[00:00] ah nice, shall I change them to fix released then? [00:00] is wubi 503 in as well? [00:00] not sure how to check that [00:02] 503> no, I'll look into that after I get back from jogging tonight or in the AM. [00:03] anyone here have a kickstart file for ubuntu 8.04 server amd64 that installs /boot + LVM? [00:04] thanks, to test 503 you need to have lupin 0.20 or manually apply the last patch at first boot [00:05] i cannot get the preseed directive to install the LVM using kickstart [00:06] it always fails and leaves me at the partitioning disks screen [00:09] ok.. so, my friend is booting boot.img.gz from USB via EEPC boot menu [00:10] linux loads and installs the usb storage before the flash storage modules :) [00:10] I dunno what modules they are right now. [00:11] so installing to /dev/sdb won't cause issues, i'd imagine.. it's mostly UUID based, but I dunno if grubs setup (hd1,0) will interfere [00:11] strike that [00:11] it should work fine [00:11] maybe I should enable edd? [01:26] Alright. [01:26] So if it's not detecting my CR-Rom drive, does it mean I'm screwed? Or can it be fixed? [01:29] usually, it's a kernel problem [01:29] worth comparing with the live CD to see if it manages to find the CD drive [01:29] that's a good way to narrow down whether it's a kernel problem or some kind of installer bug [01:30] I assume you're working with a released version of Ubuntu here rather than an alpha release or a daily build or whatever [01:30] Yeah, I'm using 8.04. [01:31] How would I figure out if it's a kernel issue or if it's a bug? [01:32] like I say, try the live CD, see if it works [01:32] As in run it in windows? [01:32] no, boot it normally [01:32] Oh, I've done that. [01:32] did you run the cd verification? [01:32] err, sorry, I was assuming you were working with the alternate CD? [01:32] I've tried that as well [01:32] perhaps you could give me a bit more context on exactly what you're doing [01:32] did it verify? [01:32] I'm assuming it is a matter of my drive not being able to read the disc [01:33] Or vice versa. [01:33] It verified when it burned [01:33] yes [01:33] Here. [01:33] when you booted it did you run the cd verification? [01:33] I burned the ISO to a CD at 40x spped [01:33] speed* [01:33] What do you mean? [01:33] There was no CD verification. >_> [01:33] :) [01:33] what iso are you using? [01:33] there is an option on the boot menu to check the CD [01:33] You mean check it for Defects? [01:33] I did that. lol [01:33] yes [01:34] however, it is unfortunately not 100% reliable [01:34] Mortis: did it pass? [01:34] Sorry. It just did the same thing, opened up busy box [01:34] it's possible for a CD to pass that and then fail in other ways [01:34] When I try to install it opens up BusyBox [01:34] CD drives are some of the flakiest bits of hardware in the universe [01:34] and doesn't install a thing. [01:34] Mortis: interesting [01:34] it is often the case that cleaning the drive can work wonders [01:34] where did you get the .iso? [01:34] When I try to verify the ISO, it does the same. [01:35] When I run the alternate install, it can't mount the image, or read a file called 'release'. [01:35] from Ubuntu.com [01:35] direct download. [01:35] do you still have the image? [01:35] I deleted it. :-/ [01:35] if so, do an md5 sum check on it [01:35] I already check summed it [01:35] It was fine. [01:35] you're a ton of help :) [01:35] ok [01:35] hardwire: I would not suspect a broken image here. [01:35] so sounds like your drive sucks eggs [01:35] Probably. [01:35] hardwire: no, it could also be a kernel problem [01:35] please don't jump to conclusions too quickly [01:36] Well, it does the same with Xubuntu. [01:36] cjwatson: not finding the right cd? [01:36] Xubuntu uses the same kernel [01:36] hardwire: yes [01:36] Oh. [01:36] Well. Let's see then. Maybe Ubuntu doesn't like my PC at all. [01:36] cleaning the drive is a worthwhile step before going any further [01:36] Mortis: maybe download the cd at http://www.insert.cd/ and burn that [01:36] try booting it [01:36] What would you suggest? [01:36] cjwatson, Mortis got some output from casper [01:36] Oh yes. [01:36] Mortis: cleaning the drive is a worthwhile step before going any further [01:36] Here I'll post what it