04:24] jdobrien, and the target folder is there [04:25] keithclark: what is the value of 'subscribed' [04:25] /home/keithclark/Docments [04:25] id=1e643d9e-4ea8-40de-a502-596b9823dfb7 subscribed=True path=/home/keithclark/Documents [04:25] id=fca9b029-1935-45f6-8dc6-dfd039ae0723 subscribed=False path=/home/keithclark/Waterloo Hosting [04:25] id=a0fbf137-3d14-45c2-a2ed-28d61e6f90f4 subscribed=True path=/home/keithclark/Videos From Home [04:25] keithclark: so the folder is not subscribed (aka syncing) yet [04:26] ? [04:26] keithclark: you can subscribe it in the control panel [04:26] keithclark: Ubuntu One doesn't by default sync all the folders on all the computers [04:27] keithclark: u1sdtool --subscribe-folder 1e643d9e-4ea8-40de-a502-596b9823dfb7 [04:27] But /Documents is synced. It has synced some of the files in that folder in the past [04:27] keithclark: for some reason it is not anymore :( [04:27] I understand that and that is why I'm here! [04:27] keithclark: oh sorry [04:27] it is True... [04:27] it's a little late for me I guess :) [04:28] hmm [04:28] keithclark: so no we need to make sure the other computer is actually pushing changes to ubuntu one [04:28] s/no/now [04:29] keithclark: can you run that same command on the other computer? [04:33] Working...... [04:36] You know what? This is all too hard to do. I'll try DropBox to see if that is easier [04:36] ok [04:37] thank you for trying [04:37] No problem (you might want to consider ease of use versus revenue) Just a hint [04:39] (and customer service as well) [04:39] i thought that's what i was doing ;) [04:39] jdobrien, you were! [04:40] jdobrien, but it took me a ton of effort to get here. [04:41] And still my machine is not synced [04:41] yeah it's strange it's not working for you [04:42] dropbox has an advantage over use in filesync (for now) since that's all they do. they have a very nice service [04:43] I don't mean to threaten. It is just that Ubuntu One wants me to pay for a service that does not work.....strange. [04:44] keithclark: i didn't take it as threatening. we know we have some bumps right now [04:45] Well pull back cash requests for properly delivery. Just an observation. [04:45] keithclark: we have 110 million free users [04:46] I am an example of a customer not served. [04:46] oh wait...let me check my decimal point [04:46] 1.5 million [04:46] free users [04:47] OK, are they satisfied? Are they daily users? [04:47] Do they depend on this service? [04:47] keithclark: there's some who aren't...the majority are satisfied i think [04:48] it's ok to fall into both categories [04:48] I will never be satisfied until it all 'just works' [04:48] Well, I'm a believer and I guess I'll have to wait for my solution [04:49] it typically does work. I have have several computers that sync fine under normal use [04:49] although I do know the weaknesses of our systems and can force failures ;) [04:50] I really just wanted help when asked. That's all. [04:51] Although, still not synced! [04:51] keithclark: we really appreciate you spending some time here and explaining what's happening, it's really the only way to work out the kinks [04:52] unlike locked-down desktops, there are so many nuances to open-source computers that create interesting challenges [04:52] I'm in the middle of restoring a system I pushed beyond what I should have :) [04:53] Yup and I've been here since '07, still learning [04:55] keithclark: I've been able to watch Ubuntu One go from scribbles on napkins to something people use. So I do have a sense of ownership, although I don't make all the decision for the choices we make, there is still a sense of pride in wanting it to suceed [04:56] Well I'm an example of where it does not work. Have fun! [04:56] keithclark: :) [04:57] Well, I'm headed for bed! Contact me at keithclark at watersubstop dot com [04:58] keithclark: all I ask is don't forget about us. I hope dropbox works out ok [04:58] no reason it shouldn't [04:58] waterloosubstop dot com [04:58] keithclark: ok [04:58] thank you... gnight [09:35] Wonderful Wednesday, people! :D [10:14] hi [10:32] is UbuntuOne always slow? I've been syncing up for over an hour now with no sign of finishing, and this is all stuff that I synced yesterday . . . . [10:33] feels really broken, and I have no feedback on what it's doing [10:34] m4r35n357: We've got someone looking into it for you. [10:34] I have the client and the home folder open on that same screen, and the "updating" emblems are on the directories, but I feels could have typed in the data quicker by hand [10:34] ah JamesTait, please forgive my bleatings, but this is doing my head in ;) [10:35] m4r35n357: No problem. :) [10:36] I have a lapton on Ubuntu Natty, and two netbooks, one with Lubuntu Oneiric, and one with the LTS netbook remix, 10.04 I think [10:37] I am trying to do a test where I share some files in my home dir on the laptop, and see what happens with the other machines [10:38] m4r35n357, we are currently investigating the