[01:39] g'night #ubuntuone ! === txwikinger2 is now known as txwikinger === ivanka is now known as ivanka-train [08:45] I can't help but point this out. Look at bug #690673 and the look back at http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-client/trunk/revision/934 now back at the bug. Yes, it is now in nightlies [08:45] Launchpad bug 690673 in ubuntuone-client (Ubuntu) (and 1 other project) "Syncdaemon MUST move files to trash on removal (affects: 2) (heat: 38)" [High,Confirmed] https://launchpad.net/bugs/690673 [09:01] morning === muffinre` is now known as muffinresearch [09:21] brb many updates === czajkows1i is now known as czajkowski [09:55] * fagan hates unity sometimes [09:56] rodrigo_, pong [09:56] erm [09:56] rodrigo_, ping [09:57] rye, pong [10:00] rodrigo_, so, it looks like if there is e-addressbook-factory running already before main evolution process starts then couchdb can be contacted. Additionally no couchdb interaction is seen when other e-processes are running (e-calendar-factory, e-alarm-notifier). [10:01] hmm [10:02] does not look like it is couch-related but clicking on the contact hangs screen redraw :( [10:02] yeah, there are a lot of problems, given what people are adding to the bug [10:03] 1st the tokens not being in the keyring [10:03] 2nd the tokens being different in the keyring than in the .ini file [10:03] and then all the evo-related ones [10:04] also, someone reported a couchdb folder being created in $HOME, and I checked mine and indeed it's there [10:04] I guess that's the cache API in e-d-s having changed [10:04] * mandel hates when launchpad is in read-only mode.... [10:04] and needing a full path [10:04] mandel: Can you have a look at my code I cant get it to work even though it seems ok [10:05] fagan: well, if you have a way to send it, yes, launchapd is on readonly [10:05] mandel: I can email it [10:05] oh or use u1 [10:05] that would be faster [10:06] yes, and +1 for using our own product :) [10:07] rye, hmm, I've lost the evo-couchdb bug, do you have the #? [10:07] mandel: http://ubuntuone.com/p/l9k/ [10:08] rodrigo_, i need to reboot, my compiz went crazy and I think i see some bug in evolution - it hangs the whole X server until killed when I click quickly on the contacts [10:09] rye, ok [10:13] fagan: I'll be on it as soon as I do a tech pit stop [10:14] mandel: its probably something dumb since im not really getting how the Qpainter stuff works yet [10:21] brb tea === Stecchino_ is now known as Stecchino [10:26] back [10:28] * mandel back [10:30] fagan: the firs thing I would do is remove the need of the designer stuff, I have not looked at it, but you dont need it [10:30] mandel: but I just didnt want to make a window :) [10:31] ok removing [10:32] fagan: no need, If you show the custom widget as the main widget, it will create the windows frame around, so lets remove it some make the thing simpler [10:32] cool [10:33] fagan: the paint is in the wrong location, take a look at the following: http://www.commandprompt.com/community/pyqt/x2765 [10:33] rodrigo_, could you please have a look at http://paste.ubuntu.com/590140/ - that is the bt i get when I manage to click addressbook entries until the whole session hangs (no reaction to mouse click or keyboard, swtching to vt works) [10:33] fagan: that is a small example of a custom widget, the important part for you is the paintEvent [10:34] ah ok [10:34] so I can get rid of the mouse events and work from there [10:34] fagan: sorry, bloody irc crashed, did you get my last sentece about the paintevent? [10:34] fagan: yes, that is, remove the mouse events and use that as a template [10:34] cool [10:34] yeah I got it [10:35] fagan: from that, you will start getting thing drawn, the next step would be to draw the correct thing, but that is a diff story :) [10:35] step by step, right ? [10:36] mandel: yeah I think I know what metods to use and all of that but just getting the start was the hard part [10:36] fagan: cool, then let me know if you have any other problems :) [10:36] sure [10:37] oh my, kubuntu looks horrible! [10:37] mandel: hah [10:38] I like how kubuntu looks but hates the theme [10:38] but unity looks pretty good now so im happy using that [10:38] (all the crashes make me regret that) [10:39] yeah, unity looks way better, and I have no problem sin my ubuntu machine, boring dells for the win! [10:40] mandel: msi wind is a little funny [10:40] its new so I hope the issues get fixed soonish [10:40] the webcam doesnt work, if I plug it out of the mains it doesnt let me use the wireless [10:41] that kind of crap [10:51] fagan: do remember that we have the spring next week, so make sure that you have a working windows vm to work with set up so that we dont loose to much time setting up the env [10:51] I'd like to take as much advantage as possible form the sprint [10:53] lisette: ping [10:53] mandel: well for doing the interface and all that do I need Windows on it? [10:53] * fagan removed windows from it because the way msi set it up was terribly annoying [10:54] fagan: every branch will have to be tested on windows, so just having two vms will not be enough, and you will have to do reviews that are not just UI reviews [10:54] mandel: ah ok ill grab windows xp or something and dual boot it [10:55] fagan: why not a vm? [10:55] mandel: its an msi wind I dont think it can handle a vm [10:55] mandel: 1.6ghz [10:55] (dual core) [10:56] fagan: dont you have something a little bit more powerful? is that you only laptop? [10:56] mandel: the one I had before it was 1.7 ghz and 7 years old [10:57] Im using a desktop for my main environment at home [10:58] bummer…. well, is not your fault I'll take a look on what we can do about it [10:58] mandel: id say there would be some other computers lying around [10:58] good morning! [10:58] morning karni [10:58] fagan: I doubt it a lot [10:59] bummer :) [11:00] mandel: I couldnt really help it too much but the dual boot should help since I cant vm it [11:00] it would do [11:01] fagan: I really dont want you to be constantly rebooting [11:01] fagan: I sort out something [11:01] mandel: yeah ok [11:01] I'll [11:04] mandel: in that example there is a bit that isnt defined and I cant find it anywhere its the bitBlt thing [11:05] fagan: well, that is just a detail of what they are drawing, you can ignore that [11:05] you wont use that at all [11:05] ah ok [11:11] rye, use 'thread apply all bt' instead of 'bt' to get the backtrace [11:14] break [11:17] rodrigo_, http://paste.ubuntu.com/590160/ [11:18] rodrigo_, basically evolution is sitting in #0 0x00007f7ac76ae7c4 in __lll_lock_wait () and all other applications are also waiting for something [11:18] btw... hmmm [11:19] rye, that's the main thread waiting for other threads to end, afaik [11:19] so it seems some thread is not releasing the lock of some data [11:19] because the other threads are waiting also [11:20] rodrigo_, well, the whole desktop is frozen when i am clicking on the contact which has all possible fields filled in [11:20] except fgor thread 8, which is 'polling' [11:20] #5 0x00007f7abddfb373 in ?? () from /usr/lib/nvidia-current/libGL.so.1 [11:21] hmm, we talked yesterday about a nvidia's GL problem, let me check [11:22] rodrigo_, the nvidia binary driver create quite some ram use for each process using gl [11:22] rodrigo_, which is basically every gtk application when cairo is built with gl [11:23] rye, not sure, but your problem might be related to that, lots of memory leaked in cairo apps [11:23] rodrigo_, ok, reproduced w/o the couchdb [11:23] rodrigo_, just created the contact in local addressbook with all fields defined, clicked on it and BOOM [11:23] rye, same backtrace? [11:23] rodrigo_, well, not boom but some field is indeed causing the hang [11:24] rodrigo_, yes, waiting on the lock [11:24] hmm, ok [11:24] in __lll_lock_wait () [11:24] rodrigo_, so that's not couchdb-evolution :) [11:24] now i need to find out which field causes that [11:25] rye, see bug #725434 [11:25] Launchpad bug 725434 in nvidia-graphics-drivers (Ubuntu Natty) (and 3 other projects) "Nvidia drivers lead to extra memory usage for each process using libGL (affects: 26) (heat: 154)" [Medium,Won't fix] https://launchpad.net/bugs/725434 [11:26] rye, did you find the evo-couchdb bug I lost? [11:26] rye, the #? [11:27] rodrigo_, well, extra memory is one thing and evolution hanging the whole UI when some contact field is defined look slike a different thing... [11:27] rodrigo_, bug #673568 ? [11:27] Launchpad bug 673568 in evolution-couchdb (and 1 other project) "Error modifying contact, other error when saving contacts (affects: 42) (dups: 4) (heat: 206)" [High,Fix committed] https://launchpad.net/bugs/673568 [11:28] rye, not that one, the one about contacts not synced [11:28] rye, out-of-memory conditions can hang the whole desktop [11:28] rye, because of swapping [11:29] rodrigo_, hm, contacts not synced? in lucid? [11:29] rye, the one from jono [11:29] rodrigo_, memory usage is ok, cpu is also ok, the whole UI hangs [11:30] rodrigo_, bug #727370 ? [11:30] Launchpad bug 727370 in evolution-couchdb (Ubuntu) (and 1 other project) "Contacts not syncing with my computer (affects: 20) (dups: 15) (heat: 164)" [Low,Fix released] https://launchpad.net/bugs/727370 [11:31] rodrigo_, and by whole ui i mean the desktop, not evolution only [11:31] rye, yes, that one, thanks [11:32] heh, clicking cancel on merge contact prompt in evolution hangs evolution... [11:36] rodrigo_, this is LOL [11:36] rodrigo_, if you set Yahoo IM info then evolution locks up the desktop [11:37] rye, ok, let me try [11:37] rodrigo_, switching to vt and killing evolution should release the desktop [11:38] rye, not for me, it works ok [11:38] rodrigo_, okay [11:38] rye, so, I'd say it's related to the nvidia driver thing, what does top say? [11:39] rodrigo_, no, try adding both AIM and Yahoo identifiers to the contact [11:39] rye, ok [11:40] rye, works ok, no hang [11:42] rodrigo_, so you have a contact with name, aim and yahoo identifiers and it works fine, right? [11:42] rye, yes [11:42] why? :) [11:43] rodrigo_, by works right i mean you can click on the contact entry, edit it and save and nothing hangs, is that correct? [11:44] rye, yes, and then open it again, remove it [11:45] clarita: ping [11:46] mandel - good day [11:46] clarita: morning! [11:46] clarita: I've got 2 questions for you, do you have a few minutes, it wont be long [11:47] mandel: sure [11:48] clarita: first one: it there is an error during the registration, shall we allow the user to go back to the last page he visited in the wizard? [11:48] clarita: second one: any design available that I can take a loot at? [11:48] ups, I meant look, not loot :P [11:49] mandel: for the error - I think we should give an option to 'try again' [11:49] mandel: within the error messaging - I still need to explore these (today) after our chat last week - for the designs I need to check with Lisette - did she ping you back? [11:50] clarita: no, that is why I opted for an diff root :) [11:50] mandel: ahaaa - ok let me check [11:51] mandel: can we aim to catch up at 4pm or so today on the error messaging? [11:51] * fagan would like to listen in [11:51] clarita: sure [11:51] mandel: will include fagan :-) [11:52] :) [11:52] clarita: hmm, with 4 pm? London 4 pm? [11:52] it would be 5PM UTC then [11:52] or 3PM [11:53] god my timezones are all over the place :P [11:53] 3pm [11:53] yeah I added an hour instead of taking away hehe [11:54] BST = UTC - 1 [11:54] yeah [11:54] fagan: this is your speaking/typing clock [11:54] mandel fagan: OK I will ping you then to check you're free then Skype [11:55] * fagan remembers the rule in this channel that you cant talk about any timezone but UTC otherwise there is punishment [11:55] clarita: sure [11:55] fagan: that is why I have a clock in my desktop for each of the locations where people are working [11:56] is always easier to say 'your 5pm' :P [11:56] mandel: thats handy [11:56] * fagan misses the old ubuntu clock applet where you could do that [11:57] fagan: there is a way to do that on unity, but I have no clue how :( [12:00] rodrigo_, hm, are you using nvidia-based machine? [12:00] rodrigo_, that poll() thread is interesting, it has nvidia's libgl in the stack trace [12:02] mandel: well it would be easier just to make a little applet in the notification area that does that [12:03] rye, yes, using nvidia driver [12:03] rye, but I'm not fully up-to-date, are you? [12:03] * fagan would do one but is way too lazy [12:04] rodrigo_, yes, i am fully up-to-date [12:08] * fagan tries to find out about multi monitor support in unity [12:25] morning all [12:26] duanedesign: morning [12:26] fagan: FWIW, I use a firefox add-on called FoxClocks for helping to keep track of everyones TZ [12:28] duanedesign: I dont use firefox :P [12:29] that won't work then :) [12:30] * fagan thinks switching to AR time would be a good idea for this team :) [12:34] fagan, multi monitor support in twin-view mode is ok [12:35] rye: twin-view? [12:35] fagan, which is basically extending the desktop, not separate displays per screen [12:35] fagan, nvidia calls it twinview :) [12:35] rye: im on nvidia [12:35] fagan, with nouveau or nvidia drivers? [12:35] nvidia ones [12:36] ah I had it on seperate x screens [12:36] brb restarting [12:37] * popey uses unity with nvidia binary driver in twinview mode. works very nicely [12:40] * mandel walking dog [12:40] rye: have I ever told you I love you [12:40] * fagan now has dual monitors [13:22] mandel: ping me when you get back [13:23] still cant get this working [13:44] ralsina: donde es? === dobey_ is now known as dobey [13:56] ralsina: when you have time https://code.launchpad.net/~thisfred/ubuntuone-client/quota-notifications/+merge/55812 [14:00] fagan: ping [14:00] mandel: pong [14:00] * fagan pastebins [14:01] sorry for taking longer… the bloody firemen are in the building, no ide why [14:01] also, https://code.launchpad.net/~dobey/rhythmbox-ubuntuone-music-store/libu1-mp3-install/+merge/56471 [14:01] karni, I can't test that branch here. Grrr. ant still doesn't build it. [14:01] mandel: http://paste.ubuntu.com/590221/ [14:01] karni, in looking it over, I didn't see anything wrong, though. [14:02] fagan: what is the problem? [14:02] thisfred: sure [14:02] mandel: I still cant get it to draw [14:02] :/ [14:02] CardinalFang: you're freaking fast. they do build, beun-o and ivank-a have been testing it on their phones as well. but hell I'd like that ant to work.. [14:03] I dont think I understand how to do it yet and googling didnt help since there arent any good examples around [14:03] CardinalFang: I've put much effort to write ListActivity as cleanly as possible and reviewed my work multiple times so that we could merge ASAP [14:04] fagan: yes, there are several issues in that code, let me get you a good example, give me a min or so [14:05] mandel: cool [14:05] fagan: how good are you with c++? I mean, is an example in C++ good? [14:06] mandel: im very good at C++ [14:06] CardinalFang: I'll be moving the U1F meta-sync core from getDelta to query (via REST, because query will deprecate), this means I will have to focus on some internal flesh now.. and make it fast. [14:06] mandel: I can probably figure it out [14:06] fast, as in: the deadline is very close [14:06] fagan: then take a look at the paintEvent of http://doc.qt.nokia.com/latest/widgets-shapedclock.html [14:06] ok cool [14:07] ill try to figure it out [14:07] fagan: that is a good example of how to deal with the painEvent, you can also try to do the same widget in pyQt if you want to see if you get it right [14:15] hi mandel, fagan! [14:16] mandel: I bet your dog climbed a tree and the firemen are rescuing it [14:16] ralsina: hey [14:16] * fagan sucks at QPainter so much [14:16] ralsina: that would be good news, I thing it has to do with the fact that the yoga people burned soemthing... [14:17] mandel: too much psychic energy. Yogis should always use fireproof mats. [14:17] thisfred, still running tests, but approved. [14:18] ralsina: yes, they are weird, the have laughting sessions.. maybe is related [14:18] mandel: what's today's exercise? I can lend a hand, probably [14:18] ralsina: QPainter [14:19] (im not getting it at all) [14:20] fagan: let's go to #fagan_pyqt and tell me all about it ;-) [14:24] CardinalFang: Has it been permission to merge to trunk or should I give you more time to review (you did it ultra fast, I was quite suprized!) [14:25] karni, I want to run it first while I read the code. [14:26] karni, a first read looked good, but I don't understand every detail. [14:26] CardinalFang: Aha. You don't have eclipse configured :< [14:26] CardinalFang: Can I be of help to explain? [14:27] * karni thinks cheap breakfast cereal are not that good -_-` [14:27] karni, yes. Let me get a bug report out of someone first. Be right back with you. [14:27] karni, 5 minutes. [14:27] CardinalFang: Sure, thanks. ACK! [14:29] * karni tests hilighting: karni [14:30] CardinalFang, do you have the latest build of Ubuntu One contacts? [14:31] karni, hilight! [14:31] CardinalFang: works! thanks :) [14:35] CardinalFang, i am not able to fetch the item from the market due to it is being in an emulator [14:38] rye, not sure, but this may be it: http://sandbox.chad.org/u1c-9001.apk [14:55] lucio_, facundobatista, verterok: q: after sending a share invitation thru the web, does syncdaemon have a valid volume_id for that share? I would guess it does not. [14:55] I'm working on bug #646059, and I need to return a volume id on the ShareCreated dbus signal, since the volume id is used later by nautilus to get at the share info [14:55] Launchpad bug 646059 in ubuntuone-client "Nautilus doesn't update emblems for shares (or, syncdaemon doesn't emit response at all for shares offered thru http) (affects: 1) (heat: 2)" [High,Confirmed] https://launchpad.net/bugs/646059 [14:56] alecu, no idea, would need to follow code [14:58] alecu: the sahre was accepted? or just offered via the web? [14:58] verterok, just offered [14:58] alecu: syncdaemon knows nothing about share offers [14:58] rye, yes, I think that^ APK is the exact same as in the Market. [14:58] verterok, this is all to show the little hand holding the folder icon that signals that the directory was shared. [14:58] alecu: a share offer isn't even a share :/ [14:59] CardinalFang, thanks, syncing! [14:59] verterok, but is it a "shared" ? [14:59] alecu: nope [14:59] it's nothing from syncdaemon POV [15:00] alecu: actually from the api server POV too [15:00] verterok, right. But I see that a marker is stored when a directory is shared thru the email [15:01] alecu: where? [15:02] alecu: I think that's only used for the protocol share creation case...where we have a response from the server [15:03] alecu: once upon a time, the only way to create a share was using the protocol...then the http share offer stuff was added [15:05] alecu: if the http code in aq fires an AQ_CREATE_SHARE_OK, you will end up with a Shared instance in the "shared" shelf [15:05] me thinks [15:06] let me check [15:06] verterok, the CreateShare command is passed a "marker" parameter [15:06] thisfred, the http code fires a different event [15:06] AQ_SHARE_INVITATION_SENT [15:06] * thisfred reads backlog [15:06] (a new one we added for zg) [15:08] verterok, so, that marker is also sent when we send the AQ_SHARE_INVITATION_SENT (http) instead of the AQ_CREATE_SHARE_OK (protocol) [15:08] sorry thisfred [15:08] I meant verterok a few lines above. [15:10] alecu: so, you don't have a Shared instance in the "shared" shelf [15:10] alecu: which makes sense...as the share isn't actually there [15:12] karni, okay, a few problems sorted. [15:12] CardinalFang: kk [15:12] Now, back to ant build. ... [15:13] CardinalFang: You courageous knight! [15:15] CardinalFang: ee.. I was pretty sure I fixed that. You apply the library with $ android update project --path path/to/proj --library path/to/gd , not $ android update lib-project .. [15:16] ok, trunk has that fixed [15:19] "package greendroid.app does not exist" -- ant has issues with GreenDroid [15:33] nessita, ping [15:34] nessita, after installation of desktopcouch-ubuntuone via the control panel I am greeted with http://ubuntuone.com/p/lCT/ [15:34] looking [15:34] rye: right, I noticed that exact behaviour yesterday [15:34] rye: desktopcouch is failing with NoPairingRecord [15:35] rye: can you please file a bug for thisfred/CardinalFang to diagnose? [15:36] nessita, i think that can be a left over from a previous installation, i see the old databases there and there was no successful replication [15:36] CardinalFang: Cyril (author of GD) writes clearly in the README how to apply the lib with ADT plugin. I've gone to Android dev website and tried the command line way, but I can't make ant recognize the lib project properly :[ [15:36] however no pairing record... [15:37] rye: I don't think is related, the U1 plugin for DC is complaining about no pairing record [15:37] CardinalFang: I do not, however, doubt your ant-jitsu. Never. :) [15:37] rye, is this running nightlies? [15:40] nessita, now it is even more interesting [15:41] CardinalFang, yes, the host is running nightlies, recently updated [15:41] CardinalFang: on the other hand.. ant does say "Ordered libraries: [setup] /home/mike/src/android/ubuntuone/GreenDroid/GreenDroid" (on my computer) [15:42] nessita, how do you like this layout - http://ubuntuone.com/p/lCa/ [15:42] rye: I love it! :-D [15:42] nessita, but wait! [15:42] rye: click on ALT having the focus on the control panel [15:43] we do magic! [15:45] nessita, http://ubuntuone.com/p/lCb/ [15:45] rye: press ALT [15:45] (only ALT) [15:45] nessita, but once I click Alt... it is all appears :) [15:45] nessita, why alt :) ? [15:46] rye: there is some weird stuff going on with compiz, I would guess [15:46] rye: we have no idea, but that's the magic key [15:46] nessita, that's not a compiz thing [15:46] nessita, that thing is running in a standard gnome [15:46] gtk, compiz, something [15:46] nessita, moreover it is running over the network :) [15:46] rye: bug #750309 [15:46] Launchpad bug 750309 in ubuntuone-control-panel (Ubuntu) (and 1 other project) "Services tab layout doesn't work right (affects: 1) (heat: 6)" [Medium,Triaged] https://launchpad.net/bugs/750309 [15:46] rye: whoever is drawing the windows is not noticing some widgets [15:47] nessita, in my unity session it was drawing at least something, now only blank bar below the file sync... [15:48] rye: I have no idea how to debug those weird rendering issues :-( we've had some here in the sprint as well [15:48] rye: we may have a workaround though [15:52] nessita: ping [15:52] mandel: pong [15:53] ... aand http://ubuntuone.com/p/lCh/ [15:53] nessita: one quick question about sso, when you are loading the image you show a nice spiral, where did you get that guy from? [15:54] mandel: is a gtk.Spinner, which allows to start it or stop it [15:54] ralsina: someone turned my router off a few times :/ [15:54] nessita: bummer… I wanted something that nice :P [15:55] nessita: thx for the help :) [15:55] mandel: does QT have something like that? [15:55] I just logged into windows since mandel and clarita and I are having a meeting on skype and my ubuntu doesnt actually work with skype :/ [15:55] nessita: I'm looking hehe :) [15:55] Bad Request on modifying the contact, latest natty with nightlies, couchdb restarted, e-addressbook-factory restarted :(... [15:56] going even further away [15:56] fagan: eek - what are the alternatives [15:56] rodrigo_: ^ [15:56] clarita: its cool im in window now so should be perfect [15:56] rye, can you activate the debugging please? [15:56] rye, export COUCHDB_DEBUG_MESSAGES=1 [15:56] grr, gio is not working right with g_file_enumerate_children_async for me :( [15:57] fagan: wasn'tme :-D [15:57] rye, and then run e-addressbook-factory on the terminal, as before [15:57] rodrigo_, are your packages in natty/main or nightlies? because nightlies do not override main now due to version [15:57] rye, Bad Request means bad oauth tokens, btw [15:57] mandel: you could use a label with a QMvie (an animated gif) [15:57] rye, which packages? [15:57] ok mandel fagan: ready for a call in 3? [15:57] ralsina: if I was on a VM it would be the same since it would inherit all the voice issues from the host OS [15:57] rodrigo_: it shouldn't. if it does that's a bug [15:57] clarita: im ready now :) [15:57] fagan: yeah [15:58] rodrigo_, Could not PUT document: \x98^ [15:58] clarita: always read :) [15:58] I think I'll listen, if you don't mind! [15:58] mandel fagan: ok, hold on to your hats [15:58] rye, that's the error? and the whole output? [15:58] hah [15:58] dobey, it's a bug for sure :) [15:58] clarita: gime me 2 min to grab a coffee and the headphone :) [15:58] dobey, but the Bad Request errors are usually because of wrong oauth tokens [15:59] mandel: okey dokey [15:59] rodrigo_: i mean invalid tokens should return HTTP 401 for oauth, 400 for some other errors with the request [15:59] 400 is not found, right? [15:59] 400 is Bad Request [15:59] 401 is Unauthorized [15:59] 404 is Not Found [15:59] right [15:59] CardinalFang: I'm giving up on ant.. moving on to next task. let me know if I can help clear up the review for you. [16:00] rodrigo_, you will be pleased to know that my evolution addressbook now hangs the vm's session too [16:00] so it is not nvidia-driver specific [16:00] mandel: answer :) [16:00] rye, I'm not pleased [16:01] rye, with the same hang in nvidia's libgl? [16:01] rodrigo_, http://paste.ubuntu.com/590287/ [16:02] rodrigo_, oops wrong process :) [16:02] ah [16:03] rodrigo_, here's evolution one - http://paste.ubuntu.com/590288/ [16:03] #1 0xb71f10b9 in __lll_lock_wait () from /lib/i386-linux-gnu/libpthread.so.0 [16:03] awesome [16:04] rodrigo_, killing evolution resumes session [16:04] so yeah, some thread seems to be locking and not releasing the lock, and the main thread just waiting [16:04] rye, do you have any remote mail/calendar/addressbook? [16:05] rodrigo_, not there [16:05] rodrigo_, but the question is - why does it block completely the session, i.e. panels, keyboard interaction etc? [16:05] Thread 6 is 'polling', so I guess that's the one that is waiting for some input [16:05] rodrigo_, to a casual user that would have looked as a locked-up machine [16:05] rye, I'd say it's because of out-of-memory condition [16:05] rodrigo_, no [16:06] rodrigo_, i can ssh to that machine and it has plenty of ram still and i am running gdb there [16:07] rye, ok, then what is it? [16:07] rodrigo_, http://paste.ubuntu.com/590289/ [16:07] karni, Exception in thread "main" java.lang.NullPointerException at com.android.sdklib.internal.project.ProjectCreator.updateProject(ProjectCreator.java:551) Woot! [16:08]