said. [01:37] hardwire: I don't think that will be very helpful [01:37] stdin: I/O error [01:37] init: /init: 1: cannot open /dev/sdc: no medium found [01:37] And that same error came up in numeros paths with variations of /dev/sdc, sdf, sdd, sde [01:38] cjwatson: you may be right, it's an older kernel and it's a small download [01:38] hardwire: (a) it has a different kernel (b) imagine a speck of dirt at a certain point on the lens; now imagine that it hits a critical bit of the Ubuntu kernel, but something irrelevant on the Insert CD [01:38] Someone suggested seeing if Gutsy may work. [01:38] I suggested it because he didn't know if it was a kernel issue, try a different kernel [01:38] hardwire: trying random other CDs doesn't really help narrow it down much, I'm afraid [01:38] Mortis: yes, that just means "kernel completely failed to read from CD", not much more [01:38] Well, how do I fix it? [01:38] Cleaning my drive? [01:39] cjwatson: I suppose he could md5sum the cd. [01:39] And if so, do you mean defragmenting the disc, or just uninstalling apps? [01:39] hardwire: would you mind leaving this one to me? [01:39] I think us both chiming in is confusing [01:39] Mortis: no no, physically cleaning the CD drive [01:39] the lens tends to get dirty [01:39] you can get cleaning kits fairly cheaply, or there are guides on the web for cleaning them [01:39] Awsome. Someone just called and rick roll'd me. [01:40] Oh. [01:40] Jesus [01:40] rick roll'd? [01:40] I [01:40] Yeah. [01:40] it sounds trivial but it's a common cause of failures like this [01:40] You ever heard the song "Never Gonna Give you Up" by Rick Astley? [01:40] Yes... [01:40] Yeah. [01:40] Someone called me and played that song. [01:40] It's a 4chan meme...bleh. [01:41] It was probably my friend. [01:41] however, if that doesn't do any good (it may not), then get a dump of the PCI IDs of your system (you should be able to extract it from some other operating system too), and dump that into a bug report on https://bugs.launchpad.net/ubuntu/+source/linux describing your symptoms as accurately as you can [01:41] the PCI ID of your CD drive should identify it accurately enough to pin down the kernel driver responsible [01:42] but if my disc drive was dirty, wouldn't that mean it can't read anything? [01:42] it varies [01:42] often it affects just particular parts of the disk as read [01:42] Okay. [01:42] Well, does cleaning the disc your self involve opening the PC? [01:42] if you get lucky, it makes no difference; if you get unlucky, it flips a bit in your kernel and the universe implodes [01:43] I believe doing it properly does, but personally I've had decent luck with cleaning kits which are basically a little brush attached to a CD-like object [01:43] * cjwatson does not work for a cleaning kit manufacturer, for the avoidance of doubt ;-) [01:43] with the latter, you stick it into the drive and run a little program on the disk, it whirs around for a bit [01:44] Eh. I don't have any money, and plus, I don't really want to wait. [01:44] I may have to try cleaning it myself. [01:44] I don't know though. [01:44] google for CD drive cleaning [01:44] I'm going to try some other things before I do that. [01:44] Mortis, Have another CD-rom drive? [01:45] I doubt you'll be able to get it fixed without either drive cleaning, fairly serious kernel hacking (assuming that it is a kernel bug at all), or hardware substitution [01:45] No, I don't. [01:45] You guys really think it's that serious? [01:45] :-/ [01:45] If anything, it's probably a shit drive. It's probably 3 or 4 years old. [01:45] well, it ain't working ;-) [01:45] you could try writing a netboot CD - http://archive.ubuntu.com/ubuntu/dists/hardy-proposed/main/installer-i386/current/images/netboot/mini.iso [01:45] I'd like to know if the image he wrote matches the md5sums [01:46] It does. [01:46] that would help verify if it's a kernel issue or not. [01:46] hardwire: not really, no [01:46] if it matches, then it's possibly a kernel issue [01:46] Either way, my md5 sums match up [01:46] if it doesn't we can almost throw out kernel issues [01:46] hardwire, Are you new to the channel? [01:46] hardwire: he's already said several times that the checksums verified [01:46] Mortis: if you