cwayne | fagan: thanks :) | 00:17 |
---|---|---|
adorilson | hi, people | 01:37 |
adorilson | my machine is outdated, but --current-transfers option u1sdtool, show 0 | 01:37 |
adorilson | why? | 01:37 |
adorilson | and --waiting option raises a exception | 01:37 |
adorilson | http://pastebin.ubuntu.com/612518/ | 01:37 |
adorilson | fagan ? | 01:37 |
fagan | adorilson: im about to head off to bed could you come back in the morning | 01:38 |
adorilson | fagan: ok | 01:39 |
=== nigelb_ is now known as nigelb | ||
fagan | morning all | 08:43 |
mandel | oh, I forgot morning all | 08:51 |
* mandel wonders if there is any point to be here for 2 hours if no one knows… | 08:52 | |
fagan | mandel: im here :) | 09:00 |
fagan | and morning | 09:00 |
mandel | hello :) | 09:00 |
* fagan wonders if ralsina is around | 09:01 | |
fagan | (I think I need a better task than trying to find a bug to do) | 09:01 |
mandel | fagan: ralsina is not here 2day | 09:52 |
fagan | mandel: damn hmmmmmm | 09:53 |
fagan | let me think what would be a good way to spend my time | 09:53 |
duanedesign | mornings | 09:53 |
fagan | duanedesign: morning | 09:55 |
* fagan takes a break while he tries to think up something | 09:59 | |
fagan | mandel: could you link me to that setup wiki page that you made | 10:27 |
fagan | the one for the depends | 10:27 |
fagan | ill link the image there | 10:27 |
fagan | wow its 2.9gb this is going to take a week to upload to u1 | 10:31 |
mattiab | mandel: where can I find the overlay icons for Windows? | 10:40 |
mandel | mattiab: design has not yet done them… and at the moment in linux we use the stock icons from the system, I'd say we can use a random icon 'til then | 10:44 |
mandel | mattiab: as long as we have an icon… it is the desing the ones that have to provide them | 10:45 |
jo-erlend | I need someone to confirm a bug for me. 1) copy a file to your Ubuntu One folder that's automatically synced. 2) look at your files on the web in a browser. 3) publish the file from that page. 4) confirm that Nautilus doesn't see the file as published so you have to first publish it again and then you can unpublish it. | 11:27 |
fagan | jo-erlend: it gives out the same publish url | 11:30 |
jo-erlend | fagan? | 11:30 |
fagan | jo-erlend: there isnt really an isue there other than the client not knowing its published it actually figures out its published and hives the same url | 11:31 |
fagan | *gives | 11:31 |
jo-erlend | fagan, if I have published a file on the web, then I cannot see that it is published in Nautilus. However, if I want to unpublish that file from Nautilus, I have to publish it first. Then, if I do the exact same thing again, then I can unpublish it. Certainly, that must be a bug? | 11:32 |
fagan | jo-erlend: yeah but it isnt really critical thats what I was saying and yeah confirmed | 11:33 |
jo-erlend | what package should I use when reporting it? | 11:35 |
fagan | jo-erlend: ubuntuone-client and someone can reassign if needs be | 11:35 |
jo-erlend | great. Thanks. | 11:36 |
fagan | np | 11:36 |
fagan | soooo quiet | 12:36 |
jo-erlend | https://bugs.launchpad.net/ubuntu/+source/ubuntuone-client/+bug/788073 | 12:49 |
ubot4 | Launchpad bug 788073 in ubuntuone-client (Ubuntu) "Ubuntu One-menu in Nautilus is not properly synched. (affects: 1) (heat: 6)" [Undecided,New] | 12:49 |
jo-erlend | ... if someone has something to add? | 12:50 |
fagan | nessita: ^ | 12:50 |
nessita | hi fagan | 12:51 |
fagan | hey nessita | 12:51 |
nessita | jo-erlend: hey there. Yes, there are a couple of issues regarding how our plugin keeps the info in synch with the backend | 12:52 |
nessita | jo-erlend: for now I would advice using the web UI | 12:52 |
jo-erlend | yes, but I'm not trying to avoid the issues. I'm trying to identify them so that eventually, I'll be able to recommend Ubuntu to users who just need stuff to work :) | 12:54 |
jo-erlend | btw; is the Ubuntu One item in the messaging indicator supposed to do anything other than launch the GUI? It seems to do exactly the same as the icon on the launcher? | 12:56 |
* fagan break | 12:57 | |
rye | jo-erlend, it is also showing new shares when they are accepted | 12:57 |
jo-erlend | rye, what exactly does that mean? Do they become available as menu items in the messaging menu? | 12:58 |
rye | jo-erlend, yes, they are opening the ubuntuone-control-panel in Folders tab where one can choose whether the folder needs to be downloaded to this particular machine | 12:59 |
rye | jo-erlend, other than that - nothing else | 12:59 |
nessita | rye, jo-erlend: it also notifies when there is a new folder available (kinda a share from you to you) | 12:59 |
jo-erlend | oh, ok. Then that must mean "sync locally" can be different on each of my computers? | 13:00 |
