=== cjwatson1 is now known as cjwatson === Nicke_ is now known as Nicke [11:13] ubiquity: cjwatson * r3048 ubiquity/debian/ (changelog init): Add LSB init script header. === evand_ is now known as evand [15:03] The user-setup change post ubiquity 1.11.11 is breaking the user setup page. Looking into it now. [15:03] oh, whoops [15:03] sorry [15:03] I was trying to cram it in pre-FF [15:04] and didn't have an up-to-date desktop CD available :( [15:04] no worries [15:05] any work on user-setup changes for kde4? [15:05] *word [15:06] wasn't feature freeze critical :) [15:06] I don't think I saw the merge request, though, odd [15:07] oh, you proposed it for the wrong branch [15:07] can you propose it for lp:~ubuntu-core-dev/user-setup/ubuntu rather than for lp:user-setup? [15:07] sure [15:07] the latter is the Debian branch [15:08] gotcha [15:09] who do I set as reviewer? [15:10] me [15:10] k [15:10] or leave it blank if you can [15:10] yea..I can leave it blank [15:11] ok, done [15:11] eek, you did a debconf-updatepo in there [15:12] :( [15:12] conflict city [15:12] oh...sorry [15:13] I made a deb to test with [15:13] so whatever debuild did [15:13] it doesn't run debconf-updatepo by default ... [15:13] I thought I committed before that [15:13] hmm [15:13] well, the only change was to the user-setup-apply script... [15:13] at least I thought... [15:14] bzr log -v [15:14] oh! [15:14] you branched from upstream [15:14] I can't merge this, you'll have to start from the Ubuntu branch [15:14] ahhhh [15:14] ok..no prob [15:14] sorry about that [15:14] is that the ~ubuntu-core-dev...etc? [15:15] yeah [15:15] k [15:15] at some point we'll have lp:ubuntu/+trunk/user-setup or something like that [15:15] k [15:15] yea..the same name multiple places was misleading [15:15] in general lp:foo is the furthest upstream in existence for foo [15:16] I see [15:20] ok..I proposed the merge [15:21] hopefully the correct way this time :) [15:43] shtylman: lp:~shtylman/user-setup/user-setup.ubuntu seems to be identical to lp:~ubuntu-core-dev/user-setup/ubuntu. Did you forget to push? [15:45] cjwatson: oops...I wasn't bound to my branch so my push didn't do anything cause I had no local commit [15:46] I have pushed again now [15:46] do I need to request a new merge? [15:47] nope [15:47] k [15:49] that looks better [15:49] :) [15:49] care to give me a changelog entry too? [15:49] um...sure...how do I do that? [15:49] just modify the changelog by hand? [15:49] dch -iU [15:50] er, sorry, just dch -i [15:50] indeed if you had done that at the same time as your original commit (don't bother redoing it now, just for reference for the next time), you can use 'debcommit' and the changelog message will be reused as the commit messages [15:50] message [15:50] which nine times out of ten is what you want [15:51] ooo...cool [15:51] I will keep that in mind [15:51] dch -v$whatever -DUNRELEASED no? [15:51] oh, user-setup [15:51] nevermind me [15:52] do you not use DEBCHANGE_RELEASE_HEURISTIC=changelog ? [15:52] saves pratting around with -DUNRELEASED [15:52] shtylman: yes, do set UNRELEASED instead of jaunty in the first line of the changelog too, avoids confusion [15:52] this is a lot easier the second time :) [15:53] cjwatson: too late :( I pushed the change [15:53] can I unset it? [15:53] I'll fix it up at my end [15:53] k [15:53] cjwatson: ah wonderful, I was missing that. [15:54] thanks [15:55] user-setup: cjwatson * r158 ubuntu/ (debian/changelog user-setup-apply): merge from shtylman [15:58] user-setup: cjwatson * r159 ubuntu/debian/changelog: releasing version 1.23ubuntu13 [15:58] shtylman: all done, thanks [15:58] cjwatson: cool, thanks for the guidance [15:58] cjwatson: might it be possible that the installer is checking for the security of passwords now? I used to preseed "ubuntu" but that seems to stall the installation now [15:59] cr3: yes. if you want to override it, user-setup/password-weak=true [15:59] oh, hmm, that isn't preseedable [15:59] can you just use a different password for the moment? :-) [15:59] cjwatson: would it make a difference if I provided a hash instead of a cleartext password? [15:59] it's difficult to make that preseedable without impairing interactive behaviour [16:00] cr3: yes, that would work [16:00] cjwatson: excellent, that will be backward compatible too so I can change all my preseeds the same way [16:00] right [16:00] sorry about that, hadn't occurred to me that you would be relying on that [16:01] how do you guys preseed the installer? [16:02] https://help.ubuntu.com/8.10/installation-guide/i386/automatic-install.html https://help.ubuntu.com/8.10/installation-guide/i386/appendix-preseed.html [16:03] ahh thanks [16:15] cjwatson: when did you sneak the password strength thing into d-i? [16:15] debian-installer: cjwatson * r1048 ubuntu/ (4 files in 2 dirs): [16:15] debian-installer: Remove console-setup from armel network-console initrds, per Oliver [16:15] debian-installer: Grawert. [16:15] davmor2: late last night [16:15] :) [16:16] I didn't think it was in yesterday :) [16:25] ubiquity: evand * r3050 ubiquity/ (debian/changelog ubiquity/components/usersetup.py): [16:25] ubiquity: Give the proper return value in the user-setup component's run [16:25] ubiquity: function. [16:33] Yay ltsp works again :) [16:33] well builds [17:10] evand: hmm, if you do that then I think you have to return True in the if branch [17:10] since question_dialog has already waited for input so you don't need to do so again [17:10] unless I'm completely forgetting how this stuff works [17:10] I originally had that, but thought it was superfluous [17:11] hrm [17:11] ok [17:12] well, seems to me that if you don't then it will always go into enter_ui_loop after the question dialog is dismissed [17:12] whereas if the user says no-I-want-to-choose-again then user-setup will go back to the previous password prompt and that will take care of that [17:14] ok, that's not what I was seeing, but I trust your reading of the code more. Perhaps I missed something in my observations. [17:15] ubiquity: evand * r3051 ubiquity/ubiquity/components/usersetup.py: Properly return from user-setup's run function (again). [17:15] obviously if 'return True' in the if branch actually breaks then that supersedes my opinion ;-) [17:16] I had tested both, so if it breaks again then I've clearly lost it :) [17:17] I'll port that to oem-config too [17:17] ah, good call [17:18] oem-config: cjwatson * r616 trunk/ (debian/changelog lib/components/user.py): Give the proper return value in the user component's run function. [17:22] ubiquity: evand * r3052 ubiquity/ (5 files in 3 dirs): [17:22] ubiquity: * Restore the OS names in partition bars label code that was accidentally [17:22] ubiquity: commented out in a previous commit. [17:22] ubiquity: * Fix the code around showing ubiquity/text/part_auto_comment_*. [17:58] cjwatson: Do you know if the python version of wubi is in? [17:59] no idea, sorry [18:00] cjwatson: np's :) === fader is now known as fader|lunch === fader|lunch is now known as fader [19:08] cjwatson, evand: does the live cd have an openssh server package installed? to push debs to it? or does that have to be installed once up and running? [23:14] wubi: Agostino Russo * r66 trunk/ (4 files in 2 dirs): [23:14] wubi: * Fixed regression, a premature deletion of the temporary folder was [23:14] wubi: preventing a correct startup [23:14] wubi: * Factored out code that looks for archive start === TheMuso_ is now known as TheMuso