make an image of the cd you made, then md5sum it.. just curious [01:46] hardwire: no, we can't throw out kernel issues due to that. [01:46] cjwatson: the iso he downloaded verified. [01:46] hardwire: enough with the checksums, please. [01:47] lol [01:47] So what is this netboot? [01:47] I'm not trying to be stupid.. sorry [01:47] cody-somerville: yes, I'm new [01:47] but I'm not *new* [01:47] Mortis: the netboot image is a very small CD image that installs by downloading practically everything from the network [01:48] (no, not the entire network, just all the packages from the network instead of the local cd ;]) [01:48] Mortis: assuming you have the network bandwidth to install that way, it may help you dodge this kind of issue [01:48] Ok. What are my options for installing without a CD? [01:48] try the netboot CD image first [01:48] How do I get it to start downloading the files? [01:48] Just open it up? [01:48] follow the prompts [01:49] Mortis, You would boot it like you would Ubuntu [01:49] Okay. [01:49] you should try the netboot image because (a) it's much smaller and so less likely to run into physical errors due to e.g. dirt on the lens (b) loading it only involves the BIOS and syslinux, not the Linux kernel [01:49] What if it can't read this CD either? [01:49] and so it has a decent chance of avoiding both plausible causes [01:49] Mortis, you can do it with floppies [01:49] cross that bridge if we come to it [01:50] Alright, I'll give it a try. [01:50] cody-somerville: not on Ubuntu [01:50] Oh definitely not. [01:50] It's 700MB [01:50] Debian supports that (ish), Ubuntu never has [01:50] cjwatson, I thought I saw something for Ubuntu [01:50] I don't know of any 700MB floppys [01:50] floppies, for the avoidance of doubt, are just the installer bit not the whole 700MB [01:50] What speed would you recommend writing this iso to? [01:50] assuming they were supported, which they aren't [01:50] 40X [01:50] ? [01:50] Mortis: as low as possible [01:50] Okay. [01:50] It'll correct itself, right? [01:50] using the lowest possible speed is always a good idea when you have problems like this [01:51] 40X is vastly ambitious [01:51] * cody-somerville has never burnt anything that fast. [01:51] That may be why [01:51] Is it possible for it to skip over files from going so fast? [01:51] Your cd passed the verifier, remember? [01:51] Oh. True. [01:51] Anyways, I'm going to try this mini.iso [01:52] What should the write method be? [01:52] I have SAO [01:52] write method? [01:52] Yeah. [01:52] oh, doesn't matter, you're not going to be doing multisession [01:52] just use the defaults [01:52] Session-at-Once, Track-at-Once [01:52] Oh okay. [01:52] Sorry, I don't burn CDs often. Heh. [01:52] Mortis, your paranoia is understandable ;] [01:53] Wow. That wa sfast O_o [01:54] I'm just trying to do everything I can before I start messing with my PC [01:54] I don't want to break my PC over a Linux distro [01:54] Alright. [01:54] Wish me luck. I'm going with the mini.iso. I'll probably be back real soon. [01:54] Thanks guys [01:55] hardwire: I didn't mean to be rough, I've just dealt with a *lot* of these issues where the checksum verifies fine and I've learned that, while it's worth checking quickly to avoid silly mistakes, it doesn't pay to fixate on the checksums [01:56] you stretched for time or something? [01:56] not especially? [01:56] but I don't like to waste users' time either [01:56] you're approach was fine, I bet its going to work and waste less of a users time [01:57] and it didn't sound like a checksum problem at all to me [01:57] checksum problems don't produce I/O errors like that [01:57] then why worry about burn speed? [01:57] no reason not to play safe :-) [01:57] * cody-somerville nods. [01:57] * hardwire sighs [01:57] sure. [01:58] you don't get "stdin: I/O error" from an incorrect image, though [01:58] we all have our own way of doing things, mines just weirded. [01:58] weirder. [01:58] that means either a physical problem, or that the kernel is hallucinating a physical problem [01:58] I thought I remembered him saying the cd defect didn't even load [01:58] there is a vanishingly small chance that the kernel got corrupted, but you'd have to be pretty unlucky for it to carry on working