[00:18] dobey, are you running a window manager inside your vm? [01:02] Hello everybody. I was thinking to join quality team as a tester and contribute little bit. My knowledge about Linux is very limited. This is actually my project for one of my classes at school. Our assignment is to join an open source community, participate and contribute. [01:03] kris_: excellent. welcome! have you seen the wiki pages already? there's a lot of detail there! [01:04] Yes, I have. [01:04] kris_: are there any questions about how the process works or where to get started? [01:08] I have been reading the wiki pages, and I can’t imagine how can I contribute. It is little bit overwhelming. So yes, what would be my first step? [01:08] kris_: so i assume, by testing, that you want to help with testing the installation images, correct? [01:09] Actually Yes, I was looking at the ISO tracker walkthrough. [01:11] ok, so basic first steps: [01:11] 1. make sure you have a launchpad account [01:11] 2. join the ubuntu-quality team on launchpad [01:11] 3. sign up for the mailing list [01:11] 4. hang out here [01:11] once you got that out of the way, then we can move on to the actual work :) [01:12] I have setup a Launchpad account already. So let me join the quality team and mailing list. [01:13] let me know when you're done [01:17] kris_: i'm out of here in 10 minutes so i'm going to break it down for you: [01:18] when it comes to ISO testing, we have sort of three different ISOs that may get tested at any particular time (tho not necessarily all of them): [01:18] 1. daily images [01:18] 2. milestone images [01:18] 3. LTS point release images which are milestones, technically, but good to point out as they're on their own cycle [01:19] so daily images represent the current state of development. both LTS and non-LTS versions have them (so right now, there's a vivid daily and a trusty daily) [01:20] milestone images are release images though not necessarily just the final release. there's alpha 1 & 2, beta 1 & 2, and ultimately the final release [01:20] LTS images have several point releases that collect updates and include support for newer hardware. like i said, they're a sort of milestone [01:21] if you look at the release schedule, the milestones always come out on thursday. the testing on those images come out the tuesday before. right now, there's nothing going on, though the 2nd point release of 14.04 (trusty) will come out next thursday. [01:21] that being said, i would focus your effort right now on daily images (specifically vivid daily) [01:22] i would suggest doing your tests on a virtual machine [01:22] you can do it on real hardware, but ultimately, it will wipe everything out, so it will need to be a spare machine [01:22] i would suggest virtualbox for this pupose [01:22] so you'll want to install that [01:23] then go to the ISO tracker, then vivid daily, then pick an image. if you're only interested in ubuntu, go to ubuntu desktop. if you want to help out with other flavors like kubuntu, lubuntu, xubuntu, etc., tehy're there too [01:23] make sure you choose the image that reflects your machine's architecture [01:23] as a general rule, i386 will work no matter what [01:23] when you click on it, there will be a link to download info [01:23] download it [01:24] then make a new ubuntu/linux virtual machine (default settings should be fine) and in the storage tab of the settings on that machine, add the iso to the empty ide drive [01:24] then boot the image [01:24] at that point you follow the test cases [01:25] if you find a bug in your testcase, file it with `ubuntu-bug ` [01:25] then include the number of the bug in your report [01:25] at the end, if the test case succeeds, then mark it a success. otherwise it's a failure. [01:25] then do some more images [01:25] and then do the same thing the next day and so on [01:25] eventually you run out of work. at that point you can help with bug triage :) [01:26] ok, i'm out [01:26] good luck kris :) [01:26] thank you [01:26] I have joined the group and mailing list. [01:27] If I run into any problem, I'll come back here, thanks again. [01:40] awesome kris_. Basically https://wiki.ubuntu.com/QATeam/Roles/Tester is everything you need to know :-) [01:42] Yes, I'm reading it right now. Lots of information, very helpful. === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === elfy is now known as hobgoblin === hobgoblin is now known as elfy === chihchun_afk is now known as chihchun === qwebirc564330 is now known as slickymasterWork === pgraner-afk is now known as pgraner [14:07] alesage: hi. So did you review my branch? [14:07] alesage: Just want to make sure