slow performance, at the moment I don't have any more info [10:38] m4r35n357: I can tell you what *should* happen. :) [10:38] And the things were really fast for me a couple a hours ago [10:38] I've been trying this since yesterday morning, last thin last night I had to abandon the attempt, and have restarted with just the laptop this morning [10:39] rye: They were really fast for me just a few minutes ago, when my wife shared her course notes with me so I can print them. :) [10:39] JamesTait, I suspect some api slaves are loose [10:39] * JamesTait nods [10:39] but the client on the laptop had lost the tick marks on the selected dirs, and the emblems had gone took, so I re-selected the dirs and have been wating ever since . . . . [10:40] rye: Losing the tick mark emblems s in Nautilus - could that be because there's no server connection, so the Nautilus plugin can't verify the file status? [10:40] JamesTait, I'm suspicious of the 10.04 client, but can't confirm anything until I get some results [10:41] m4r35n357, you can find out the current connection status with "u1sdtool --status" command [10:41] JamesTait, no Idea I'm sure ;) I'm a first-time user ATM . . . [10:41] of U1 that is [10:41] rye, "bus error" [10:42] the clien telss me there is a sync in progress . . === eu is now known as Guest43344 [10:43] Ironically, my previously perfect connection has now disconnected. ;) [10:43] I'm thinking about disconnecting and reconnecting, but not sure if it will just make things worse . . . [10:43] m4r35n357, is status QUEUE_MANAGER ? [10:43] rye, any comment on the bus error? [10:44] no, "bus error" [10:44] did I remember to say "bus error"? [10:46] m4r35n357, well, bus error can be triggered by illegal instruction in the code, given the code is python this looks like something deeper, like dbus [10:46] JamesTait, it's coming up to two hours now, so think I'll try reconnecting . . . [10:46] m4r35n357, yes, please try reconnecting, and try quitting syncdaemon completely in order to reset it completely [10:47] the weird thing is, all the files seem to be on the server, so WTF is U1 actually doing, I presume all files are checksummed? === Guest43344 is now known as adorilson [10:50] well I've now restarted the client, and file sync is now disconnected, reconnecting manually as it's not happening automatically . . . [10:50] and now I'm back to "bus error" [10:50] but the client telss me there is a file sync in progress . . . [10:51] m4r35n357: There's a problem at the moment with some of our servers not accepting new connections. We're looking into it. [10:51] I tend to believe u1sdtool [10:52] With a big hammer. [10:53] JamesTait, dunno if it's just me, but I've shied away from U1 for some time because of this kind of experience. The diffference on this occasion is that I have persevered [10:54] are there many non-hardcore users on the system? I would consider anyone (granny?) who gets all this stuff working first time to be extremely fortunate [10:56] m4r35n357: Your perserverence is appreciated. I don't think we have any kind of information that would tell us if our users are "hardcore" or "granny" users. :) [10:56] JamesTait, how will I know when to try again? [10:58] m4r35n357: If you have Identica, Twitter or Facebook accounts, we usually put notes out there for severe service disruptions. We also have a status page here: https://one.ubuntu.com/help/faq/where-can-i-check-the-status-of-ubuntu-one-services/ [10:58] m4r35n357: Nope, sorry, here: https://wiki.ubuntu.com/UbuntuOne/Status [10:58] JamesTait, yeah thanks I've seen the status page, says file sync is fine . . . . [11:01] now getting an Oops mesage from u1sdtool, and GUI client won't launch [11:03] Oops, an error ocurred: [11:03] Traceback (most recent call last): [11:03] Failure: dbus.exceptions.DBusException: org.freedesktop.DBus.Error.Spawn.ChildSignaled: Process /usr/lib/ubuntuone-client/ubuntuone-syncdaemon received signal 7 [11:03] but my disk light is flickering . . [11:04] m4r35n357: Right, I might know what that is. [11:05] m4r35n357: Bear in mind I'm not that familiar with the client side; I think it happens when it's performing a local rescan and the daemon process is too busy processing that to handle the dbus call. [11:06] ok the flickering has stopped and u1sdtool is now responding [11:08] unfortunately I now have two U1 entries under the mail icon, and neither of them shows the client, reckon my GUI is screwed now [11:09] unless you can suggest otherwise, I think it's reboot time . . . [11:10] m4r35n357: A logout/login might be equally effective in that case. [11:10] what is signal 7??? [11:10] ah, SIGBUS [11:10] JamesTait, gotcha, I'll see if I can find something to kill first . . . . [11:11] rye: Beat me to it. :) [11:11] rye, would that be "bus error" ? ;) [11:11] not listed in kill(1) [11:11] m4r35n357, yes, exactly [11:11] brb [11:12] m4r35n357, i think that having a pythin structure misaligned in memory is quite hard, so that's c extension or