and pretend like the CD was busted [01:58] cd defect detector [01:59] it failed in the same way - the CD defect check involves mounting the filesystem on the CD [01:59] which is the step that was failing [01:59] yup [01:59] it's a file-by-file check [01:59] I'm familiar [01:59] yeh.. I'm on it [01:59] that's why I wanted the iso checked [01:59] just so he knew what happened [01:59] checking the ISO is worthless in the case of an I/O error [01:59] it will almost never help [01:59] I missed the I/O error part [02:00] hardwire, next time you might want to take things a bit slower to ensure you read everything that is being said. [02:00] I've got another window open where I'm helping somebody who can't seem to download a checksum verified image for the life of him right now. [02:00] certainly, that kind of thing does happen (a lot ...) [02:00] cody-somerville: thanks for the advice guys, you're probably right on the money. [02:01] * hardwire ungrinds teeth [02:01] not a huge fan of admitting i've been overniced. [02:02] hardwire, well, atleast you don't throw chairs. :] [02:03] I rather like this chair. [02:03] It would be a shame to toss it [02:03] * cjwatson blinks at gfxboot [02:04] why are you calling ProgressUpdate before ProgressInit [02:04] meh.. I'd guess so it gave a more "instant" feel :) [02:04] or fail.. meh. [02:04] doing so crashes [02:05] eePC users are so very green [02:05] I've helped a few now put ubuntu onto them [02:05] it's almost like I need an image I can just flash onto them [02:05] I haven't messed with OEM mode ever. [02:08] oh, ouch, I bet this is an interaction between com32 and gfxboot [02:09] that would explain why suse didn't see it [02:09] I've been meaning to try out wubi [02:09] but I have no windows to try it on [02:10] I'll have to sneak it onto some other persons laptop [02:10] It is substantially more fragile [02:10] If the ntfs disk gets flagged dirty, *buntu will drop to busybox. [02:11] on a different note, coLinux is teh wild. [02:24] Ok. [02:24] I think I messed up. [02:24] It was partitioning, but it wasn't showing any progress...I figured it mucked up so I restarted it. [02:24] Now it won't resize any partition. [02:24] that was indeed a mistake [02:25] It just sat there at 0% D: [02:25] I think windows fixed it though... [02:25] you should be able to rectify it by running filesystem checks on those partitions [02:25] if they're Windows partitions, booting Windows ought to do that, yes [02:25] Ok [02:25] Well, it fixed it then [02:25] I think I'm just going to use an entire spare drive for linux [02:25] and back up all my music [02:25] there are indeed parts of the partitioner that will spend time working but without accurate progress information, I'm afraid [02:25] That's all I use my D: drive for anyways. [02:26] Okay. [02:26] Well, next time I'll be sure to go make a sandwhich and watch TV or something [02:26] Instead of sitting in front of my PC being inmpatient. [02:26] impatient* [02:27] the mini.iso does indeed work though. [02:27] Thank you so much [02:27] great, glad to hear it [02:27] Actually, I think I will have to resize my partition. [02:27] I need FLstudio and a ton of other things on this drive. [02:27] I'm going to go try again and see if it will let me resize. [02:28] Is there anyway to check if my partitions are messed up or anything [02:28] before I go wasting my time if it doesn't work [02:28] easiest way to do that is to boot the mini.iso and see if it likes it :-) [02:28] lol [02:28] Okay. [02:28] I'm going to back some stuff up on my D: drive first. [02:29] And when it sets up a partition, if I use the max space available, does that mean it's using both drives? [02:29] or hard discs rather [02:30] no, just one [02:30] That's odd. Why does it say I can use 117 gigs then? [02:30] you have to use LVM or RAID to spread a partition across multiple disks [02:30] When my two CD drives together only have about 70-80 gigs of space. [02:31] free space, mind you. [02:31] without knowing exactly which bit of text you're referring to, perhaps it means the total you could use if you resized your Windows partition to the smallest possible size [02:31] Well, doing that can cause serious problems in windows, right? [02:31] If I ever do use