rye | jo-erlend, yup, in natty the default is not to sync your UDFs unless you say so | 13:01 |
jo-erlend | UDF? | 13:01 |
mandel | jo-erlend: User Define Folders | 13:02 |
mandel | jo-erlend: are those folders that are not the 'Ubuntu One' folder that are sync to the could from your home directory, like ~/Music for example | 13:02 |
jo-erlend | right. I thought a folder would be either synced or not. That's good news. | 13:03 |
nessita | jo-erlend: you can choose in a per-device manner which folders and which shares your want to be synch | 13:04 |
nessita | jo-erlend: you can even have 'web-only' shares/folders | 13:04 |
mandel | nessita: ping | 13:07 |
nessita | mandel: pong | 13:07 |
jo-erlend | that's very promising. Is there a way to sync these folders locally? That is, if my desktop and my Ubuntu One account are synced, can I sync my laptop with the desktop without going through Ubuntu One? | 13:07 |
mandel | nessita: I'm going to have to go for lunch in a bit, but do you want to take a look at the CredentialsManagementTool issue that I added in the branch during the sprint? | 13:08 |
jo-erlend | my LAN is somewhat faster than my broadband connection :) | 13:08 |
mandel | nessita: at least to understand what is your plan? | 13:08 |
nessita | jo-erlend: not at the moment, but we have that feature planned | 13:08 |
nessita | mandel: sure, let's talk. I have no current plan :-). Wanna mumble | 13:08 |
nessita | ? | 13:09 |
mandel | nessita: sure, starting it up | 13:09 |
* nessita too | 13:09 | |
=== m_conley_away is now known as m_conley | ||
dobey | hmm | 13:53 |
nessita | me | 14:00 |
thisfred | me | 14:00 |
nessita | mandel, dobey, fagan | 14:00 |
mandel | nessita: stand uo? | 14:01 |
mandel | up? | 14:01 |
dobey | me i suppose | 14:01 |
nessita | yeap | 14:01 |
nessita | mandel: say me! :-) | 14:02 |
nessita | fagan: only you missing | 14:02 |
mandel | me | 14:02 |
nessita | DONE: more of the same (QT windows control panel port, reviews, emails) | 14:02 |
nessita | TODO: guess what!!! yes, just like yesterday. | 14:02 |
nessita | BLOCKED: nopes | 14:02 |
nessita | NEXT: thisfred | 14:02 |
thisfred | * NEEDSREVIEW https://bugs.launchpad.net/bugs/781119 | 14:03 |
thisfred | * INPROGRESS https://bugs.launchpad.net/bugs/781538 | 14:03 |
thisfred | * TODO https://bugs.launchpad.net/ubuntuone-control-panel/+bug/781875 | 14:03 |
thisfred | DONE:reviews | 14:03 |
thisfred | NEXT: dobey | 14:03 |
ubot4 | Launchpad bug 781119 in ubuntuone-couch (Ubuntu) (and 1 other project) "Crashes if not logged into Ubuntu One (affects: 1) (heat: 6)" [Medium,In progress] | 14:03 |
ubot4 | Launchpad bug 781538 in ubuntuone-couch (Ubuntu) (and 1 other project) "OAuth support doesn't handle query parameters (affects: 1) (heat: 300)" [Undecided,New] | 14:03 |
ubot4 | Launchpad bug 781875 in ubuntuone-control-panel "ERROR - ReplicationSettingsChangeError: args (<ubuntuone.controlpanel.dbus_service.ControlPanelBackend at /preferences (affects: 1) (heat: 17)" [Undecided,Confirmed] | 14:03 |
dobey | λ DONE: Nigh finished nightlies fixes, logilab-common patch on O | 14:03 |
dobey | λ TODO: Reviews | 14:03 |
dobey | λ BLCK: None. | 14:03 |
dobey | mandel | 14:03 |
mandel | DONE: Worked on ubuntuone-dev-tools to add reactors as plugins. Discovered that we have the same problem as the one that alecu found importing the qt reactor from a func. The trick is to use __import__ instead. QTF? | 14:03 |
mandel | TODO: Finish the dev work. Re-add the CredentialsManagement code after talking with nessita. | 14:03 |
mandel | BLOCKED: no | 14:03 |
mandel | s/QTF/WTF | 14:03 |
nessita | NEXT: fagan | 14:04 |
nessita | any comments? | 14:04 |
mandel | sometimes python does weird things | 14:05 |
thisfred | could use quick reviews on https://code.launchpad.net/~thisfred/ubuntuone-couch/lp-781119/+merge/62176 | 14:05 |
dobey | s/sometimes// | 14:06 |
nessita | thisfred: I'll take it! | 14:09 |
thisfred | thx! | 14:09 |
nessita | thisfred: can I test it IRL? | 14:21 |
thisfred | nessita: you can, if you're willing to throw your SSO credentials away | 14:21 |
nessita | of course! | 14:21 |
nessita | I throw them away regularly (?) | 14:22 |
thisfred | So if those are gone from the keyring | 14:22 |
thisfred | then try: | 14:22 |
thisfred | u1oauthrequest https://example.com | 14:22 |
thisfred | and it should tell you it can't find the credentials, rather than give a nasty traceback | 14:23 |
nessita | it worked | 14:24 |
nessita | approved | 14:24 |
thisfred | yay! | 14:24 |
fagan | sorry took a quick sleep to get a bit more lively | 14:26 |
fagan | me | 14:26 |
* fagan writes notes | 14:26 | |
fagan | DONE | 14:28 |
fagan | * Made a vm image to use for testing | 14:28 |
fagan | * Read down through some of a book I have about networking | 14:28 |
fagan | TODO | 14:28 |