something. In case it does not start at all with dbus activation I'd suggest restart dbus session, which in our case is logging out and in [11:13] rye, well I'd like to try killing one or both of my stuck clients first, do you know the program name for the client? [11:13] before I destroy the evidence, as it were [11:15] hmm looks like at least one is a zombie, so that's reboot time . . . [11:16] going down . . . [11:21] hmm, that reboot didn't go smoothly, and now I have threee Ubuntu one entries under the mail menu . . [11:22] and now it's down to two, and one of them has a triangle to the left, does that mean it's the "active" one? [11:23] and in the file manager, two of my synced directories have no emblem . . . [11:24] * JamesTait really needs to learn more about this side of things. [11:25] well I've really tried hard this time, but now my computer is not rebooting cleanly and the last login took ages [11:25] and I don't know how to fix my mail menu . . . and I still don't have synchronised files [11:26] BTW I've been using Linux since the early nineties, and have worked as system admin and programmer, so I'm not a newbie at this stuff [11:27] but if I had a hosted server, I could have had this done & dusted in about twenty minutes using SSH . . . [11:27] so I don't know what to think at this point . . . [11:31] think I'll reboot again . . . [11:33] m4r35n357: Sorry, remind me, which version of Ubuntu is this machine running? [11:35] this is Natty, I tried Oneiric yesterday, but changed my mind and restored today [11:35] m4r35n357: And are you using standard repositories, no PPAs? [11:36] yep, bog standard [11:36] m4r35n357: OK, thanks. [11:36] well the last reboot went cleanly [11:36] and I only have one client entry in the mail menu, and I'm trying to look at stuff uning u1sdtool [11:37] I don't have a Natty machine to test locally, but I might be able to run up a VM. [11:37] ah dammit, now I have two again! WTF? [11:37] maybe this is how it's supposed to work ? [11:38] but now I'm getting bus errors all over the place again [11:43] m4r35n357: Could you confirm which version of ubuntuone-client you have installed please? [11:43] m4r35n357: dpkg -l ubuntuone-client [11:46] hello everyone! [11:46] nessita, hi...... saliendo para alla! [11:46] gatox: ack! [11:47] JamesTait, ii ubuntuone-clie 1.6.2-0ubuntu2 Ubuntu One client [11:48] m4r35n357: Thanks. [11:53] still stuck here, getting bus errors from u1sdtool, and still two U1 entries under the mail menu, and no HD light flickering, and client won't launch . . . [11:54] perhaps it's just busy, who knows? [11:56] m4r35n357: One thing I hadn't realised is that the syncdaemon code is now partly implemented in a C library, rather than all Python. There's an ubuntuone-client-dbg package that has debug symbols which might provide more clues when those bus errors occur. [11:57] rye: What do you think? Is it worth installing ubuntuone-client-dbg? [11:58] rebooting again . . . [11:59] JamesTait, no, that's debug symbols for gnome extension [11:59] Hm, OK. [12:01] aquarius, why do you hate gtk treeviews? anything I can do to help? [12:02] rodrigo_, because what I want to do is say: here is a python list. Display it. What I actually have to do is create a treeview, a treestore, a treeviewcolumn, a cellrenderer, and stitch them all together. I know *how* to do it, I just hate having to do it :) [12:02] JamesTait, and it seems that IS found the reason of partial filesync outage [12:03] aquarius: I know the pain [12:03] aquarius, ah, the columns thing indeed is not good, I think you can create it all in glade [12:03] aquarius, and in the code just fill the model [12:04] rodrigo_, yeah, I can, but this little program is too stupid to need a glade file :) [12:04] ah ok :) [12:04] rye: Indeed. So server-side we should be back in business. Not sure about these strange client problems though. [12:06] JamesTait, did some digging before the last reboot, found my /tmp dir was full! I suspect that was the cause of the bus errors [12:06] but that was from a clean reboot, so I expect it to happen again [12:06] m4r35n357: That's worth noting, good catch! [12:06] U1 was the only thing writing to /tmp, which is a 1.5 SHM partition [12:07] 1.5GB I mean [12:07] in fact it's already up to 1.3GB now, without me doing anything [12:07] . . . . and now it's full again [12:08] and I get a bus error [12:08] so this is all automatic, not sure how to avoid it [12:09] none of these figures exceed the U1 specs as far as I know [12:11] m4r35n357: I don't know enough about the client to understand what might be doing that, but I'm trying to get hold of someone who will. [12:11] now, I would like to start over with the main directories, but I can't stop them synchronising once the client is crippled like this [12:11] unless I reboot and get in really quickly . . . don't like that idea [12:12] m4r35n357, /tmp full of what? [12:12] data files, presumably from u1 but I can't prove it [12:12] m4r35n357, with "bus