it again? [02:32] And when I was partitioning, I think it was doing it on my C: drive...how do I switch it to the other drive? [02:32] you ought to back everything valuable up first, of course, but it ought not to trash Windows [02:32] Or is partitioning only for your home drive. [02:32] C: and D: are Windows-style names for drives [02:32] Well, I have the XP disc, so I'm not too worried about windows. [02:32] Linux calls them /dev/sda and /dev/sdb (usually, possibly hda and hdb instead) [02:32] Oh. [02:32] Well [02:33] select the one you want [02:33] It didn't give me that option when I was resizing. [02:33] the automatic resizing thing won't, no [02:33] but you can use manual partitioning [02:33] Eh. I don't want to mess anything up... [02:33] It's not complex is it? [02:33] if you're going to stick everything on your second disk, then there's no reason to resize Windows on your first disk [02:34] the automatic resizing widget assumes that you want to resize Windows for a good reason, i.e. to use some of the space it was using [02:35] there is fairly substantial documentation on the web; you'd do well to look over it first [02:35] Dang, Spyware Terminator was taking up 33 gigs [02:35] I'm just going to use my D: drive. [02:35] I freed up enough space for all my music [02:35] there is an automatic partitioning option to use an entire disk [02:35] erasing anything that was previously on it [02:36] you select use entire disk, then you select the disk you want to use. it will ask you for confirmation before actually doing anything [02:36] check the size of the disk it suggests carefully; it might be that Linux detects your disks in the opposite order to Windows [02:37] Yeah [02:37] Well, when I did that, it aid both had 120 gigs of space [02:37] It didn't list how much was free :-/ [02:37] ah, difficult [02:37] Yeah... [02:37] * cjwatson <- not responsible for what happens if you don't take backups [02:37] Obviously XD [02:38] Is there any way you know of that I could check? [02:38] the manual partitioner may be more informative here [02:38] Okay. [02:38] I'll try that one then. [02:38] you could drop into it and see if it helps you [02:38] you can delete or resize a partition there, then select guided partitioning and tell it to just stick Ubuntu partitions in a free region [02:39] Uh oh... [02:39] just note that, unless it says otherwise, it doesn't touch your disks until after you've acknowledged a confirmation message [02:39] I can't delete anything from my recycle bin -_- [02:39] so if you think you've screwed up, just reboot [02:40] Okyy. [02:41] I'll be trying that in a moment. [02:43] * cjwatson -> bed [02:43] Later. Thanks for your help === cr3_ is now known as cr3 [03:55] hi! can anyone answer a question about the ubuntu-desktop metapackage? [03:55] I am trying to modify the list of packages installed by default on my ISO [03:57] giosue_c: What kind of modification? [03:58] hardwire: i want to not install a certain package [03:58] ubuntu-desktop? [03:58] hardwire: i removed a dependency from the xubuntu-desktop metapackage and rebuilt the ISO... but the package still gets installed [03:59] have you checked to see if it is required by other packages as well? [03:59] i see that is is "suggested" by another packages [03:59] ah, that wouldn't matter [03:59] that's what i thought. [03:59] are you at a full xubuntu install right now? [04:00] after the system boots I can remove the package without other packages being removed... so i don't understand. [04:00] yes [04:00] if you apt-get --purge --auto-remove remove packagename does it try to uninstall anything that xubuntu-desktop uses? [04:00] i have the modified system and a clean system at my fingertips [04:00] lemme check... [04:02] so i executed apt-get remove --purge --auto-remove apmd [04:02] and it says apmd* will be removed [04:02] that is it. [04:02] to me that means my package modification is good, but i don't know why apmd was still installed in the first place. nothing depends on it. [04:04] hmm [04:04] when i look at the preseed for d-i it looks like they just install xubuntu-desktop and that causes everything else to be installed. [04:04] so that is why i modded that package [04:04] are there any other mechanisms that they