that it is good enough to be used in the power indicator testings. [14:18] balloons: i think the ubuntu-touch-session setup script for adt-run doesn't start one [14:42] good morning. [14:45] morning elopio [14:46] hello dobey. Last I read is that your problem is only with one button, right? [14:46] elopio: yes [14:49] dobey, I find running a window manager is required in order for some things to work correctly. Not sure if this is a case of it or not, just tossing it out [14:50] balloons: i'll try making it start unity8 then [14:50] you are running under mir? [14:51] no [14:54] dobey: have you pushed to your branch with the instructions to create the qemu? [14:54] I need to copy somethings you have to the webapps tests. I might as well give it a try and see if I can find something related to the button problem. [14:55] elopio: no, i don't remember how i created the images [14:55] dobey, if not, I use something like ratpoison then [14:56] elopio: i think maybe it was adt-buildvm-ubuntu-cloud [14:56] basically something that autofocuses the app on launch [14:57] balloons: having no wm means whatever is on top of the stack has focus. and focus is surely not the issue as other things in the window can be clicked [14:57] "adt-buildvm-ubuntu-cloud -r vivid" i think is how i built the vivid image [14:59] dobey, just passing along that I had issues in the past.. my issues where with clock app actually, and noticing a qml property was never set unless run under some sort of wm [15:02] sure, i don't know what your specific issue was, but it sounds different [15:03] definitely was.. just trying to empty the brain of all the quirks I've ever experienced and pass them along ;) [15:05] hmm, i guess i can't start unity8 [15:06] will try unity7 then === chihchun is now known as chihchun_afk [15:16] hrmm, unity7 won't start, probably due to video of course [15:22] paulliu, sidetracked by other stuff last evening, will look at asap [15:24] whee and sometimes i don't even get a screenshot or such with the 'failure' [15:31] alexabreu: ping. I have three things to talk to you about your branch. [15:31] elopio, sure [15:32] alexabreu: first, we need it to be python3. So on your dpkt-architecture call, you will need to add the argument universal_newlines=True. [15:32] to make it return a string. [15:32] also, that means we need the PPA thomi started, fgimenez is giving it the last touches. [15:32] elopio, yes (for python3) but I made it python2 bc of selenium, which was python2 afaik [15:34] elopio, ah no my mistake, there is a python3 version, so I'll fix that for sure [15:35] alexabreu: second thing, the only way I got it to work when using launch_click_package was to use initctl set-env for dev tools host and port. So I think that we need different ways to set the env vars depending on how we launch the package. [15:38] alexabreu: and last thing we need to discuss is how to get this tests running. If we put them in a debian package, we can tell ci to configure it for us using the selenium ppa. [15:38] elopio: alexabreu a first version of the ppa is ready for testing at https://launchpad.net/~canonical-platform-qa/+archive/ubuntu/selenium [15:39] alexabreu: another nice alternative is to split the tests, one per click package. On the manifest we specify the tests to run, and then they are runnable with adt-run using qemu as a testbed. [15:40] according to dobey's recent adventures, it's likely that if we go the adt-run way, we'll hit some problems because it's mostly unexplored. [15:41] fgimenez: thank yyou. I will give it a try. [15:42] experimental selenium packages? elopio fgimenez what do you have up your sleeves? [15:42] thanks for the info in the crd fgimenez. That quilt push -a was the command I missed. [15:43] elopio: maybe you should just have the EnvironmentVariable fixture set the variable in the upstart and dbus sessions [15:43] elopio, could we control when the tests are being run? or do they have to be run manually? [15:44] elopio, regarding the envvars, any suggestion? [15:45] alexabreu: if we go the adt way, it's going to be manual because ci is not yet ready to run those tests. But there are some good points for this: it's the future and ci is going to get it working for merge proposals soon, and by documenting how you create the qemu, you will have an environment that really easy to reproduce, so everybody should always get the same results. [15:45] elopio: it's the quick solution, probably it would be better to change the source format and have the patches automanaged [15:46] elopio: but that means more changes to the package itself [15:46] alexabreu: about the env vars, my suggestion is