errors" you mean dbus? [12:13] m4r35n357, give me one example of file name [12:13] there in /tmp, I mean [12:13] facundobatista, u1sdtool gets killed with SIGBUS [12:14] facundobatista, tmpRoAw4D [12:15] m4r35n357, yes, that can be syncdaemon... or any python process using the standard library tempfile call [12:15] m4r35n357, can you kill syncdaemon and see if /tmp keeps growing? [12:15] I've managed to stop the dirs updating so I think I'll reboot again and see if it happens again [12:15] facundobatista, it's already full, happens in a few seconds [12:16] facundobatista: Background info on "bus error": http://irclogs.ubuntu.com/2011/11/09/%23ubuntuone.html#t10:41 [12:17] facundobatista: Actually this might be more useful: http://irclogs.ubuntu.com/2011/11/09/%23ubuntuone.html#t11:03 [12:18] JamesTait, maybe dbus fails because /tmp is full? [12:19] well the reboot was not smooth either [12:19] facundobatista: Seems likely, yes. [12:19] m4r35n357, did you kill syncdaemon? [12:19] the cleint process was still open when I rebooted [12:19] m4r35n357, are you syncing some file that is around 1.5Gb ? at the moment? [12:19] and I had to use magic sysrq to finish the boot [12:20] rye, no, I would have thought of that ;) [12:20] m4r35n357, can you please kill syncdaemon and let us know if /tmp filling stopped or continues? [12:22] m4r35n357, any 5 files that will give this amount? The uploads are happening in parallel [12:22] yeah I've discovered some files that should not have been there, now deleted [12:23] I'm just doing the smaller directory ATM [12:23] I think I deleted the big files while I was testing Oneiric yesterday, and when I restored of course they were still there [12:24] none of this is over my limits though [12:24] and none of the files is itself big enough to fill /tmp [12:24] m4r35n357, as rye said, uploads are done in parallel, 10 at a time, so 10 files could fill your /tmp if its slow [12:25] but if the whole dir is being processed then /tmp will not be big enough [12:25] m4r35n357, you can change that in the config [12:25] m4r35n357, sorry, not if your /tmp is slow, but if its small [12:25] facundobatista, didn't know there was any config [12:25] m4r35n357, yes, there is [12:26] facundobatista, OK, any clues? ;) [12:27] m4r35n357, check /etc/xdg/ubuntuone/syncdaemon.conf for global setttings, and you can use any of those in your ~/.config/ubuntuone/syncdaemon.conf if you want to change something specifically for any user [12:27] mandel: ping [12:27] I normally touch the config for me only [12:27] m4r35n357, the config for the simultaneous uploads is "simult_transfers" [12:28] (which actually affects downloads too) [12:28] facundobatista, OK thanks but I think I'll work with the defaults ;) [12:29] tmp folder is well behaved now, so once this is all synced up I can fire up the netbook and see how it all works . . [12:31] OK so is there any way this /tmp full condition can be checked in the client? [12:32] looks like it's just chucking stuff in there without looking [12:32] then the process blocks [12:33] in such a way that it can't be killed, u1sdtool shows bus errors, and rebooting is also blocked [12:34] m4r35n357: Sounds like a subject for a bug report. [12:35] m4r35n357, the problem we have in the client is that if the compressing fails, we leave the half compressed file there [12:35] actually taking space [12:35] the fix for this is actually already coded, will be in trunk today or tomorrow [12:35] nessita, ^ [12:36] Even better! :) [12:36] OK so some good will come of all this! [12:37] now I can experiment and find out how ubuntuone actually works ;) Thanks everyone, hopefully this is the end of my whining! [12:38] hi folks. My u1 installed in my android never sent my pictures automatically, yet it is set to do it. What can I do? [12:39] m4r35n357: Thanks for your patience through this, and feel free to drop in again if you continue to have problems. [12:40] afeijo: hi! [12:40] afeijo: What phone model do you have? :) [12:40] motorola droid 1 [12:40] android 2.2 [12:40] afeijo: Ah, right. Do you have any file browser? [12:40] ok and thanks for you guys for your patience ;) I might have a question or two, but I think this is the main issue I've been seeing [12:40] yes [12:40] facundobatista: I need the review!!! [12:40] in particular, I don't expect the 10.04 client to play well . . . [12:40] facundobatista: https://code.launchpad.net/~nataliabidart/ubuntuone-client/compress-flush-to-file/+merge/81595 [12:41] afeijo: Could you please tell me if your picture path is /mnt/sdcard/DCIM or /mnt/sdcard/dcim ? [12:41] nessita, I know, I have it queued [12:46] facundobatista: thanks! [12:46] karni, sorry the delay, I confused my devices. I have a file browser in my Xoom, I just installed in my Droid [12:46] afeijo: no probs, perfect. [12:46] karni, my pics are at /mnt/sdcard/dcim/Camera [12:47] afeijo: so, dcim is lowercase, right? [12:47] right [12:48] afeijo: It's an Ubuntu One Files bug, it has been fixed in the