use to specify what it installed? [04:04] what about ubuntu-minimal [04:04] and ubuntu-standard [04:05] are those metapackages too? [04:05] sure [04:06] i'll rip those open and see what they depend on. [04:06] apt-cache rdepends apmd [04:07] i get a list... but how can i tell what from that list is on my system? [04:07] is there a neat trick for that? [04:08] hard work :) [04:08] darn! [04:08] actually, apt should know how it all works [04:08] hence removing it should remove the foul creature that installed it [04:08] whats up, why is apmd bugging you? [04:09] well... it is sort of a proof of concept. apmd is one of many packages that I will be removing. [04:09] the systems that these ISOs will be installed on have very limited HDD. normally i run a script that does a bunch of installs and removes. [04:09] i'm getting fancy though... ;) [04:09] and is aptitude being told to install suggested packages? [04:10] giosue_c: sounds like you're better off managing your own metapackages [04:10] hmm.. dunno. i just let the installer run and this is what i end up with. [04:11] also.. xubuntu-desktop is in upstream pools [04:11] yea. i agree. this is sort of dipping my toe in the water. I wanted to see if it would work how i expected. [04:11] err.. it's in pools that you can't control [04:12] so if you're installing xubuntu-desktop from a cd you modified, eventually that package is going to get updated somewhere else in the world and your computer will install apmd [04:12] my final solution I'll have a whole new metapackage... similar to xubuntu-desktop but with a different name [04:12] right [04:12] if you are doing an xubuntu install *AND* using a network mirror, you're probably going to pull down the latest xubuntu-desktop package instead of use whats on the cd [04:12] all during the install [04:12] ok. [04:13] oh... [04:13] headdesk? [04:13] what is headdesk? [04:13] nevermind :) [04:14] http://www.urbandictionary.com/define.php?term=headdesk [04:14] maybe it's not the right term for what you probably just felt. [04:15] headdesk. [04:15] well. the master plan is now to create a new metapackage very similar to xubuntu-desktop [04:15] and put it in the preseed file [04:15] you should maybe make your own metapackages, make your own package pool, and then make a preseed that uses them [04:15] call it giosue-desktop or something ;) [04:15] can do [04:16] or hardwire-desktop [04:16] you'd name your baby after me? [04:16] how sweet. [04:16] :) [04:17] i'm still not sure i understand how apmd got installed on my system though... [04:17] :( [04:17] when i remove, nothing else gets removed. [04:17] you mentioned aptitude can be told to install suggested. surely that isn't the behavior at installer time... [04:18] you would end up with all sorts of junk in there. [04:18] happy junk [04:18] but it was just a thought, not really an answer [04:19] I'm guessing while you are installing off your new cd image you are using the "use a network mirror" option before you finish the install [04:19] and it pulls down a list of more recent packages than what is on CD [04:19] one of which.. probbably, being xubuntu-desktop [04:19] so it's superseding your changes [04:19] hmm.. i am using the alternate-installer [04:20] and what modifications did you make to the alternate installer cd? [04:20] and if i installed an updated xubuntu-desktop i wouldn't have been able to remove it. [04:20] it doesn't have to be installed for xubuntu to do it's magic [04:21] it just needs to be installed once [04:21] hmm. interesting. [04:21] I'm not sure how you're handling things, at least well enough to tell you exactly what's causing this [04:21] so the dependencies come from the packages file in the mirror. [04:21] giosue_c: most of the time more recent version of packages are available on mirrors [04:22] I should probably give my package a ridiculously high version number to test this theory [04:22] that's one way of doing it :) [04:22] or make sure it is pointing at my own special mirror. [04:22] the hardwire mirror ;) [04:23] anyway. these are all helpful leads you've given me [04:23] certainly a charming little fella [04:23] I gotta go try it [04:23] will drop in later if it works... or doesn't [08:32] iam getting problem while booting