fagan | * Find something to do or get a bug | 14:28 |
fagan | Blocked | 14:28 |
fagan | * nope | 14:28 |
fagan | sorry bout that nessita | 14:28 |
nessita | fagan: do you feel better now? | 14:29 |
fagan | nessita: yeah im good | 14:29 |
nessita | fagan: :-) | 14:29 |
fagan | I was good just needed a brain refresh | 14:29 |
fagan | reboot if you will | 14:29 |
fagan | Oh and one more to DONE * college admin crap that I needed to sort | 14:30 |
fagan | nessita: is there anything in the cp that I can look at for the day | 14:32 |
fagan | Im still open to ideas on what I could make myself useful on for the day | 14:32 |
nessita | fagan: not that I can think of from the top of my head. We do have some issue with our nautilus plugin, are you familiar with it? | 14:32 |
fagan | nessita: not really but I can investigate | 14:33 |
fagan | nessita: since I do know gtk and twisted now | 14:33 |
nessita | fagan: if you feel like it, it would be great. Latest bug report, for example, is: bug #788073 | 14:33 |
ubot4 | Launchpad bug 788073 in ubuntuone-client (Ubuntu) (and 1 other project) "Ubuntu One in Nautilus is not properly synched regarding published files (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/788073 | 14:33 |
nessita | fagan: no twisted needed for this plugin, just C and some syncdaemon and dbus knowledge | 14:34 |
fagan | nessita: yeah cool ill have a poke | 14:34 |
fagan | nessita: well me and the dbus arent friends yet other than in theory but ill figure it out | 14:34 |
dobey | ah, published files listing | 14:35 |
dobey | didn't even need to open the bug and i already knew what the problem was | 14:36 |
dobey | and it has nothing to do with C and dbus | 14:36 |
fagan | ahh ok | 14:37 |
fagan | Well ill poke the plugin anyway and look at the code instead then | 14:38 |
jo-erlend | I have enabled sync of Firefox bookmarks in Ubuntu One on both my laptops but nothing seems to be synced. They're both running 11.04. Do I need to do anything else? | 14:39 |
jo-erlend | or perhaps only newly added bookmarks will be synced? | 14:39 |
dobey | __lucio__: how do you feel about adding the file publishing stuff to storage-protocol? | 14:45 |
nessita | dobey: I know public files are not part of the protocol, but the plugin should show consistent info, right? The public file list can be queried and the emblems can be updated accordingly | 14:50 |
dobey | nessita: we would have to poll, which would be devestatingly horrible | 14:51 |
dobey | nessita: and even if we polled, there's no guarantee that the client would be in sync when the user right-clicks on a file in nautilus | 14:52 |
dobey | nessita: so the client *should* show consistent info, yes; but it is impossible with the current implementation. the right fix is to add the stuff to protocol and do it through there. | 14:53 |
jo-erlend | is sync of Firefox bookmarks supposed to be working in 11.04? | 14:53 |
dobey | jo-erlend: should work i think. but i don't know enough myself to be able to give you good answers | 14:53 |
thisfred | jo-erlend: it is supposed to be | 14:53 |
jo-erlend | I can't see any sign of it being functional at all. Can someone help me troubleshoot? | 14:54 |
dobey | jo-erlend: are you syncing contacts? | 14:54 |
thisfred | jo-erlend: can you look in ~/.cache/desktop-couch/log/desktop-couch-replication.log | 14:55 |
thisfred | and see if that has any errors when replicating the bookmarks db | 14:55 |
jo-erlend | dobey, what I've done is to check the box in the Ubuntu One GUI on both laptops. Something was installed and I haven't done anything else. Am I supposed to? | 14:55 |
nessita | dobey: I fixed https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/move-data-gtk/+merge/62209 | 14:56 |
dobey | jo-erlend: well for contacts, the contacts have to be added on the web site, or in evolution in the additional address book there | 14:56 |
jo-erlend | dobey, oh. I am syncing contacts, yes, but I was tolking about Firefox bookmarks. Contact sync seems to be working as expected. | 14:56 |
cwayne | anyone have any idea how to oauth-sign a request in js? | 14:57 |
nessita | cwayne: js == javascript? | 14:57 |
dobey | jo-erlend: ok; i was asking to see if it worked as well, to either place or take away, blame on desktopcouch :) | 14:57 |
dobey | nessita: yes | 14:58 |
fagan | nessita: yep :) | 14:58 |
nessita | dobey: yes, you're approving? :-) | 14:58 |
* fagan hates shortenings too | 14:58 | |
cwayne | nessita: yerp | 14:58 |
nessita | cwayne: I have no idea then :-) | 14:58 |
jo-erlend | dobey, it's actually been a little while since I added or removed a contact. Let me see. | 14:58 |
dobey | nessita: yes js == javascript | 14:58 |
nessita | dobey: oooohhhh for a minute there I was happy ;-) | 14:58 |
fagan | one of the web guys would know id say | 14:58 |