upcoming update. [12:48] karni, how long until that update will be released? [12:48] afeijo: I hope to release it shortly, perhaps this week. [12:48] great [12:48] I'm waiting for it for months hehe [12:49] karni, I visited u1 website and noticed it now has a windows version! tonight I'll test it at home :) [12:49] afeijo: Sorry, I'll put more focus on frequent updates. [12:49] afeijo: Enjoy! :) We have a team of great engineers here working on the Windows version, correct! [12:49] karni, this week is great, np [12:49] nessita: package arrived yesterday [12:50] haven't opened yet, will do so now [12:50] thisfred: WOW, thanks! [12:50] thisfred: is it too big? (drop the boxes, of course) [12:50] thisfred, tape the unboxing hehehe [12:50] videotape [12:50] nessita: without the boxes it will be fine. 2 controllers: check [12:50] thisfred: awesome [12:51] karni, do you have a tablet? [12:52] nessita: oh, and so you know: Anthony gave me the kindle cover [12:52] thisfred: double awesome! [12:55] afeijo: I have a tablet. [12:56] karni, I got my 1st one last week, loving it... so much to figure out, I don't know the best apps yet, etc [12:56] afeijo: I use my Adam for reading, really :) [12:56] I tested the kindle app, very nice [12:57] afeijo: I use Aldiko :) [13:06] hello #ubuntuone! hello gals and boys! [13:14] hello alecu! [13:14] alecu: how is it going? [13:14] alecu, hi! [13:14] nessita, great! [13:15] hi there gatox, I'll do your review today [13:15] alecu, you read my mind :P [13:20] alecu, nessita, gatox I forgot to tell you that today I have a day off, is a hliday in Madrid :P [13:21] sorry.. [13:21] mandel: but... is it a national holiday? [13:21] mandel: I googled and did not find it [13:23] nessita, here is the laboral calendar http://www.seg-social.es/Internet_1/Masinformacion/CalendarioLaboral/Fiestas/index.htm?Ejercicio=2011&prov=28&loc=280790000 [13:24] mandel: if I understand that properly, today is not a national holiday but a local holiday? [13:24] nessita, yes, we can take those, right? [13:24] nessita, cause if we cant I have been doing this all year.. [13:24] mandel: from my understanding, no, but we should confirm [13:24] nessita, ups! yes, let me ask hr [13:31] me [13:32] me [13:33] alecu, mandel ? [13:33] me [13:34] DONE: bug #887150, reviews, bug triage, teaching duties [13:34] TODO: start digging into packaging qt4reactor, start moving QT bits from installer to sso [13:34] BLOCKED: nopes [13:34] Launchpad bug 887150 in ubuntuone-client "Windows: test failure at tests.syncdaemon.test_action_queue.TestZipQueue.test_compress_gets_compressed_data (affects: 1) (heat: 6)" [High,In progress] https://launchpad.net/bugs/887150 [13:34] NEXT: gatox [13:34] DONE: [13:34] Fix some branches (removing mixed changes). Work on unicode issues. [13:34] TODO: [13:34] Keep fixing unicode issues. [13:34] BLOCKED: [13:35] No [13:35] alecu, go [13:35] DONE: fight with qtnetwork + qtreactor [13:35] TODO: backport server time sync [13:35] BLOCKED: no [13:36] alecu: why fighting? is not working? :-/ [13:37] nessita, qtreactor seems to be dirty and won't shut down properly when using it with qtnetwork [13:37] alecu: what a ... pain [13:37] nessita, so a few DelayedCalls end up as errors in the tests runs I do with qtnetwork [13:37] alecu: so, you're debugging qt4reactor, basically? :-/ [13:37] nessita, not yet [13:38] nessita, I thought the problem was with my mockers, so I started doing some tests with libsoup, but I run into issues like this: https://bugs.launchpad.net/ubuntuone-control-panel/+bug/842614 [13:38] Launchpad bug 842614 in ubuntuone-control-panel "Nightlies build failures in maverick and oneiric (affects: 1) (heat: 16)" [Medium,Triaged] [13:38] and this: http://mail.gnome.org/archives/nautilus-list/2011-October/msg00007.html [13:38] looking [13:39] nessita, is that the problem that dobey mentioned about GI + twisted? [13:39] alecu: ok, that is an issue when mixing static and dynamic gobject binding [13:39] nessita, right. But it seems that the twisted glib reactor is using the static bindings [13:39] me [13:40] TODO: fin out if I'm on holidays [13:40] alecu: yes, known bug... dobey will have that fix as one the first things when he comes back, but may take a week [13:40] meh, I'll continue when is sorted out === adorilson_ is now known as adorilson [13:45] ok, mandel is oficially on holiday today (btw, did you add the request in the admin?) [13:45] nessita, yes :) [13:47] nessita, it would be nice to explain gatox and alecu the confusion so that they take their holidays correctly :) [13:47] mandel: yeah, I'm telling them in our private .ar channel ;-) [13:47] mandel, we don't have "local" holidays, so don't worry about that :-) [13:48] nessita, oh, the secret channel.. hehe [13:48] alecu, yo know, lazy spaniards :P [13:53] alecu: so, going back to the glib issue. You're trying to run tests in the controlpanel with qtnetwork? [13:55] alecu: or just using trial + qtnetwrok? [13:56] nessita, in sso, with runtests -qt (using qtnetwork). [13:56] nessita, let me upload a branch. [13:56] alecu: right... so trial will not work with dynamic glib :-/ [13:57] I'm wonder what thing is importing the dynamic bindings [13:57] nessita, well, I am importing the dynamic bindings when trying to use libsoup. [13:59] nessita, here's the code so far: https://code.launchpad.net/~alecu/ubuntu-sso-client/proxy-web-client/ [13:59] looking [14:01] nessita, and here's the error when running u1trial with the qt reactor: http://pastebin.ubuntu.com/733089/ [14:01] I'm running it with: u1trial --reactor=qt4 --gui ubuntu_sso/utils/tests/test_webclient.py [14:02] alecu: I have the same output, looking a bit more [14:04] alecu: a tiny note, the tests for webclient should be inside the webclient folder [14:04] noted [14:06] hi [14:06] alecu: so, what of you fix all the "valid" errors? you still get the dirty reactor ones? [14:06] alecu: I [14:06] oops [14:06] alecu: I'm just running tests for get_url like this: [14:06] u1trial --reactor=qt4 ubuntu_sso/utils/tests/test_webclient.py -t WebClientTestCase.test_get_url [14:07] and I'm having an AttributeError [14:07] nessita, yes, I comment those other tests, but the first one is still unclean when it tears down, hence the error. [14:07] ah [14:07] debugging a bit more... [14:07] I have a scenario that breaks ubuntu one [14:08] alecu: out of the box questions: don't you need to shutdown the mockwebservice? [14:08] ah, you do, sorry [14:08] Beret, please tell us [14:08] nessita, yes, on addCleanup [14:09] yes yes [14:09] I had a machine that was syncing ~/Music - I then did a fresh install on that machine and chose not to sync ~/Music from U1. I logged onto the U1 web site and turned off syncing of ~/Music. I then rebuilt the ~/Music folder from scratch on the new machine (cleaning things up) and then right-clicked, u1, sync this folder on the machine. [14:09] Turns out, the machine says it's fully synced [14:09] the folder is empty on the u1 site [14:09] so it's thoroughly confused [14:10] my machine thinks it's synced when it isn't, and there are no files on the website [14:11] u1sdtool --list-folders is showing only one Music folder shared [14:11] Beret: what do you mean with "onto the U1 web site and turned off syncing of ~/Music"? [14:12] login to one.ubuntu.com, go to files, click "More" under the folder and hit "Stop syncing this folder" [14:12] rye, duanedesign: Beret is describing above a complex synchronization issue. Have you guys seen anything like that? [14:12] nessita, I don't see a "delete" anywhere, so that's the best I could do [14:12] alecu: so, running the suite for WebClientTestCase with all the tests commented out but setUp as given will not give dirty reactor [14:12] my intent was to nuke hte files on u1 [14:12] so I could resync the good folder [14:12] Beret: right, so, the ~/Music folder no longer is listed on the web site, then? [14:13] it wasn't no [14:13] it is now of course [14:13] but it's empty and my client isn't uploading files as it should [14:13] it's showing no files on one.u.c [14:13] Beret, are you using windows or Ubuntu ? [14:13] Beret: right, can you please restart the client? u1sdtool -q && u1sdtool -c [14:13] nessita, right. And if you enable the first test, and in "ubuntu_sso/utils/webclient/__init__.py" you uncomment the line 30 and comment the 31, it works fine [14:13] rye, smack [14:13] rye, j/k oneiric [14:13] nessita, this is UDF [14:13] ah [14:14] UDF filtering [14:14] nessita, (using the twisted web client instead of qtnetwork) [14:14] Beret, ok, then this is most likely bug #869920 [14:14] rye: yeah, my thought exactly [14:14] Launchpad bug 869920 in ubuntuone-client (Ubuntu) (and 2 other projects) "Files in new UDFs are not uploaded due to filtering (affects: 1) (heat: 6)" [High,New] https://launchpad.net/bugs/869920 [14:14] grrrr [14:14] nessita, can we get SRU for that? [14:15] rye: you will need to chase facundobatista for that one :-/ [14:15] rye, so I should be able to restart u1 services and it'll sync? [14:15] rye: I'm happy to help making reviews and all, of course [14:15] facundobatista: any ETA on the localrescan/UDF unsubcribe bug? [14:15] alecu: confirmed, looking a bit more [14:15] nessita, not at all [14:16] facundobatista: but there is no plan to have that done, eventually? [14:17] alecu: so... I added a "return defer.succeed('test')" in the qtnetwork request method [14:17] and that makes the test work ok (it fails due to assertion error, but no dirty reactor) [14:17] nessita, yes! eventually [14:18] alecu: so I'm wondering if we don't need to close/shutdown/something else the QNetworkRequest [14:20] facundobatista: but do you have an estimation of when? this month? next? next year? [14:20] :-) [14:20] nessita, that sounds possible... but the delayed call traceback suggests that the problem is not on the client, but on the server: [14:20] nessita, no, I don't [14:20] File "/usr/lib/python2.7/dist-packages/twisted/web/server.py", line 219, in render [14:20] (the