up the livecd " [08:32] mount: Mounting /dev/loop0 on //filesystem.squashfs failed: Invalid argument [08:32] Can not mount /dev/loop0 (/cdrom/casper/filesystem.squashfs) on //filesystem. [08:32] squashfs [08:32] " [08:34] can anybody know the solution [08:35] stdin: erroe 0 [08:38] plz tell me [10:04] iam getting problem while booting up the livecd " mount: Mounting /dev/loop0 on //filesystem.squashfs failed: Invalid argument Can not mount /dev/loop0 (/cdrom/casper/filesystem.squashfs) on //filesystem.squashfs" [10:04] cjwatson: fyi have merged the gobby dump to the wubi intrepid wiki, since noticed you mentioned that in the meeting [10:47] iam getting problem while booting up the livecd " mount: Mounting /dev/loop0 on //filesystem.squashfs failed: Invalid argument Can not mount /dev/loop0 (/cdrom/casper/filesystem.squashfs) on //filesystem.squashfs" [10:51] (a) please don't repeat yourself (b) is this a modified live CD? otherwise check dmesg after you get dropped to a busybox prompt [10:51] "Invalid argument" (a.k.a. EINVAL) is a famously uninformative kernel error code [10:54] xivulon: it's usually a good idea to avoid specifications that are largely lists of bullet points; furthermore, I'm sure there are items in there that we discussed and said were infeasible for intrepid, and these should not be in the wubi-intrepid spec [10:55] this spec should be a concrete description of things to do for intrepid, rather than a wishlist [11:08] os-prober: cjwatson * r217 ubuntu/ (5 files in 4 dirs): merge from Debian 1.26 [11:13] os-prober: cjwatson * r218 ubuntu/debian/changelog: releasing version 1.26ubuntu1 [11:14] cjwatson I will clean it up in the coming days [12:07] is there a wiki for proposed usability enhancements in ubiquity for intrepid? [12:43] cjwatson, https://wiki.ubuntu.com/WubiIntrepid take-2, did a bit of cleanup will spend more time on it later on [12:47] mpt can you please review the "Ubiquity Partitioning Page Mockup" in there? [12:48] "Wubi" is the coolest program name ever [12:48] hehe [12:51] however [12:52] xivulon, I don't think "Migrate the existing Wubi installation" is quite human-centric enough [12:52] ah sure, feel free to edit [12:53] something like: [12:53] You currently have Ubuntu installed on Windows. [12:53] (*) Copy all files and settings from this installation into a dedicated partition [12:54] ( ) Install Ubuntu from scratch [12:55] done [12:55] hmm [13:40] evand did you talk to slangasek re ntfs-3g patch? [13:48] hmm I wonder if dm-loop can be used with LVM2 to create resizable virtual disks, looks like they are part of the same project... [13:55] cjwatson, you mentioned you had a discussion with don re umountfs, from the notes it seems the safest approach is to unmount /proc/mounts in reverse order, do we still need -l option? [13:58] I wouldn't think it hurts, particularly if we unmount items before / (/host in particular). man shows -l is only supported in 2.4.11+, shouldn't be an issue. [13:59] I don't think it matters, if you're doing things in reverse order [13:59] I'd think that it is relevant for /host (unless we stop at /) [14:00] stopping at / seems to make sense [14:00] true, but lazy unmounting /host is also desirable [14:00] since now we do not really unmount it... nor remount it ro [14:01] I suppose [14:01] as for -f, man mentions it might be useful for "unreachable NFS system" [14:01] not sure what happens in that case [14:02] but surely -f creates problems to bindmounts (which I think it is a umount bug by the way) [14:02] I'd leave -f alone [14:02] if we keep -f, then we need weakmountpoints or fixing umount [14:03] re /host, on second thought, not sure -l would help [14:03] since /host contains / which is only remounted ro, so /host would never be unmounted anyway [14:11] xivulon: No, I got back late last night. I'll talk to him during his core hours today. [14:12] I think we could ask pitti now [14:12] he sponsored other patches on ntfs-3g already [14:13] would you agree if we bring it on #ubuntu-release? [14:15] agree to bring it on #ubuntu-release? Sure. [15:26] would it be feasible to handle wubi-migration stage 2 with a custom d-i based initrd? [15:27] so we use ubiquity for stage 1 and d-i in stage 2 [15:27] that would spare