thisfred | cwayne: you could look at what couchdb does for it's test suite, which is all js, and does test the oauth bits, I think | 14:59 |
dobey | cwayne: http://oauth.googlecode.com/svn/code/javascript/ | 14:59 |
thisfred | cwayne: http://oauth.googlecode.com/svn/code/javascript/ | 14:59 |
thisfred | ah :) | 14:59 |
cwayne | dobey: u1 doesn't give a callback url, correct? | 15:00 |
dobey | nessita: i also approved your branch | 15:00 |
nessita | dobey: thanks! | 15:00 |
dobey | cwayne: not sure how to answer that, but callback is a required part of oauth 1.0a which we use on u1 | 15:00 |
dobey | thisfred: why did you make CredentialsNotFound.__str__ look like a repr? | 15:02 |
jo-erlend | dobey, actually, no.. Contacts doesn't seem to be synced either. | 15:02 |
dobey | jo-erlend: ok, then look at the replication log that thisfred mentioned | 15:02 |
thisfred | dobey: I copied that from another exception class basically | 15:02 |
thisfred | cargo cult programming FTL | 15:03 |
thisfred | dobey: it doesn't make sense, I'll take that out | 15:03 |
nessita | thisfred: would you be available for a simple review? | 15:04 |
thisfred | dobey: r16 pushed | 15:04 |
thisfred | nessita: sure | 15:04 |
nessita | dobey: I added a new revision to remove all the previously set <property name="pixbuf">services-bookmarks.png</property> (I was getting GTKWarning) | 15:04 |
nessita | dobey: you may wanna confirm the approve | 15:04 |
nessita | thisfred: https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/move-data-gtk/+merge/62209 | 15:05 |
__lucio__ | dobey, from ambivalent to disinterested, depending on who does it. | 15:06 |
jo-erlend | dobey, actually, I find a few python tracebacks in that file when syncing couchdbs. Want me to paste some of them? | 15:06 |
dobey | jo-erlend: pastebin would be good yes | 15:07 |
dobey | nessita: that makes sense; still approve | 15:07 |
thisfred | jo-erlend: yes please, but not here, could you use paste.ubuntu.com? Also verify that they don't contain anything you don't want to be public knowledge | 15:07 |
nessita | thanks | 15:07 |
jo-erlend | thisfred, hehe... | 15:08 |
jo-erlend | dobey, thisfred: can you have a look and tell me if I need to remove something? :) http://ubuntuone.com/p/vWf/ | 15:13 |
thisfred | hmm, this I've seen before, recently ServerError: (500, ('json_encode', '{bad_term,<0.10918.2>}')) | 15:15 |
thisfred | This is new, but weird: SSLError: [Errno 8] _ssl.c:499: EOF occurred in violation of protocol | 15:16 |
* jo-erlend thinks it's a little strange that you cannot publish an entire directory. | 15:17 | |
dobey | jo-erlend: are you behind a proxy? | 15:17 |
jo-erlend | dobey, no. I'm on a NATed network, but I don't use any proxies. | 15:17 |
jo-erlend | and I've never had any problems with SSL or TLS before on this computer on this network. | 15:19 |
dobey | well that's all from yesterday as well. there are no entries from today? | 15:20 |
thisfred | jo-erlend: it looks like bookmarks did replicate successfully from that machine to the cloud at least once yesterday | 15:21 |
thisfred | jo-erlend: are you sure they're not on the other machine? They may have ended up in a separate bookmarks folder. | 15:22 |
mandel | oh my… bloody 32 Ceclcious… no way to work like this.... | 15:24 |
mandel | bloody spanish weather! | 15:24 |
jo-erlend | thisfred, I've searched and searched, but I haven't found any sign of any bookmarks being synced. | 15:27 |
nessita | dobey: when you have a moment, would you check this packaging-dailies branch please? https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/gui-package/+merge/62309 | 15:29 |
thisfred | jo-erlend: well they are being replicated from this machine, so maybe the problem is on the other machine? | 15:29 |
dobey | nessita: some needs fixings | 15:38 |
nessita | ack | 15:38 |
dobey | jo-erlend: right, so that replication log says "no change" between that machine and the u1 server. check the replication log on the other machine as well. | 15:39 |
CardinalFang | karni_, that android:layout_alignWithParentIfMissing is really tough. If I have it correct, I'm triggering bugs in Android. | 15:42 |
karni_ | CardinalFang: hah! so it wasn't "just me" :) cool, thanks you've looked into that. it gave me a headache | 15:42 |
jo-erlend | dobey, the other replication log: http://ubuntuone.com/p/vWq/ | 15:44 |
thisfred | jo-erlend: that one seems to have a problem talking to the server | 15:46 |
thisfred | the wrong oauth tokens maybe | 15:46 |
jo-erlend | ok. How do I fix that? | 15:47 |
jo-erlend | it does sync files properly. | 15:47 |
karni_ | CardinalFang: in worst case we could try different views for rows which don't have "last_modified" label, but AFAIK with CursorAdapter ListView that's not trivial. | 15:47 |
