=== soren_ is now known as soren [10:16] partman-auto: cjwatson * r255 ubuntu/ (debian/changelog lib/auto-shared.sh perform_recipe): [10:16] partman-auto: * Ensure that there is always at least one primary partition after [10:16] partman-auto: autopartitioning. An existing primary partition on the disk satisfies [10:16] partman-auto: this, as does a recipe partition explicitly marked $primary{ }; if [10:16] partman-auto: neither of those exists, then arbitrarily promote the first recipe [10:16] partman-auto: partition to be a primary partition. [10:18] partman-auto-lvm: cjwatson * r199 partman-auto-lvm/ (debian/changelog lib/auto-lvm.sh): [10:18] partman-auto-lvm: * Ensure that there is always at least one primary partition after [10:18] partman-auto-lvm: autopartitioning, considering only the non-LVM parts of the recipe. [10:18] partman-auto-lvm: Requires partman-auto (>= 73ubuntu4). [10:20] partman-auto: cjwatson * r256 ubuntu/ (debian/changelog recipes/atomic recipes/home recipes/multi): [10:20] partman-auto: * Remove explicit $primary{ } specifiers from standard recipes. We only [10:20] partman-auto: need them to be primary if no other primary partitions exist already. [10:24] partman-auto: cjwatson * r257 ubuntu/ (2 files in 2 dirs): [10:24] partman-auto: * Relax auto-resize constraints now that a primary partition is not [10:24] partman-auto: unconditionally required; this will allow auto-resize partitioning on [10:24] partman-auto: systems that already have three primary partitions (LP: #134950). === cjwatson_ is now known as cjwatson [15:31] I noticed you are short of space on the CD, I can squeze ~700K off wubi since now debugging is on [15:32] unfortunately last time I turned it off I had a crash in vista (which I cannot reproduce when debugging symbols are in) [15:32] I will give another go tonight and see what happens [15:32] that would be great if it can be made safe [15:33] I will try my best, but debugging without symbols is a major pita... [15:33] Hopefully some of the other changes since last time "addressed the issue" [15:33] but I am usually not that lucky... [16:26] The netinstall directory has a file "splash.rle" that's a simple ubuntu splash graphic. It's shown to clients. [16:26] How can I edit this file or make a new one? What format is it? [16:29] it's syslinux lss16 format [16:29] best documentation for creating a new one is probably in build/boot/x86/pics/polverini_b.README in the debian-installer source package [16:30] though the colour assigned to palette index 7 may vary [16:30] Brilliant, thanks cjwatson [18:12] michael_e_brown_: While the install in noninteractive mode completes, something is preventing oem-config from getting installed. I'm looking into it and will keep you posted. [19:08] evand, ping. Tried with todays daily CD and it gets further, but I get the same ValueError [19:09] assuming you didn't see this as well: [19:09] 13:12:41 < evand> michael_e_brown_: While the install in noninteractive mode completes, something is preventing oem-config from getting installed. I'm looking into it and will keep you posted. [19:09] ah. yeah, that would be my home account [19:09] mebrown: can you post logs? [19:09] havent seen it since I havent been home in the past couple days [19:10] ah, ok [19:10] I'm looking into it... I might not have yesterdays daily... (I was told it has been updated, but /.disk/info says 20080303, which seems wrong [19:10] oh! [19:10] I should note [19:10] there were a few CDs today [19:11] only the most recent works [19:11] as the previous ones didn't have live filesystem updates [19:11] huh [19:11] ok [19:11] can you give me a URL and I'll manually pull down and update? [19:11] bbiam... [19:12] http://cdimage.ubuntu.com/daily-live/20080305.3/ [20:31] ubiquity: evand * r2540 ubiquity/ (3 files in 2 dirs): * Fix the --no-continue option. [22:18] good news: can make wubi 700K smaller, and do not get any crash :) [22:18] evand will upload soon [22:19] xivulon: great! [22:20] did the crash just stop occuring or did you find the root cause? [22:20] pure luck, some of the other changes might have fixed it [22:21] I'll keep an eye on user feedback anyway and ask users to check well the new build [22:21] ok [22:22] tonight I am also uploading the umenu pot... Last chance for text changes... [22:22] evand: You know there is a way to gather the metadata needed to get dmraid to sdupport that ATI controller? I'd have to dig it up, but its possible to do, and its then a matter of sending upstream, which can be then used to write supporting code. [22:23] TheMuso: ooo, that I did not. I'll definitely have to look into that. [22:29] xivulon: I'd suggest going with the ones suggested on the wiki, except for the bit about WinFOSS applications unless you can convince slangasek and henrik that we have enough time to persue that. [22:32] Do you mean proposal 3 in the umenu wiki? [22:32] That does notmap correctly since it does have 4 bullets, and am not sure the last one applies [22:32] yes [22:33] do you want to change the english text yourself? [22:33] then I will pull [22:33] will do [22:33] thanks [22:35] What is the windows installer written in if I may ask? [22:36] I pulled the source the other day, and saw it was some form of scripting... [22:43] evand, looks like