[00:21] Could someone point me at a good doc for writing partman modules? I think I understand what I want to do based on code inspection, but I'd rather avoid cargo-culting. [00:23] * persia finds useful stuff in installer/doc/devel and retracts the request [00:25] yes, that's the best I know (and the sgml doc there is in fact very detailed and pretty complete) [00:26] albeit written in somewhat Slavic English :-) [00:27] heh [00:29] Seems like I'm going to have to play some games to make it make sense for non-block devices. [00:32] did you talk to Per? [00:32] he spent months on this in a GSoC project, we shouldn't waste time duplicating that [00:32] I haven't. I'm just trying to find my way around now. [00:34] Oh cool. That device happens to be one of my two test cases :) Doesn't use the filesystem I want, but does provide 90% of the glue. Thanks for the pointer. [01:42] ubiquity: superm1 * r3870 ubiquity/debian/changelog: releasing version 2.1.30 [01:44] i just realized that debcommit decided to stop making tags for me recently. is there some fancy switch that is supposed to do it now on lucid? === robbiew is now known as robbiew_ [03:40] has anyone here installed on a macboook? [03:41] macbuntu11: Yes. What version are you trying to install? [03:42] I've had no problems since installing hardy on my early 2008 model MacBook Pro. [03:43] Themuso: 9.10, installing on macbook 2,1. When i try to load it from reFIT, it just keeps repeating GRUB in an infinite loop [03:43] macbuntu11: have you synced the GPT and MBR tables? [03:45] TheMuso: Not really sure what you are talking about (noob to linux, not my first linux os though) Do you mean doing the partitioning tool on reFIT? I get an error when trying to run it. [03:45] macbuntu11: When you boot, you get icons for OS X and linux. There is a line of icons underneath your OS icons. The second one on the left, i.e the one after the terminal screen icon syncs your tables. [03:45] Yes, that gives me an error [03:47] What error do you get? [03:47] TheMuso: I'm likely going to just reinstall, just checking if there is a quick fix [03:47] hold on [03:49] TheMuso: Ah, just found it on the Ubuntu site "f you see the error "GPT partition of type 'Unknown' found, will not touch this disk" instead of the 'yes' prompt, follow the instructions in comment #6 of this thread and make sure to install gptsync_0.13-10 or higher. " [03:49] macbuntu11: Ok, I've never got an error so this is new to me. [03:51] One more quick question, how do you remove the swap partitions in gparted? On mine, there is a lock next to it making it unable to be deleted [03:52] macbuntu11: Never used gparted sorry, so don't know it enough to help. [03:53] TheMuso: Alright, thanks for your help [03:53] macbuntu11: Welcome. [08:24] superm1: awesome, thanks for the upload! [09:32] Is it possible to resize an existing Windows install during a preseeded (netboot) install? [09:34] We already deploy to our servers over PXE and we're now looking at running dual boot Windows/Ubuntu desktop machines but we'll need to resize the existing Windows system during the install process. [09:39] ubiquity: evand * r3871 ubiquity/ (debian/changelog ubiquity/components/ubi-language.py): [09:39] ubiquity: Do not show the contents of the language page until the strings have [09:39] ubiquity: been substituted and translated. === mpt_ is now known as mpt [10:06] cjwatson: hey sorry, remember I was talking to you about including certain xen modules in the 64bit installer initrd? === ogra_ is now known as ogra [10:50] ubiquity: evand * r3872 ubiquity/ (debian/changelog gui/gtk/stepPartAdvanced.ui): [10:50] ubiquity: Ellipsize the advanced partitioner progress label so that it doesn't [10:50] ubiquity: cause the window to expand. Ideally, we'd also right-align the [10:50] ubiquity: spinner and left-aligned label, but that doesn't appear to be [10:50] ubiquity: possible without a custom composite widget, and putting the spinner [10:50] ubiquity: to the right of a right-aligned label looks ugly. [10:51] sometimes GTK+'s constraints can be annoying, though admittedly this is more me not wanting to have to write said-widget. [11:40] dmarkey: