[09:10] morning all [10:42] Noskcaj: that would be awesome [13:25] wich iso should i download to test? daily or final? [13:39] final [13:40] ok [13:58] afternoon all [13:59] hello slickymaster [13:59] hello elfy, how's your chest infection? [13:59] not too good - but nvm :) [14:00] elfy: I trust you saw a doctor and are under treatment [14:00] of course ... [14:01] elfy: well you'll have to wait until the antibiotics start to do their job [14:01] slickymaster: I'm not world renowned for doing what I should do ;) [14:02] but it's all o/t in here anyway - better in -offtopic :) [14:02] elfy: I know what you mean. I do the talk, when it concerns others, but rarely do the walk, when it concerns me ;) [14:03] :) [15:06] i'm not seeing any progress bar while the installation goes [15:06] have you filed a bug? [15:08] no, i dont know if its a bug [15:09] http://i.imgur.com/5twUZIu.png [15:11] knome, ^ is that correct? [15:11] no [15:13] i assume i need to create a bug report agains ubiquity [15:14] yes [15:14] ok [15:22] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1240117 [15:22] Ubuntu bug 1240117 in ubiquity (Ubuntu) "ubiquity installed doesnt show a progress bar while installing on saucy RC" [Undecided,New] [15:26] its there any particular thing needed to check? [15:26] GridCube: I'll see if I can replicate that [15:26] did you try resizing the window at all though [15:27] GridCube: it looks like the progress bar has been shoved off the bottom because the window shouldn't have a title bar and scrollbars [15:27] yes [15:28] i tried to resize but it only allowed me to move the window [15:28] do the scrollbars work? [15:28] yes [15:28] it just shows the same image a little more or less [15:29] ok - I'll look [15:29] no that I can do anything but confirm or not :) [15:31] GridCube: in a vm here with 2gb ram - can see progress bar ok [15:32] http://imagebin.org/273732 [15:35] did you encountered the kernel error before? [15:35] and thats the installer from the desktop live session [15:35] Bug 1220165 [15:35] bug 1220165 in parted (Ubuntu) "Error informing the kernel about modificatons" [Critical,Fix released] https://launchpad.net/bugs/1220165 [15:35] that bug GridCube ^^ [15:35] yes that one [15:36] I have seen it of late - but not today [15:36] ok, my bug seems related to that [15:36] 'should' have been fixed with todays rebuild [15:36] i'm using the iso from the "final" not "daily" [15:38] yep - it gor rebuilt - did you get it today - I got notification at 10:56UTC [15:40] just now [15:40] mmm - not sure then - it's ok here [15:40] D: wait [15:41] i though the szync was correct but it actually fetched 79% of the iso [15:42] great [15:42] i will delete the results :/ [15:42] i was testing the beta2 image all again [15:43] oh [15:44] how do i cancel the bug report aswell? [15:45] GridCube, first check if it happens with the final images; if not, then mark it as fix released and leave note in the comments that it doesn't appear with the final images anymore and that your previous test was with an older image [15:46] GridCube: I see we were marking it invalid at the same time then [15:46] i marked it invalid myself :/ [15:46] sorry [15:46] marking as invalid is fine as well [15:46] GridCube: I know - I was doing the same thing [15:49] ok [15:49] i will test tomorrow then :( i lost a lot of time doing this one and i have to get some work done [15:50] GridCube: I expect you'll need an up to date one again tomorrow [15:50] from the same final folder? [15:50] or from daily? [15:51] same place - Final [15:51] in any case, it wont be the same as 50% difference from beta2 [15:51] :) [15:51] GridCube: whatever is here http://iso.qa.ubuntu.com/qatracker/milestones/305/builds [15:57] well let's hope it's more than 50% same than todays image. [15:57] lol [16:03] haha [18:15] ali1234: did you already take a look at bug 1185396? maybe even done some debugging? [18:15] bug 1185396 in gnome-system-tools (Ubuntu) "users-admin crashed with SIGSEGV in gst_user_profiles_get_for_user()" [High,Triaged] https://launchpad.net/bugs/1185396 [18:15] i had a look but it's gnome [18:17] it's all plugins inside plugins and i have no idea how you are even supposed to run it let alone debug it [18:17] what i did find is that it crashes at the same time as the user creation actually happens [18:18] and the new user does seem to get created correctly [18:18] it still uses the normal gtk stuff [18:18] and the segfault when you close the application? [18:18] yes [18:18] only if you made a user though [18:19] right [18:19] something corrupts the list which contains all the groups [18:19] where is it in the source even? [18:20] what exactly? [18:21] the invalid function call? [18:21] well, the list with the groups i guess [18:22] groups-table.c [18:22] this does not use GSlist [18:23] not directly anyway [18:23] it's GtkListStore [18:23] it's some sort of wrapper [18:23] static GtkListStore *groups_model = NULL; [18:23] that one? [18:24] yes, and gtk_list_store_clear (groups_model) fails when terminating the app [18:27] hmm [18:27] fishy fishy code [18:28] after it sets up the tree model it calls g_object_unref(groups_model) [18:28] it seems to me that using groups_model after unref'ing it would be a bad idea [18:28] i don't know though... need to read the docs [18:28] unless the model gets copied [18:29] *if* the model gets copied [18:29] unref just decrements the reference count [18:29] when it hits clear it is free'd [18:29] *zero [18:30] if gtk changed so that the object was copied that would make this code stop working [18:30] since there would no longer be another reference to keep the object alive [18:31] i doubt that is the problem though [18:32] mmh, but there is no problem with the model until you add new user [18:32] which also adds a new group [18:32] http://stackoverflow.com/questions/2848273/should-a-g-object-new-have-a-matching-g-object-unref [18:32] deleting a user/group does not cause any notable trouble, but adding does [18:33] if i read that correctly... the unref's are completely unnecessary [18:33] and will cause the object to be freed prematurely [18:34] wouldn't that kill the app way sooner? [18:36] Soooo, why is it when you boot an installed system in Vbox it'll use plymouth-theme-xubuntu-logo, but booting the ISO it'll use plymouth-theme-xubuntu-text? [18:36] Unit193: nvidia graphics? [18:37] vbox, so does it matter much? Yes. [18:37] thad: no. reading freed memory works fine. writing to it as well, as long as nothing else started using it. extending it however [18:37] I see [18:38] Unit193: actually no. i was thinking how plymouth doesn't work with nvidia driver, but that shouldn't affect the ISO [18:40] still segfaulting [18:42] thad ali1234 - I did some checking on that user admin thing - the ONLY thing I've crashed it with is adding a user [18:42] yes, right [18:42] does it crash in unity? [18:42] no idea ali1234 [18:42] or gnome classic? [18:42] I can have a look [18:42] or anything that isn't xfce? [18:43] (does unity even use it, or something different?) [18:43] would that really make a difference? [18:43] i dunno [18:43] users-admin is not maintained anymore [18:44] ali1234: I can check if you want me too [18:44] ali1234: Yeah, no idea why it doesn't use the logo boot. [18:45] only xubuntu/ubuntustudio/lubuntu depend on it [18:45] hmm ok [18:46] can i hack this so it doesn't actually make users, but still crashes? [18:49] I didn't try this yet [18:51] hmmmmmmmmm [18:51] i think i know what is happening [18:54] the "create new user" function fails? [18:55] adding a group manually works without causing a segfault afterwards [18:57] yeah it crashed when clear is called from gst_users_tool_finalize (GObject *object) (which is in another file) [18:58] exactly [18:59] if i rearrange the calls so it clears the models first then it crashes in a different place :/ [18:59] i think manually clearing and unreffing everything is the problem here [19:01] what about "** (users-admin:14123): CRITICAL **: gst_user_profiles_apply: assertion 'profile != NULL' failed" ? [19:01] this one appears when creating a new user [19:02] it's a double free issue [19:02] the thing has already been cleared/freed [19:02] if i comment out the call entirely it doesn't crash [19:02] i don't know for sure what the consequences of that are though [19:02] but it also does not complain [19:03] right [19:03] well, the app is closed anyway :) [19:03] also i still don't understand why it only crashes after a user was added [19:03] it does cause a few extra calls... maybe that is the problem [19:04] the groups_list gets created and destroyed a lot during program lifecycle [19:04] probably the function to create a new user [19:04] fishy fishy code [19:04] and the critical error [19:04] but i don't feel comfortable changing stuff that is security related [19:05] mmh, identifying the culprit does not cause any harm [19:06] and you can test it inside a virtual machine [19:13] i also don't understand why the exact same thing doesn't appen with the users table [19:13] the code is identical [19:15] it does complain [19:15] (users-admin:25934): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed [19:17] does the error appear when creating a new user? [19:18] I see this one -> ** (users-admin:14951): CRITICAL **: gst_user_profiles_apply: assertion 'profile != NULL' failed [20:28] good night all [22:35] any xubuntu release manager online? [22:35] hi doko, what's up? [22:35] doko, heya. [22:35] oh good, knome is here [22:35] :) [22:35] https://launchpad.net/ubuntu/saucy/+queue?queue_state=1 [22:36] doko, saw that discussion; is there any actual code change? [22:36] no, just the config.* change. see the diff [22:36] doko, i'm fine with those [22:37] knome, cool, could you repeat that on -release? [22:37] done [22:38] thanks, hopefully these were the last ones [22:38] :)