noninteractive is going now. gotten way past where it failed before. [22:43] TheMuso that is nsis [22:43] and calling it scripting is a big word [22:43] plugins are mostly in c/c++ [22:43] xivulon: Ah ok, showing my ignorance clearly. :) [22:45] nsis? [22:46] http://nsis.sourceforge.net/Main_Page [22:46] was not my choice... [22:46] Ah. [22:46] evand, percent complete is a bit wonky, though. Goes from 99% complete copying files back to 75% complete configuring target system. [22:47] I took over from the original interface and kept adding... Would have been much wiser to migrate everything to python/venster... [22:47] mebrown: yeah, the display needs a bit of polish. I have some changes in my local branch. [22:47] evand, any way to skip "getting time from network time server?" [22:48] also, not properly clearing the line when it scrolls [22:48] doh. [22:48] AttributeError: [22:48] Wizard instance has no attribute 'reboot' [22:48] clearing line> yeah, that's one of the things I have a fix for in my local branch [22:49] hrmm [22:49] also doesnt appear to have run my scripts [22:49] checking [22:49] some weird behaviour, not sure what to make of this.... [22:50] It gives the attribute error (noininteractive.py, line 122 in progress_loop) [22:50] Then, it displays this: [22:50] 0 percent complete: Detecting file systems... [22:50] No root file system: No root file system is defined. [22:50] Please correct this from the partitioning menu [22:51] Bad password: You entered an empty password, which is not allowed, please choose a non-empty password. [22:51] Password: [22:51] re time server> I think preseeding clock-setup/ntp false will fix that [22:52] hrm, perhaps it called return_to_partitioning [22:53] (not a huge deal, just sits there for about 10 secs, which I wanted to eliminate.) [22:53] biggest problem is that it does not appear to have run my post-install scripts [22:58] ah, found the problems, I believe. One moment. [23:10] ubiquity: evand * r2541 ubiquity/ (debian/changelog ubiquity/frontend/noninteractive.py): [23:10] ubiquity: * Properly handle success_command and rebooting in the noninteractive [23:10] ubiquity: frontend. [23:12] mebrown: there's some more changes I want to do here, but I'll let you know when that lands on a CD. You might be able to get around it for the time being by removing ubiquity/reboot from your preseed file. [23:12] will that fix the issue with it not running my scripts? [23:14] oh, no it wont. The only way to do that is to modify the code. [23:14] in the initramfs, before ubiquity runs, that is. [23:15] but you know what the problem is? [23:15] how long until it hits a daily CD? [23:15] yes, I forgot a call to run_success_cmd in the noninteractive mode [23:16] hrm [23:17] I can try to get it in for tomorrow's CDs, but I'm not sure how sucessful I'll be as we're in a freeze leading up to Alpha 6. [23:17] I'm working on getting our factory install revamped to use the noninteractive feature and a few other things. [23:18] ok [23:18] how long is the freeze, as I am trying to get something workable this week [23:18] so I can move on to some other improvements [23:19] If you could let me know, I'd appreciate it. Time for me to go home for today. [23:19] it's until the Alpha is released, which should be tomorrow. I'll do my best to get this in asap though. [23:20] if it is delayed until Friday, it wont be the end of the world for me. [23:20] fwiw [23:20] I was just afraid if the freeze is for another week, that is problematic [23:20] ah, understood [23:20] just keep me informed and I'll be happy. [23:20] will do :) [23:21] thanks. [23:25] evand rev 449 for wubi [23:30] evand: I think https://bugs.edge.launchpad.net/ubuntu/+source/dmraid/+bug/112402/comments/14 has the info you need to at least try and collect the needed data to be sent upstream. [23:30] Launchpad bug 112402 in dmraid "Fasttrack Promise not recognized: "No RAID disks"" [Medium,In progress] [23:31] xivulon: thanks [23:31] TheMuso: thanks [23:31] evand: np [23:32] However, that could only be for Promis controllers... In general though, the metadata is somewhere at the end of the disk. [23:32] I think its just a matter of having to search around for it... [23:45] evand the 449 build should be 1.3M as opposed to 2.1 [23:46] xivulon: committed my attempt at polishing the wording on the options for umenu. You might want to have another person look over that as well though. [23:46] you might want to run make clean first [23:46] ah thanks [23:50] not sure about "Reboot assistance" [23:51] that is basically the title of the page with the options: reboot now, reboot later, boot helper [23:51] yeah, me either, but at the same time I'm not sure about "boot into a new world" [23:51] boot helper works for me [23:51] suggestions welcome anyone... [23:52] "reboot required" maybe? [23:54] oh, I misread completely [23:54] yes, that works [23:57] hmm at least in wine there is not enough space to display option 2 [23:57] xivulon: I seem to be missing Hungarian.nsh when building wubi. [23:58] argh [23:58] yes I took it out as explained in the changelog [23:58] since it was creating problems with encodings [23:58] but it should compile anyway [23:58] at least it does on mine [23:58] or better I renamed the po files to hu.po.ignore [23:59] one sec