mock server) [14:20] Hmmmm... [14:20] alecu: more "guessing": perhaps the request is still "connected" to th server [14:20] I may try having it download some other url [14:21] alecu: so the server can't shutdown because it has pending stuff "in it" [14:23] nessita, I'll take a look at that. [14:23] alecu: another question, shouldn't this self.nam.finished.connect(reply_finished) be reply.finished.connect(reply_finished) ? [14:23] nessita, anyway, I think we should be working on this later, since today I'm supposed to backport the timesync stuff. [14:24] nessita, I tried that too. According to the docs both signals are sent at the same time, the first one with reply as the only arg, the second one with no args. [14:25] alecu: ah, ok. And ack to the backport, I can give any help you need about it [14:33] nessita, I think I might have figured it out [14:33] nessita, I've straced the test run, and the socket is not being closed [14:33] nessita, so I looked into the data stream, and it's HTTP/1.1 [14:34] nessita, that means that more than one http request may be done on the same connection [14:34] nessita, so the networkaccessmanager keeps the connection open, so it can do another http request to the same server [14:35] so, it's a good thing, because the webclients would be faster. But it's annoying for the tests :-) [14:36] alecu: great news! (in a average) [14:57] Hi, I can't get my U1 to upload a file, it's a backup from evolution, 497.3 MB. If I kill the daemon and then start it again it starts but then it just stops after a while and won't continue [14:58] well the daemon doesn't stop but the transfer stops and daemon is idling [14:58] Is the file to big to upload? [15:07] nessita, I added just one line to the webservermock that fixed this issue! [15:07] request.setHeader("connection", "close") [15:08] alecu: YOU ROCK BIG TIME [15:08] nessita, and now, I'll put this in the freezer and get porting. [15:08] alecu: YOU ROCK EVEN MORE [15:08] :-) [15:08] * alecu headbangs [15:08] * alecu misses his very long hair. It's been... 10 years! === m_conley_away is now known as m_conley [15:50] gatox, approved. It was a really small branch without the other branch merged in!!!! [15:50] alecu, thanks! === zyga is now known as zyga-afk === nessita1 is now known as nessita [16:16] Hello? [16:21] Anyone for a little support? [16:24] A horse a horse my kingdom for a horse [16:25] hello rakshasa [16:25] Hello again [16:26] Did you recieve my mail [16:26] i did. I just responded :) [16:26] Good cause I'v traced it back to the DBCouch not willing to start [16:26] Destop couch [16:27] http://paste.ubuntu.com/733229/ [16:28] But i'm stuck there [16:31] rakshasa: could you please open a terminal and ttry these commands [16:31] killall beam.smp; killall beam [16:32] then: rm ~/.config/desktop-couch/desktop-couchdb.ini [16:32] and: dbus-send --session --dest=org.desktopcouch.CouchDB --print-reply --type=method_call / org.desktopcouch.CouchDB.getPort [16:34] put in your username: xdg-open file:///home/YOURUSERNAME/.local/share/desktop-couch/couchdb.html [16:34] your username on your computer [16:34] the last one will confirm whether or not desktopcouch is running [16:35] if the couch interface opens it is running [16:35] Still didn't start yet [16:35] http://paste.ubuntu.com/733235/ [16:36] And the local html said the same that it hasnt started yet [16:36] rakshasa: ping [16:36] oops [16:36] rye: ping [16:37] mornign [16:37] oops [16:37] duanedesign, pong [16:37] goooood morning rye [16:37] rye: http://paste.ubuntu.com/733235/ [16:38] rakshasa, hi, could you please pastebin the contents of ~/.cache/desktop-couch/desktop-couchdb.log.1 ? === beuno is now known as beuno-lunch [16:40] http://paste.ubuntu.com/733240/ [16:42] rakshasa, ok, nothing peculiar in the log, looks like it was working (I suppose 10 minutes ago), is there any beam.smp process? [16:42] or beam [16:44] No no processes with beam.smp or beam where killed [16:44] no process found [16:44] rakshasa, ok, let's start directly - /usr/lib/desktop-couch/desktopcouch-service (in a separate terminal, it will hold it) [16:46] same prob http://paste.ubuntu.com/733251/ [16:47] rakshasa: oh wait, you are behind a proxy? [16:48] Yes but that has never been an issue before [16:49] to my knowledge db couch connects on 127 local and previously that always worked [16:49] I did however see a keyring error in a evolution mail log [16:49] If that helps? [16:50] But iv been unable to reproduce the errer [16:50] *error [16:51] rakshasa, are there any desktopcouch-service processes running? [16:52] rakshasa, you are right, proxy does not make it fail, something else does [16:52] i wonder whether couchdb-bin package is actually installed [16:52] Yeah couch has always worked for me [16:52] rakshasa: have you followed these steps before: https://one.ubuntu.com/help/tutorial/contact-sync-for-ubuntu-1004-lts/ [16:53] Not yet I will do them right now [16:55] Installing 1 of 52... [16:56] I will restart my