to try to always launch the click package and set the initctl env vars, but that would mean to either install the click package or to make a test desktop file to launch it. [15:47] if we go the qemu way, adt-run takes care of installing it. [15:47] for the test desktop, I need to give it a try. I'm not sure if it'll work. I'll be playing with your branch today. [15:48] elopio, ok, feel free to MR something [15:49] fgimenez: I understand that. I would love if we could update the package in debian, so the patch won't be needed. But that's something to discuss for the next sprint, it might take us too far from our goal. For now, your quick solution is perfect. [15:49] I'll give it a try after walking the dog. bbs. === chihchun_afk is now known as chihchun [15:58] well [15:58] i got metacity working and qmlscene can actually do maximized now, but the click still fails [16:12] ok, so i changed the resolution of the xvfb to 540x960, and hav metacity running, and the qml is maximized, and it still times out === chihchun is now known as chihchun_afk [16:19] anyway, i guess i should go get lunch. [17:10] hrmm [17:10] * dobey wonders who else to bug about this problem [17:11] or if i should just try to land my branch with skipTest added to these tests that fail in this way === pgraner is now known as pgraner-afk [17:46] balloons: I'm working now on updating the manual testcase wiki page - I will add a note about not adding possible test numbers to mp's [17:48] elfy, always seems to confuse folks :-) [17:48] yea - it shouldn't in future :) [17:49] I remember doing exactly the same thing :p [17:57] balloons knome - https://wiki.ubuntu.com/QATeam/DevelopmentToolsUsage [17:58] this page has 'bit's of bzr in it - there are other bits, like adding and commit messages at https://wiki.ubuntu.com/QATeam/ContributingTestcases/Manual/Writing [17:59] which I'm looking at now - should I add add & commit to the tools usage page and remove from the writing one? [18:00] elfy, probably so because that page can be reused [18:01] (without having looked at the pages just right now) [18:01] :) [18:06] ok - and yea - I can just clean up the write page and add thing to usage now [18:10] knome, what's the ticket for https://bugs.launchpad.net/ubuntu-qa-website/+bug/1416893? [18:10] Launchpad bug 1416893 in Ubuntu QA Website "cannot login to http://iso.qa.ubuntu.com/" [High,Triaged] [18:10] i don't remember from the top of my head [18:11] apparently it's still an issue; I didn't realize [18:12] bug 1416893 [18:12] bug 1416893 in Ubuntu QA Website "cannot login to http://iso.qa.ubuntu.com/" [High,Triaged] https://launchpad.net/bugs/1416893 [18:12] I hope [18:12] yay [18:12] it's still an issue if the SSO account doesn't have a username associated with it, eg. it's not linked with LP [18:12] elfy, ticket meaning the canonical IS one [18:13] oh sorry [18:18] can't find the ticket now [18:18] maybe pleia2 knows [18:20] interesting.. so people with sso accounts, but never signed up for launchpad? [18:20] yeah, there's a forever ticket, sec [18:20] so if they sign up for launchpad, does it fix the issue or no? [18:21] #24297 [18:21] also, dkessel did you see pleia2's feedback? [18:21] balloons, fixes [18:21] filed in April of 2014, last response I got was in May of 2014 (no fix), I followed up on January 4th again [18:22] only last year? :) [18:22] knome, so is it a matter of a better error message then? [18:22] good evening [18:22] balloons, that would be a start [18:22] the transition from lp to Ubuntu SSO was a serious disaster with no communication with the community, so we have problems all over the place :( [18:22] balloons: yes. i read the part about the list of known bug, and about the hardware URL box [18:23] balloons, but ultimately, it should just allow doing whatever is needed to log in on the fly [18:23] help wiki, loco.ubuntu.com, iso&package tracker are the ones I know of [18:23] might affect other sites too [18:23] what about pad? forum? [18:26] dkessel, yes.. I'm thinking about the hardware link especially.. I know you have a nice idea for the bugs [18:28] balloons: honestly, i don't know if anyone ever used those URLs. I would not have any problems with dropping it ;) [18:29] but i am not part of the target group of that information ;) [18:31] dkessel, I was thinking to do the same [18:32] that is, dropping the field completely. Originally we envisioned a hwdb, but that never really came to pass [18:35] elfy, what do you think? ^ [18:35] infinity: the metapackage issue is not affecting vivid daily is it? [18:38] brendand: should we also skip the wizard test if the sim is not inserted? [18:41] re hardware info - never use it - never ask people to use it :) [18:42] I've set up a xubuntu place for that information for if *we* ever need it [18:43] oops [18:48] elopio, why? [18:48] given Liz's feedback and knowing I've been asked about it before it seems to be mainly a point of confusion. I too, do not use it [18:50] brendand: I'm getting: [18:50] File "/tmp/adt-run.ePQNOn/build.FOM/real-tree/ubuntu_sanity_tests/tests/test_settings_wizard.py", line 120, in test_complete_settings_wizard [18:50] self.assertEqual(sim_inserted, self.is_sim_inserted) [18:50] um, but that shouldn't happen. That's weird [18:54] pleia2: i updated bug 1017207 to note the hardware URL input box should be removed [18:54] bug 1017207 in Ubuntu QA Website "Clarify what a "URL to the hardware profile" is in tracker" [Wishlist,Triaged] https://launchpad.net/bugs/1017207 [18:54] dkessel: hooray :) [18:54] dobey: how are you running your tests? I still get the launching error. [18:54] I'm using adt-run --click-source . --click build/com.canonical.payui_15.01.last_amd64.click -U --apt-pocket proposed --setup-commands ubuntu-touch-session --- qemu /media/elopio/vms/images/adt-vivid-amd64-cloud.img [18:56] elopio, shouldn't that just match the fact of whether the sim inserted screen is shown with the sim status? [18:56] brendand: yes, I'm waiting for the screenshot. Maybe the wrong screen was opened. [18:58] elopio, is_sim_inserted will be true if a sim is inserted in any slot [18:59] it comes with no screenshot, how convenient :/ [18:59] brendand: I'm not able to reproduce your uninstall errors. === [BNC]elfy is now known as elfy [18:59] elopio, i'm not able to not reproduce them :) === pgraner-afk is now known as pgraner [19:00] elopio, i will try and figure out what's happening [19:01] brendand: I left comments on the branch. In theory, I have code to make sure they don't happen. So if you see it again, please paste traceback and screensthot. [19:05] elopio: what revision of the branch do you have? [19:06] elopio: i've added skips for the tests which fail due to the buyButton click failing. and i just pushed a new revision which updates the HACKING file [19:07] dobey: I'll follow the instructions from there. [19:10] dobey: is this expected? http://paste.ubuntu.com/10176933/ [19:12] elopio: yes. add --maint-mode and it should "do the right thing" there. i left --maint-mode out of the HACKING file so it would show up for people who might need to care about that [19:12] i don't think having the one extra package installed will matter for them though [19:13] dobey: got it. [19:13] elopio, alesage - could one of you give this a second review? https://code.launchpad.net/~canonical-platform-qa/ubuntu-sanity-tests/print_using_argparse_conf/+merge/249304 [19:14] dobey: I don't have the right chroot. http://paste.ubuntu.com/10176950/ A step with the command to create it would be useful [19:14] brendand: sure, one second. [19:15] elopio: hmmm [19:17] ah ok [19:19] elopio: done [19:20] elopio: i've finally made an MP for it too: https://code.launchpad.net/~dobey/pay-ui/adt-run-ap/+merge/249387 [19:41] elopio, to confirm - the thing you would like to be fixed is passing the artifacts_directory instead of the config stack? [19:49] elopio, someone accidentally set https://code.launchpad.net/~canonical-platform-qa/ubuntu-sanity-tests/print_using_argparse_conf/+merge/249304 to top-approved [19:51] elopio, if you want i can make another proposal to implement your suggestion [19:58] brendand: well, that's related to the mail I sent to vila with copy to the mailing list. I think I would prefer to unpack the values from the config and pass them as arguments if they are just a few. [19:59] oh what the hell? [19:59] so get artifact directory before calling the methods, and pass the directory as argument. If you like the idea, yes, a branch for it would be nice. [19:59] it will change your tests a little too. [19:59] elopio, i pushed and it went to trunk... [19:59] elopio, bzr did something dumb [19:59] elopio, i'll fix it [20:00] brendand: :) not a big deal, as your branch is actually fixing the issue. [20:00] about the test suggestion, I will try to make an MP so we discuss it there. [20:03] elopio, ok. i just checked i didn't introduce any flake8 issues too [20:05] balloons, hi [20:06] elopio, looks like we are broken again: [20:06] elopio, http://pastebin.ubuntu.com/10177521/ === roadmr is now known as roadmr_afk [20:10] brendand: coudl it be that your image is too old? [20:10] if there is a new one, you should flash it. [20:11] veebers: looking at all the prints for finger movement we are now getting in the log, I think that autopilot's -v should