nessita | dobey: I'm not familiar with Conflicts/Replaces. Are you saying that python-ubuntuone-control-panel shoudl conflict with ubuntuone-control-panel? I would think it needs to conflict with the package of the same name (python-ubuntuone-control-panel). Or the one you're referencing is the source package? | 15:47 |
thisfred | jo-erlend: weird. Did you add this machine today? | 15:47 |
thisfred | jo-erlend: in theory it could be that the new token has not been added to couchdb, though I thought that was instantaneous | 15:48 |
jo-erlend | thisfred, no, it's been added since I installed 11.04. | 15:48 |
thisfred | ok | 15:48 |
dobey | nessita: yes, because you moved files from ubuntuone-control-panel to python-ubuntuone-control-panel | 15:49 |
karni_ | CardinalFang: I need your brainz for a sec, got a moment? | 15:49 |
jo-erlend | thisfred, then.. If I remove that computer and add it again, it should fix the problem? | 15:50 |
thisfred | jo-erlend, not sure, so let's not do that yet | 15:50 |
karni_ | CardinalFang: I was thinking of reusing the ACTION_UPLOAD_PICTURES for retrying failed transfers as well, but it's obviously a bad idea - if user has not enabled the picture auto upload, we won't try to resume his uploads | 15:50 |
=== karni_ is now known as karni | ||
thisfred | I'm thinking of a good way to diagnose this | 15:50 |
dobey | nessita: so the package that now has the files, needs to conflicts/replaces the package that used to have the files | 15:50 |
karni | CardinalFang: so I'm thinking if we should only for for a 1) notification ("tap to retry transfers") and 2) option menu -- these two instead of the action_upload_pictures. still, I'd like to auto-retry picture upload if "picture auto-upload" is enabled. | 15:52 |
karni | CardinalFang: so, in the end, I'm thinking of: 1) notification "Tap to retry transfers." 2) option menu (same) 3) retry *picture* uplaods when action_auto_upload is invoked | 15:52 |
CardinalFang | karni, we should only ...? | 15:53 |
karni | CardinalFang: *go for | 15:53 |
karni | CardinalFang: question is, how do we tell the auto-uploaded (failed transfers) from regular uploaded pictures (other than take from camera, or manually uplaoded, when instant upload is off) | 15:53 |
karni | CardinalFang: to sum up - how should we approach auto-retry of failed transfers in case of uploads. | 15:55 |
CardinalFang | karni, maybe a "Sending X, Receiving Y, Waiting Z" notification should point to an activity showing the current state, which has a retry button is there are any that are stalled or could be retried. | 15:55 |
karni | CardinalFang: both manual and automatic | 15:55 |
nessita | dobey: I understand, thanks | 15:55 |
* mandel => eod | 15:56 | |
karni | CardinalFang: cool, but that's a little to much for one day :< So, at the moment, you can retry uploads and downloads just by clicking on the list item if it failed. | 15:56 |
dobey | wtf planet gnome | 15:56 |
CardinalFang | karni, if a transfer is stalled and it was initiated automatically, it should retry automatically. Yes. | 15:57 |
karni | CardinalFang: AH! imporant thing. you should ping __lucio__ about the music app. He mentioned yesterday, that he got 45.000 requests for the same song one day or something like that. | 15:57 |
karni | CardinalFang: the rest should not retry automatically? | 15:57 |
CardinalFang | Hrm. Maybe they should. Exponential backoff on failure.. | 15:59 |
CardinalFang | ^ For both automatic and manual. | 16:00 |
karni | CardinalFang: yes, I've asked beuno about exp backoff | 16:00 |
karni | CardinalFang: so you're saying, we should retry all transfers automatically | 16:00 |
thisfred | jo-erlend: you'll be happy to know I have the exact same behavior on my own machine, though I'm completely mystified as to why | 16:01 |
thisfred | jo-erlend: I wonder if the servers are having problems | 16:01 |
karni | CardinalFang: exp backoff is not that trivial having to wake up(?) the device? or just use Intents with no-wake locks (laggy exponential backoff hahhaa) | 16:01 |
CardinalFang | Laggy is fine. | 16:01 |
jo-erlend | thisfred, I'll be happy once you no longer have the same problem and can inform me that I no longer have it either :> | 16:01 |
CardinalFang | Immediate, immediate, 15 minutes, 15 minutes, 1 hour, 2 hours... Approximate with the inexact alarms. | 16:02 |
CardinalFang | karni, ^ | 16:02 |
CardinalFang | Just suggesting. | 16:02 |
karni | CardinalFang: 2x immediate, 15m, 30m, 1h, and every 0.5 day + on every WIFI connected event. all right | 16:05 |
CardinalFang | karni, sounds perfect. | 16:07 |
karni | CardinalFang: all right, on it | 16:07 |
jo-erlend | thisfred, I'm not even able to file a bug about this one. Can you do that? | 16:07 |