yes [11:46] What is the process to get the ball rolling on that again? [11:55] dmarkey: at the moment I don't have a memory of what needs to be done [11:55] oh, didn't I direct you to the kernel folks for that? [11:55] the kernel packaging controls which modules go into which udeb; I can advise but can't make changes directly [11:55] ubiquity: evand * r3873 ubiquity/ (3 files in 3 dirs): Fix the spacing on the console-setup page (LP: #528695). [11:57] cjwatson: hmm, do you see any problem with including the modules? [11:57] nope [11:58] * cjwatson looks for the last conversation in his logs [11:58] 23:49 anyway, file a bug against the 'linux' package in Ubuntu asking for those to be added [11:59] 23:49 refer them to me for the details if you like [11:59] did you file that bug? [11:59] 23:49 oh, i'll do it tomorrow, bed time for me now [11:59] 23:49 from the arrangement in Debian, xen-blkfront goes in scsi-modules, xen-netfront in nic-modules, and netxen_nic in nic-extra-modules [11:59] is it still possible to get into the next release? [11:59] yes [11:59] did you file that bug? [12:00] nope [12:00] doing it now [12:00] thanks, let me know the bug number and I'll ack it [12:02] ubiquity: evand * r3874 ubiquity/debian/real-po/ (81 files): debconf-updatepo [12:02] michaelforrest: are the slides and new timezone map nearly ready? [12:14] ubiquity: evand * r3875 ubiquity/debian/changelog: releasing version 2.1.31 [12:15] superm1: for what it's worth, tags are working for me [12:25] cjwatson: https://bugs.launchpad.net/ubuntu/+bug/531883 [12:25] Launchpad bug 531883 in ubuntu "Include Xen Modules in 64bit install initrd" [Undecided,New] [12:26] cjwatson: https://bugs.launchpad.net/ubuntu/+source/hw-detect/+bug/304984 [12:26] Launchpad bug 304984 in hw-detect "When using a customised xen kernel(xen modules), xen virtual disks are mis-detected" [Undecided,New] [12:26] that can probably be close [12:26] d [12:32] ev: I haven't had a chance to do that stuff :( [12:33] I could brief Dylan now that the brand has been launched though [12:42] dmarkey: thanks. I've moved the bug to the right place. [12:43] dmarkey: I'd rather leave 304984 open until it's actively confirmed not to happen after these modules are aded [12:43] *added [12:43] but it's your bug, if you insist you can close it yourself [12:46] michaelforrest: please do. Could you CC me on that email as well, please? [12:46] cjwatson: cool, How would one get canonical to officially support ubuntu as a Xen guest [13:00] you'd have to ask our support folks [13:01] but making it work properly would be a fairly major component [13:03] now that pygrub has grub2 support im hoping the next version will work OOTB [13:14] ev: where is dylan based? [13:14] is he in here? [13:14] Vancouver [13:14] and nope [13:15] I don't think he's on IRC during the day [13:15] https://edge.launchpad.net/~dylanmccall [14:39] cjwatson, were you able to look at that blkid issue? only enquiring. [14:39] not yet, sorry - have been working on the kernel for this console-setup bug [14:39] it's a bit time-consuming [14:39] np's [14:40] I'll check back next week [14:43] cjwatson: you're from belfast? [14:53] dmarkey: yep [14:54] born and bred, left when I was 18 [14:55] cjwatson: aww why did you leave? [14:55] not tough enough? :) === robbiew_ is now known as robbiew [14:56] dmarkey: university :) [14:56] then settled down [14:59] oh i see, well, if you're ever back home come down to monaghan for a pint [14:59] my parents moved over too later on, so I'm not back often, but now and again ... [15:00] ta :) [15:07] ev, np, i saw you planned on doing one earlier but were out when LP came back to life [15:08] In need of a honest opinion. Would a Mac frontend to usb-creator be a worthy Summer of Code project? I wont be offended if you say no :) [15:08] don't macs still require refit and what not though for actual install? [15:09] well, a more crackful idea would be to incorporate grub2's efi support as part of the project, but I'm not sure how well we support EFI just yet. [15:10] ev: When we did the usb-imagewriter code, I was specifically requested to write instructions for Mac. I didn't end up