people troubles with burning CD, bios, overrding usb content and such [15:27] since it is an automated installation d-i might even use usplash as frontend [15:35] IMO, we should stick to the plan we came up with at UDS, but if cjwatson disagrees then I can be persuaded as well. [15:38] hello [15:39] did anyone work on something similar to oem-config for server? [15:41] evand, I must admit I haven't touched the d-i initrd since lupin 7.04 so I might misjudge the complexity of the task, but if doable it would be a more userfriendly approach, and we wouldn't need to edit raw disk bits to flag gfxboot... [15:45] we will need to add local HD preseeding to d-i anyway, and once that is done and provided we keep a list of udeb dependencies, is that much more work than the Ubiquity+gfxboot approach? [15:52] offhand, I'm not sure [16:01] evand: I'm not especially happy with redesigning it on the fly now unless there is a fatal problem with the agreed design [16:01] nijaba: not as yet [16:02] xivulon: sticking usplash in front of d-i would be very hard work [16:02] xivulon: let's stick with what we agreed [16:02] cjwatson, usplash is a detail, I don't think we thought about d-i in stage 2 at UDS and wanted to bring it up now [16:02] cjwatson: what would be the best course of action to start working on it? [16:03] as it might be relatively easier to implement and certainly more user friendly [16:03] nijaba: write a text frontend to oem-config [16:03] cjwatson: ok, agreed [16:03] cjwatson: ok thanks [16:03] xivulon: I don't think that approach is likely to be more user-friendly [16:04] much though I love d-i [16:04] its strengths are power and flexibility, not user-friendliness [16:04] well if it is non-interactive that does not matter... [16:05] but it avoids problems with burning a CD (bad medium) and booting off CD/USB (bios or people ejecting) [16:06] basically all the user has to do is reboot and choose Ubuntu again [16:07] then d-i kicks in and completes the installation, then next reboot you are in a dedicated partition installation [16:08] sure the progress bar won't be as pretty but I think that is ok [16:10] so in stage 1 all we do is create a preseed, download d-i initrd/kernel in /boot, and replace menu.lst [16:22] well have added a note to the wiki for reference [19:56] debian-installer: cjwatson * r924 hardy-proposed/debian/changelog: No-change rebuild to pick up new components. [20:01] debian-installer: cjwatson * r925 hardy-proposed/debian/changelog: releasing version 20070308ubuntu40.4 [20:11] debian-installer: cjwatson * r941 ubuntu/ (4 files in 3 dirs): Move mainline architectures to 2.6.26-2 kernels. [20:44] so if I understand correctly, rosetta-merge-all needs to be run over debian-installer/ and installer-po/ now, correct? === evand_ is now known as evand [21:01] evand, does the iso you tested with include the latest partman-auto-loop? [21:02] 6V,BHP90G8 [21:02] ^kids [21:14] oh lovely, there's no ubiquity with partman-auto-loop 0ubuntu14 in hardy-proposed [21:15] so if I understand correctly, rosetta-merge-all needs to be run over debian-installer/ and installer-po/ now, correct? [21:15] cjwatson: ^ [21:16] I think so - if you're uploading anyway, certainly [21:16] I think debian-installer/ is more important [21:17] from a quick scan of the languages, it has the packaged languages whereas installer-po/ has the ones which are new in LP, but I'd do both if I were you just to be sure. rosetta-merge-all will skip any ones that aren't already in ubiquity anyway [21:17] ok, thanks [21:18] Given that I have to upload a new ubiquity, is it reasonable to include the missing translation updates as well? [21:18] I think so, yes [21:18] ok, I'll make an SRU request for the both [21:18] check with Steve whether that's actually necessary [21:19] or whether it just counts as syncing up with other stuff [21:22] Hrm, the Khmer translation appears to be broken. [21:22] I'll leave that one out [21:26] Arr, and part of the Spanish one. [22:06] check the Portuguese (and maybe Brazilian, I forget) ones - there was a report of #-#-#-# type breakage there [22:08] they look OK [23:31] evand is http://cdimage.ubuntu.com/hardy/daily-live/20080619.3 any good? [23:40] guess not, it contains ubiquity 1.8.11 :(