os after its done [16:58] rakshasa, hm, does it install couchdb-bin as part of upgrade? [17:02] Back [17:02] Succes [17:02] nice [17:03] My ubuntu couch seems to be running [17:03] So an out of date package was the issue [17:03] Seems strange since its a local process [17:04] To my knowledge [17:22] duanedesign ? [17:22] 1 more question [17:22] yes [17:23] I got a todo list ./home/USERNAME/.evolution/tasks/local/system/tasks.ics but its not loaded by my evolution mail (i know its not desktopcouch specific) [17:23] My contact get loaded agiang [17:23] *contacts [17:23] *again [17:28] rakshasa: did you create it in Evolution or are you importing it? [17:28] Created it in evolution [17:29] But evolution only seems to load contacts again, not todo list [17:31] Yes [17:31] Problem solved [17:31] The file was locked by me [17:32] Thanks for all the support guys, you guys were awesome === beuno-lunch is now known as beuno [17:33] aha [17:33] rakshasa: awesome [17:33] glad yyou got it working. [17:34] and I learned not to confuse ubuntuone with the local couch [17:35] :p [17:56] nessita, ping [17:57] nessita, I'm working on the ubuntuone-storage-protocol/stable-1-6 [17:57] nessita, and tests on that branch are not passing, because that stable branch is not calling "yield super(...).setUp()" like it should. [17:57] nessita, so, how should I approach this? [17:57] alecu: yeah :-/ [17:57] nessita, should I fix these tests in my branch? [17:58] yeah... [17:58] nessita, ok. [17:58] (I know, it would be a pain) [17:58] specially for u1client [17:58] nessita, ugh! [17:58] nessita, my question was if stable-1-6 branch was being built using a previous u1devtool [17:59] alecu: no, is built using latest, afaik [17:59] nessita, perhaps for u1-client I can work on a different branch first, that fixes all those tests [17:59] nessita, and then work on the branch that fixes this. [17:59] nessita, I have no preference, so that's why I'm asking [17:59] alecu: in trunk we have all fixed in every project, but stable branches are not fixed [18:00] alecu: so you will need to fix every stable branch in similar ways [18:00] nessita, right, I know that trunk is fixed, so I meant different branches to fix all the stable u1-client tests [18:00] nessita, and then different branches with this bugfix [18:01] alecu: a single branch will be better for the SRU process [18:01] nessita, great, that's what I needed to know. [18:02] nessita, thanks a bunch! [18:02] alecu: thank *you* === yofel_ is now known as yofel [18:05] nessita, another thing: I'm getting tons of lint issues in the stable-1-4 branch, probably because of the newer pylint version. [18:17] nessita, this is the place where the failing checks were removed in stable-1-6: http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-storage-protocol/stable-1-6/revision/122#setup.py [18:18] nessita, so, what shall we do with that in stable-1-4? === beuno is now known as beuno-lunch [18:23] alecu: looking [18:24] alecu: I'm not sure what you'er asking (after looking at the url) [18:24] alecu: need to skype/mumble? [18:25] nessita, sure, let's [18:26] alecu: skype for me today, I'm with the laptop [18:26] looking the headset === beuno-lunch is now known as beuno [19:16] alecu: shall I review https://code.launchpad.net/~alecu/ubuntuone-storage-protocol/timestamp-autofix-1-4/+merge/81758 ? [19:17] nessita, it has the issue with pylint [19:17] nessita, I have not tested it in maverick yet, it's still downloading. [19:18] alecu: ok then, I'll wait [19:36] Hi, I'm having trouble with the music store. Is someone around who can help? [19:57] @ping [19:57] pong [19:59] @ping [19:59] pong [20:30] EOD! bye all [21:30] * nessita -> eods === m_conley is now known as m_conley_away [23:31] Hey guys, can I point Ubuntu One at other directories in my home directory for backup? [23:32] amusselman: yes, or no [23:32] amusselman: that depends on what you mean by 'backup' [23:32] amusselman: you can sync arbitrary folders in your home directory. But it's not backup -- there is backup built on top of Ubuntu One, but it's a separate thing. [23:32] Chipaca: I mean for incremental backups to the Ubuntu One "cloud" [23:33] Alright [23:33] amusselman: I'd try deja-dup's Ubuntu One mode [23:33] Okay [23:33] amusselman: it's got some glitches, but it essentially does what you describe [23:33] I don't want to roll up "backups" [23:33] amusselman: (the glitches are more on the server side, and we're working on them -- meanwhile, if it gives you a 500, retry) [23:34] Maybe the thing I want is not called "backup" [23:34] I want to sync my files [23:34] up to an off-site location [23:35] Oh, I see that is exactly what you are saying [23:35] Thanks Chipaca [23:35] amusselman: if it's just sync, then, in nautilus, right click on the folder, ubuntu one, sync this folder [23:37] I want rolling syncs [23:37] that if something disastrous happened will allow me to reinstate my files [23:56] Chipaca: thanks, deja-dup looks like it does what I want.