getting a GUI, but rather just instructions to write dd, which the (few) users seemed happy to use. [15:10] but yes, without an EFI bootloader it would only be able to generate images for regular PCs, as legacy bootloaders don't work on USB on Mac. [15:10] (but that was for image creation: I doubt anyone tried to *install* Ubuntu MID on a Mac) [15:12] I haven't actually tried grub2 efi yet; it's supposed to work [15:12] I do have a machine over <- there but my wife keeps doing inconvenient things like actually using it ;-) [15:13] I have the iMac upstairs, I might just give it a try tonight === ara_ is now known as ara [15:41] Is it possible to resize an existing Windows install during a preseeded (netboot) install? [15:43] id imagine it'd be a bit risky, but im no authority [15:44] We're not worried about that, these will be fresh builds anyway so if anything goes wrong it doesn't matter. [15:45] I'm just wondering if it's possible by using a combination of parted and early_command. [15:54] Omahn: you could do it manually with partman/early_command; there's no built-in preseeding for it [15:54] the parted command has a command-line batch mode [16:04] cjwatson: Ah, is parted on the alternative iso? [16:04] yeah; you might have to use anna/choose_modules=parted-udeb [16:05] cjwatson: That's what I'm missing, perfect. [16:10] cjwatson: 'No Implementation: Support for opening ntfs file systems is not implemented yet.' <--- is that support in the parted-udeb that's missing NTFS support or do I need to pull in a further udeb? [16:11] hrm, how on earth did this user-setup code for ecryptfs-setup-swap ever work? It checks for the existence of /target/dev/null, but that always exists as its in /rofs. [16:11] That was from: parted /dev/sda resize 1 32.3kB 29GB [16:16] Looking at the source it seems that use of parted for resizing filesystems is deprecated anyway. [16:18] ntfsresize [16:19] see lib/resize.sh in partman-partitioning [16:21] ah, yeah, you might need to do it by steam [16:23] ev: Thanks, that looks like it has everything I need to get this working. [16:31] user-setup: evand * r214 ubuntu/ (debian/changelog user-setup-apply): Fix check for already-mounted /dev in ecryptfs setup. [16:37] user-setup: evand * r215 ubuntu/debian/changelog: releasing version 1.28ubuntu4 [17:41] ev, http://imagebin.org/87512 I'm seeing this with 03/04 daily, but wasn't seeing it w/ 03/02 [17:42] very odd [17:42] is this with a preseed? [17:42] and ubiquity/partman-skip-unmount was set to true [17:42] Yeah [17:44] http://pastebin.com/Jwa0WSvG [17:45] oh wait [17:45] surely this is somewhat expected behavior [17:45] I mean, the message is expected [17:45] but I imagine you want it suppressed :) [17:46] well i thought that partman-skip-unmount was normally suppressing that, but maybe that was something else it was doing [17:46] but yeah, most certainly don't want to be seeing it [17:47] is this the same thing as what happened to usb-creator, where parted is returning a return code lately? [17:48] parted seems to be paying attention to the kernel re-reading the partition table now, yes [17:49] in this case it triggers an exception [17:49] (in partman, not ubiquity) [17:51] actually [17:51] if you preseed partman/exception_handler to null, does it continue? [17:51] as this is an exception with only a single choice, it should work I think [17:52] err exception_handler_note [17:52] oh damn, I should keep reading before I start talking [17:55] yeah, ignore the above suggestion, I apparently cannot read [17:55] just tried it before reading your response and it didn't do anything anyhow [17:56] indeed [18:19] hmm, this might be worse off than i thought, i'm not seeing the device nodes either anymore [18:24] parted 2.2 has a bug fix purporting to be related - I plan to get that in before b1 anyway [18:25] cjwatson: do you have a reference off hand? I'm quite intrigued. [18:34] i'm not sure i'd say it's parted offhand though. sfdisk wasn't complaining on 03/02 and is complaining now too [18:34] did the kernel interface for reloading the partition table change? [18:55] ev: just the NEWS file [18:56] ah, yeah, already looking at git. Thanks.