only print info messages. [20:11] -vv should print debug. [20:12] elopio: Sounds reasonable, I totally agree about the prints for finger movement [20:14] elopio: can you file a bug against ap stating that? [20:15] veebers: yes, give me a second. [20:15] and that means that we need to print more info in our tests, which is good. [20:15] no pressure, just when you can. I can't fix it right now anyway ;-) [20:16] dobey: http://paste.ubuntu.com/10177617/ [20:16] you missed the 'create' [20:16] veebers: the only pressure is that I will forget if I don't do it now :) [20:21] https://bugs.launchpad.net/autopilot/+bug/1420949 [20:21] Launchpad bug 1420949 in Autopilot "Autopilot run with -v prints debug information" [Undecided,New] [20:23] elopio, i just updated that's why i'm saying we got broken :) [20:24] brendand: I don't get how the apt list in a newly flashed image can point to non-existing packages. I will reflash. [20:25] elopio: oh, oops indeed [20:40] brendand: agh, same problem here. [20:41] elopio, might be a small window between image builds where we get hit by this [20:41] so it seems we are with no good solution here. If we add the update, we might get unity upgrade. If we don't, we might get 404 [20:41] brendand: we will be hitting this until a new image is made, unless we update. [20:43] for now, and to let us keep working, I suppose adding the -u option for the update is the way to go [20:43] I have no idea what an image tester should do in this case. [20:46] balloons: ok - redone https://wiki.ubuntu.com/QATeam/ContributingTestcases/Manual/Writing and added stuff tohttps://wiki.ubuntu.com/QATeam/DevelopmentToolsUsage [20:58] wxl: you going to get that gnome monitor one merged - or I can I'm wandering about in the tracker atm [20:59] elfy: oh yeah i guess i can do my first merge :) [21:00] don't forget to do the tracker first so you get the number and can rename the testcase :) [21:00] help me out with that, elfy ? [21:01] http://packages.qa.ubuntu.com/qatracker [21:01] then the admin option on the left [21:03] jeez weird that you have to copy/paste everything [21:03] :) [21:03] then new testcase - don't give a number - that's automatic [21:04] once you've got that - you can add the number and merge it [21:04] balloons has a tool in the branch but I can never make it work [21:11] balloons, you're about ? [21:15] he was - he's probably hiding from me again Letozaf_ :) [21:16] or me [21:17] if we're both here - he's definitely going to hide :p === roadmr_afk is now known as roadmr [21:21] elfy: done, but had to make another commit with the pass/fail bit corrected [21:21] elfy, LOL [21:21] noticed that after the fact [21:21] maybe i should have done a merge request but naaaah :) [21:21] elfy, good evening to you :) [21:21] wxl: ok - thanks :) [21:22] Letozaf_: and you too :) [21:22] :) [21:22] Letozaf_: did you get a phone? [21:23] elfy, yes [21:24] elfy, a white one, it's really cool ? [21:25] nice :) [21:25] elfy, did you try to get one on bq site ? [21:26] nah - first of all I've no cash, secondly I'm one of those people who's phone contract is for hardly anything as I text and talk to people [21:27] elfy, :) [21:34] sorry, missed her [21:34] balloons: seems that it's no longer easy to see how to report testcase bugs from tracker [21:34] bug 1420980 [21:34] bug 1420980 in Ubuntu QA Website "Testcase bug reporting now no longer apparent" [Undecided,New] https://launchpad.net/bugs/1420980 [21:35] balloons: not sure she's deliberately left right now [21:38] elfy, nice catch actually [21:41] someone I am talking to in #xubuntu-devel brought it to my attention :) [21:41] xubuntu ftw :D [21:43] elfy, btw the new page looks lovely! [21:43] wiki ones? [21:43] elfy, are you going to go ahead and move the script into the branch also? https://wiki.ubuntu.com/QATeam/ContributingTestcases/ManualStyleGuide/test_case_format_script [21:45] balloons: I was looking at that a short while ago - I assume just more or less the same as the existing tool? [21:46] elfy, not sure what you mean, but we could certainly grab the code and place it inside the branch itself [21:46] it would also be helpful to have a nice README inside the branch; even if you just point people to the wiki to avoid duplicating info [21:46] that's what I meant - wget it then - make it a file with readme things ? [21:46] elfy, yep :-0 [21:47] please remove the instructions from the wiki :P [21:47] ok - I'll make that next balloons [21:47] or at least don't make people go there to read the readme.. [21:47] I usually just run it locally rather than copying to /usr/bin as wekk [21:47] knome: I'd not do that :D [21:47] knome, lol, don't