thisfred | jo-erlend yeah will do, I'll give you the bug # when it's there so you can subscribe | 16:08 |
jo-erlend | thanks. | 16:08 |
thisfred | can anyone with ubuntuone-couch installed try 'u1couchquery contacts' or whatever database you like really? | 16:09 |
dobey | thisfred: uhm, i guess mterry should approve your branch too? there is a "pending" review from him on it | 16:09 |
thisfred | dobey: yeah, he commented that he liked it, but did not approve | 16:09 |
thisfred | Since he filed the bug I wanted his feedback, but maybe a review was overkill | 16:09 |
thisfred | nessita: what do you do to nuke your U1 sso keys and then get them back again? Mine don't seem to be working after the upgrade to natty, and I have someone reporting the same thing | 16:11 |
thisfred | I want to try if readding them solves the issue, though I still have no idea what's causing itr | 16:11 |
nessita | thisfred: I open the u1cp and I go to the devices tab, and I remove the current device | 16:12 |
dobey | thisfred: remove them from the 'devices' tab in cp? | 16:12 |
dobey | thisfred: if not there, just delete them from your keyring i guess | 16:12 |
thisfred | ah ok, I was looking in seahorse :) | 16:12 |
dobey | well | 16:12 |
nessita | thisfred: is better to remove it in the 3 places (seahorse, u1 database, sso database) | 16:12 |
dobey | i don't think u1cp removes the desktopcouch oauth tokens in the keyring | 16:12 |
nessita | thisfred: u1cp will do the proper thing | 16:13 |
dobey | thisfred: but i guess it local desktopcouch oauth is working ok | 16:13 |
thisfred | dobey: yeah that seems ok | 16:14 |
thisfred | nessita: dobey: hmm, after readding it still gives unauthorized on couch access on the u1 servers... :S | 16:14 |
nessita | thisfred: how did you re-add it? | 16:15 |
nessita | thisfred: do you have the proper setting in /etc/couchdb/default.ini? | 16:15 |
thisfred | nessita: by clicking 'I already have an account' | 16:15 |
thisfred | nessita: yep | 16:15 |
dobey | i think he's talking about couchdb.one.ubuntu.com | 16:15 |
dobey | not local couch | 16:15 |
* fagan errands | 16:15 | |
thisfred | dobey: correct, but not having that setting also prevents replication | 16:16 |
thisfred | baffling | 16:17 |
dobey | thisfred: but not due to auth error to the remote server :) | 16:17 |
thisfred | perhaps it's yet another server bug | 16:17 |
thisfred | dobey: right | 16:17 |
thisfred | but it's not broken for everyone, or even for all devices of a single user | 16:18 |
thisfred | The question I have to ask myself is, where is the %2F | 16:19 |
dobey | it's where the darkness lies | 16:19 |
thisfred | maybe my token has a weird character | 16:19 |
dobey | oh that was a bad joke | 16:19 |
dobey | i got them mixed up | 16:19 |
thisfred | tokens are per device right? | 16:19 |
dobey | uhm, what %2F? | 16:19 |
thisfred | dobey: every single time before when we've had weirdness with replication, it turned out to have something to do with (un)encoded slashes | 16:20 |
thisfred | SO I've learned to look there first | 16:21 |
dobey | oh | 16:21 |
nessita | dobey: gui-package changes pushed and re-scanned | 16:23 |
dobey | nessita: approved/merged | 16:26 |
nessita | dobey: thanks | 16:26 |
oimon | nautilus on lucid is crashing when i try to publish via ubuntu one - anything i can check? | 16:41 |
fagan | holy crap my rent is paid 2 months in advance and I dont know how it happened | 17:01 |
* fagan is going to have a crap load of disposable income next payday | 17:02 | |
joshuahoover | dobey: ping | 17:16 |
nessita | ok, lunchtime for me | 17:45 |
dobey | joshuahoover: hi | 17:48 |
joshuahoover | dobey: hi! do you know if we should be able to share a udf from the nautilus context menu in natty, right? i'm seeing that you can't do this (with nightlies at least) and a user has the same question | 17:49 |
dobey | yes and no. i think you can share folders within it, but not the root folder | 17:54 |
dobey | joshuahoover: does it work differently in maverick? | 17:55 |
joshuahoover | dobey: can't remember...seems like you could...now i need to get a maverick vm setup...had one but not sure where it went ;) | 17:55 |
dobey | iirc, syncdaemon won't let you share the root folder; but if that's not the case, we can fix and sru | 17:56 |
joshuahoover | dobey: k, i'm checking...thanks! | 17:57 |
cwayne | anyone have any thoughts on why the REST api is telling me the oauth token it just gave me is invalid? | 18:18 |
nessita | cwayne: how did you get the tokens? and what error are you having? | 18:42 |
cwayne | nessita: did a call to get them from GET https://login.ubuntu.com/api/1.0/authentications with http auth | 18:43 |
cwayne | nessita: got them, then did a request for GET https://one.ubuntu.com/oauth/sso-finished-so-get-tokens/<email address> signed with the token | 18:43 |