like the idea of linking to the wiki? [21:47] balloons, i don't like to force people to use the wiki if they already download the tool from a branch [21:48] knome, :p I know. The instructions are really simple. Run it and pass your file as the argument [21:49] yes, that too === pgraner is now known as pgraner-afk [22:24] elfy: i think i'm going to copy what xfce did and make a lxde folder in the testcases. i can do that without affecting anything else, right? [22:24] yea [22:24] i don't need to make a merge request, right? :/ [22:25] hmm [22:26] well - probably better to do something - I can approve it as soon as I see it [22:26] wxl: ^^ [22:27] k thx elfy [22:35] elopio, veebers, paulliu asked me to review http://code.launchpad.net/~paulliu/unity8/notification_helper/+merge/249211 , he could probably use your feedback too [22:39] elfy: your turn :) [22:41] not sure why all those bugs are linked grr [22:41] alesage: ack, will push it on the stak [22:41] stack [22:42] wxl: propose it for merging so I can approve [22:42] elfy: https://code.launchpad.net/~ubuntu-testcase/ubuntu-manual-tests/trunk/+merge/249427 [22:42] what's with all the bugs linked? [22:42] i have no idea [22:42] wait a minute [22:43] something is messed up here me thinks [22:43] i'm going to do this over [22:43] ok [22:43] I might have crashed before I see it again - put my name as reviewer and I'll do that as soon as [22:45] elfy: ok NOW it's right [22:45] i did my merge backwards XD [22:47] all done [22:47] thank you sir [22:48] welcome :) [22:49] what *we* did was only have things in xfce which were just us, so things like abiword which could have been used by anyone were in the main [22:49] right right [22:49] so i copied [22:49] made sense to me [22:50] i've used lxterminal on other flavors ;) [22:50] so for instance - gnome system monitor would be where it is now [22:50] yep [22:52] elfy: you might want to move Xubuntu GTK3 Indicators and Power in there, no? [22:52] or Xfce4 Assistive Technologies? [22:52] I might once [22:53] we don't actually do it anymore :p [22:53] ah [22:53] should prolly rm then [22:53] wut? [22:53] if you don't do it anymore, the testcase can be removed? [22:54] no we use them :) [22:54] oh [22:54] then move them in Xfce/ ? [22:54] ormight [22:55] no - xfce was JUST xfce - xubuntu was kept out - it's from the way we used to do our testcases [22:56] well [22:56] there's something xubuntu specific in there [22:56] settings of some kind [22:56] anyway - the long and short of is - we're not looking to fiddle about currently :) [22:56] hahhaa ok [22:57] :) [22:57] theoretically the tracker "should" have a tagging system which allowed anything to appear under anything [22:57] if I was ever likely to do anything it would be empty xfce on the branch [22:57] but practically it's just much more useful to group them to this kind of groups [22:58] "stuff that mostly xubuntu users use" [22:58] yeah [22:58] well [22:58] merge requests welcomed, knome :) [22:58] of course you can put a rhino on an elephant [22:58] hahahahahah [22:58] i mean i'm happy the way we have it now... [22:59] wxl: when knome and I first worked out how we wanted to set it up - this was the easiest and most logical way [22:59] we now don't test in quite the same way [22:59] ah [22:59] so it's less of an issue [22:59] well i kind of like the separation [22:59] it will make it easier for lubuntu folks to know where to help out [22:59] and of course the tracker admin and branch are 2 seperate things [22:59] or rather where to put their stuff [22:59] it's better than a list of all testcases.. [22:59] there is that [23:00] elfy: which, to me, is really silly [23:00] but again, merge requests welcomed XD [23:00] wxl: yep - but these are the tools we have :) [23:00] yuuup [23:01] it's just nice to have someone else watching what goes on in the manual testing code tbh [23:05] night wxl - have fun wondering what to do with the branch :) [23:07] hahah [23:07] night elfy [23:08] balloons, do you know if there's a better remedy for the adt-run 404 errors you mention in your orange notebook post? [23:09] I'm on the latest and greatest but getting some 404s [23:09] :( [23:24] alesage: we need to do the apt-get update, again. [23:25] elopio, is writable-mode necessary for that? and if so how to undo? [23:25] I've made a card for it. I'll implement it now, unless you want to take it. [23:25] alesage: writable is necessary, temporarily. [23:26] umount to undo it. But there's a script that will take care of it for us. [23:30] elopio, umount what? [23:30] alesage: http://anonscm.debian.org/cgit/autopkgtest/autopkgtest.git/commit/?id=a51f26f61c376