cwayne | nessita: which gave me "ok 2/0" | 18:43 |
cwayne | nessita: then when i try to get account info, i get "Invalid access token: ;lakjsdf;lakjw3;LAW " | 18:44 |
nessita | cwayne: that means that 0 of 2 tokens where imported, let me confirm | 18:44 |
nessita | cwayne: ah no, the other way around | 18:45 |
nessita | hum | 18:45 |
cwayne | also i just rechecked i was wrong | 18:45 |
cwayne | 0/2 | 18:45 |
cwayne | sorry | 18:45 |
nessita | so, 0 tokens were added from 2 listed from SSO. Your tokens probable have a "non U1" name | 18:45 |
nessita | cwayne: what s your tokens name? | 18:45 |
cwayne | nessita: Ubuntu%20One%20@%20 + the UDID of the system | 18:46 |
nessita | cwayne: they need to be called "Ubuntu One @ <some custom string>" | 18:46 |
nessita | cwayne: hum, the %20 is the problem. Where is that coming from? | 18:47 |
thisfred | lunch! | 18:47 |
nessita | thisfred: on it! | 18:47 |
nessita | (?) | 18:47 |
thisfred | :) | 18:47 |
cwayne | nessita: hrmm, i think i added that yesterday cus i thought it wasn't uri encoding it or something... but when i remove it and put spaces i get 1/3 now | 18:48 |
nessita | ;-) | 18:48 |
cwayne | when i do the second call (sso's done so get tokens) im not passing it the name anywhere. should i be? | 18:48 |
nessita | cwayne: nopes, you shouldn't. You should be setting the display_name in SSO | 18:49 |
cwayne | nessita: just as a param? (aka not int he header with all the oauth stuff) | 18:49 |
cwayne | nessita: also, thanks! that seemed to do the trick... ps i got the %20 from here i guess https://edge.one.ubuntu.com/developer/account_admin/issue_tokens/cloud/#get_httpsloginubuntucomapi10authentications | 18:50 |
* thisfred is reheating some delicious masaman curry | 18:50 | |
cwayne | 'token_name must be set to "Ubuntu%20One%20@%20" + machine_name.' | 18:50 |
nessita | cwayne: hum, I think you misunderstood me. The name of your user is not something tied to Ubuntu One, for to Ubuntu Single Sign On | 18:50 |
cwayne | nessita: ok, yeah i pass it to sso | 18:50 |
cwayne | and now i get ok 1/3 when i tell sso bout the tokens | 18:51 |
nessita | cwayne: the name needs to be set in your profile in login.ubuntu.com, and we (as Ubuntu One) don't handle that | 18:51 |
cwayne | and i can actually get my account info from a seperate call | 18:51 |
cwayne | nessita: right | 18:51 |
nessita | cwayne: can you please report the bug regarding the documentation? | 18:51 |
cwayne | nessita: sure thing. so its okay im getting 1/3 now? should it not be 3/3? | 18:52 |
nessita | cwayne: is ok 1/3, because the other 2 does not have a valid Ubuntu One prefix | 18:53 |
cwayne | nessita: awesome, thanks! | 18:53 |
nessita | cwayne: you're welcome! | 18:53 |
cwayne | nessita: i wouldve been stuck on that forever and never even thought to get rid of the %20 :P | 18:53 |
nessita | cwayne: is ok, that's why we're here :-0 | 18:54 |
nessita | :-) | 18:54 |
=== kklimonda is now known as Guest62081 | ||
dobey | ah url encoding fun | 19:04 |
dobey | wonder if i should just add this stuff to the protocol, and hook it up on the server | 19:05 |
dobey | hrmm, i need maverick | 19:10 |
=== Guest62081 is now known as kklimonda^ | ||
MBarvian | __lucio__: ping | 19:29 |
pfibiger | MBarvian: lucio's off today, it's a national holiday in Argentina. | 20:01 |
MBarvian | pfibiger: ah, tanks | 20:58 |
MBarvian | *thanks | 20:58 |
MBarvian | pfibiger: would you happen to know if an ubuntuone client for os x is in the works? I looked around at some bug reports but it was kind of vague | 20:58 |
jderose | JamesTait: so was it you that gave rockstar a Novacut sticker? :) | 21:12 |
JamesTait | jderose: Guilty as charged. :) | 21:12 |
jderose | hehe | 21:12 |
jderose | JamesTait: well thanks! how's Argentina? | 21:12 |
JamesTait | jderose: Cold! But it's worth it for the meat. ;) | 21:13 |
JamesTait | Oh, and the quality time with my awesome team, of course. :-P | 21:13 |
jderose | JamesTait: hopefully it's not as overcooked as in Budapest :P | 21:13 |
jderose | yeah :) | 21:13 |
jderose | JamesTait: so this is the U1 sprint going on then? | 21:14 |
JamesTait | jderose: One of them, yes - this is just the Web & Mobile team for now. | 21:14 |
JamesTait | jderose: In a couple of weeks we have a full team sprint in London. | 21:15 |
jderose | JamesTait: nice... well, tell everyone hi for us :) | 21:17 |
JamesTait | jderose: Everyone says hi back. :) | 21:17 |
jderose | :) | 21:18 |
cwayne | hmm, what does the android app use as a machine name? it's not UUID... | 21:20 |
JamesTait | cwayne: The Android files app? | 21:22 |
JamesTait | karni: ^^ | 21:23 |
karni | cwayne: machine name? do you want to track users? | 21:23 |
karni | cwayne: we use the device name, which is often the device model | 21:23 |
cwayne | karni: isn't it supposed to have a unique id? machine_name as it were | 21:23 |
cwayne | hmmm | 21:23 |
karni | cwayne: a friendly (semi random?) name would be better than a UUID | 21:24 |
karni | cwayne: why? you and me can both have "HTC Hero" phone | 21:24 |
karni | cwayne: why would that matter? | 21:24 |
cwayne | karni: yeah, udid sucks. its like 2lpkjwrlksdfikhj4r5kljhsdkfj | 21:24 |
karni | ;) | 21:24 |
cwayne | karni: i suppose it wouldn't... | 21:24 |
karni | cwayne: it's only for the user, to see that on the 'device list' | 21:24 |
karni | JamesTait: tnx ^ ^ | 21:24 |
cwayne | karni: right, its just harder to pull out any other info from webos phones... | 21:25 |
karni | cwayne: right, I don't have anything better for you :< | 21:25 |
* cwayne could just use webOS and hope they don't have more than one :P | 21:26 | |
cwayne | karni: also, a user should only have to authenticate once right? then you can reuse the oauth token? | 21:27 |
karni | cwayne: correct :) | 21:27 |
cwayne | karni: whew, that makes oauth more worth it :) | 21:27 |
karni | cwayne: as long as the user doesn't de-authorize the device by removing the token | 21:27 |
karni | cwayne: :D | 21:27 |
karni | cwayne: and cool, because you can disable access from the device remotely, just by revoking the token access | 21:28 |
cwayne | karni: ooh that is pretty cool... | 21:28 |
=== m_conley is now known as m_conley_away | ||
dobey | cwayne: can't you use the machine name? | 21:57 |
dobey | or "device name" | 21:57 |
dobey | cwayne: if you go to "device info" settings, there is a "Name" field there which can be changed by users | 21:58 |
cwayne | dobey: this is with the REST api.. on the device the only info available is udid | 21:59 |
dobey | cwayne: it doesn't matter because the machine name is only used in the SSO bit; but you should be able to read that device name somewhere. doesn't "hostname" on the device give you the same string as in that Name field? | 22:01 |
cwayne | dobey: that name field is currently using the device's UDID. i did that since it's supposed to be unique (at least unique for the user). that seems to be the only info i can pull from the device about the device | 22:03 |
dobey | huh? i set mine to "Pre" | 22:04 |
dobey | it isn't a uuid | 22:04 |
dobey | and i think it was "Palm Pre" by default | 22:04 |
dobey | cwayne: are you testing in SDK/PDK or on a real device? | 22:05 |
cwayne | dobey: sdk right now | 22:06 |
cwayne | dobey: how can you get that string? | 22:07 |
fagan | holy crap in a day of upload u1 only has 100 mb of the 3.9gb im uploading | 22:08 |
dobey | cwayne: i don't know, i haven't ever developed on webos; but i would presume it is the hostname for the device or something, and there would be API to do it | 22:08 |
cwayne | dobey: you'd think so, but it doesn't appear to be the case :/ im looking around though | 22:09 |
dobey | fagan: uhm, how fast is your connection? | 22:09 |
fagan | dobey: 1mb/s upload | 22:10 |
fagan | dobey: it was max going 400kb/s and stopping every few secs | 22:10 |
beuno | FWIW, I've been trying to download 14 files for days | 22:10 |
beuno | the logs indicate the connection gets terminates constantly | 22:11 |
fagan | beuno: that sounds like whats going on with me | 22:11 |
fagan | beuno: it just stops for some reason a lot | 22:11 |
fagan | (in natty) | 22:11 |
beuno | yeap | 22:12 |
fagan | I wouldnt mind but it actually effected my network speed so much and didnt do a lot | 22:12 |
beuno | it's server issues, I'm sure | 22:12 |
dobey | cwayne: i bet it is a preference | 22:12 |
dobey | fagan: 1 MB/s or 1 Mbit/s? | 22:13 |
fagan | dobey: 1mb/s max upload | 22:14 |
fagan | (I dont know what that is in mbits | 22:14 |
beuno | dobey, it's a few mb that I need to download here | 22:15 |
beuno | so it's not network speed | 22:15 |
beuno | it's server suckyness | 22:15 |
dobey | sure | 22:15 |
fagan | yeah im sure it isnt on my end | 22:15 |
dobey | but 4 GB on a 1 MB/s upload, is not goign to be anywhere near fast | 22:16 |
fagan | dobey: well it would have a lot faster than 70mb in a day | 22:17 |
fagan | like ive been uploading since 10AM | 22:17 |
dobey | cwayne: and looks like getPreferences just returns everything, so you could call that and then look through the result to find the key for the device name i guess; perhaps it is called 'deviceName' though :) | 22:20 |
dobey | fagan: well assuming that it only got 400 KB/s upload, and it stayed at that for the past 12 hours, then you would have only uploaded about twice what you have; *shrug* | 22:21 |
fagan | dobey: well its still more than what was uploaded though | 22:23 |
fagan | I was expecting it to go faster | 22:23 |
dobey | have a good evening all | 22:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!