[01:35] ibex alt installer log "Oct 28 01:19:47 debootstrap: BAD signature from "Ubuntu Archive Automatic Signing Key " [01:35] this worked yesterday. I do go though apt-cache. anyone know what's going on? === persia_ is now known as persia [09:05] xivulon: I wouldn't worry about the virus. [09:06] It's a scanner that I've never heard of before. I'd be more concerned if it was Symantec. [09:17] evand: I'd been more concerned if it was avg [09:18] :) [09:44] yep, I have installed that scanner (quite dodgy thing, half in polish) and when installed it did not find anything [09:51] by the way same results for 8.04.1 and nobody complained :) [09:55] wonderful [10:28] Just thought if it's a new antivir. It might not have tools for opening 7z file only zip and rar so everything else might be highlighted as a possible threat [10:51] cjwatson: you fix for edubuntu on netboot may of broken it :( [10:51] just trying again [10:56] davmor2: you mean the thing that shouldn't be being tested? [10:57] davmor2: in any case, I didn't upload my fix [10:57] so it can't have broken anything [10:57] cjwatson: Ah that would explain it, it get's stuck at 2% any way [10:58] it's just thinking, it'll get there eventually I predict [10:58] How long would you say? [11:03] dunno :) [11:06] cjwatson: just done an install of kubuntu netboot along side it was on 2% for no more than 30seconds [11:07] it sticks at 6% for ages and then shoots up to 22% I think so I expect that pause [11:22] meh still stuck at 2% [17:52] hey guys, I'm having trouble with a package's postinst script in livecd-rootfs, basically its trying to see if /etc/gdm/gdm.conf-custom has been changed [17:52] the script calls for CUSTOMGDM=`cat $GDMCONF|grep "="|grep -v "#"` [17:52] and if it hasn't been changed, it exits with a nonzero error [17:53] can apt handle |? [18:13] pkgsel/include looks for debs or udebs? [18:22] cr3: dedbs [18:22] debs [18:22] acoc: apt has very little to do with postinst execution. That's several processes down the chain [18:23] acoc: I need to see the whole script in order to comment [18:23] cjwatson: aha! so in the ubuntustudio image, what might make use of the openssh-server udeb? [18:26] cjwatson: http://pastebin.com/m3fc791f4 - line 19 fails if gdm.conf-custom is not changed, no rush and thanks [18:27] cr3: it's on all images, due to network-console, a customised installation mode. see the installation guide [18:27] cr3: openssh-server-udeb is not really set up for use by anything other than network-console [18:27] cjwatson: interesting, I'll have a gander at the guide [18:28] acoc: I have no idea what that thinks it's trying to achieve [18:28] acoc: it's horribly overcomplicated for the only thing I can possibly imagine it's trying to do [18:29] acoc: is this your code? can you explain it? [18:29] cjwatson: it's hardware from Intel, so we're not "certifying" it in the strictest sense of the term but it's under "enablement". [18:29] cjwatson: oups, nevermind that [18:29] cjwatson: basically, if gdm config has been changed, the oz-desktop metafile will not replace the config [18:29] cjwatson: and no it's not my code, I'm just trying to figure out why it's failing [18:30] acoc: sure, but why would grep "=" | grep -v "#" tell you that? [18:30] it's probably failing because it's totally wrong [18:30] I'd just get in touch with whoever maintains that package if I were you, and get them to figure it out [18:31] cjwatson: it's just a config file, so he was trying to strip the comments and see if anything was set (ie =) [18:31] an unmodified file would have nothing set [18:31] reading some tutorials on shell quoting might be a good idea ... [18:32] personally, I'm learning shell scripting as I go, but besides bad looking code, is there anything wrong with what he's doing [18:32] acoc: honestly, I don't know what the problem is, but it's far more likely to be a problem in the package than in livecd-rootfs. The Swiss Army Knife of shell debugging is 'set -x' at the top of the script [18:34] I tried sed to do the same thing and it fails too (sed to stip comments and then grep) [18:34] fails how? [18:34] same error exit 1 on postinst [18:35] anyway, this really isn't a channel for teaching shell scripting, sorry [18:35] exit 1 => could mean anything [18:35] use set -x to get an execution trace [18:35] agreed [18:36] alright main reason I wanted to ask here was to see if while installing there were any commands that basically couldn't be used in a postinst [18:40] acoc: yes, but none that that script is using [18:41] acoc: the non-zero exit might have nothing to do with that line for all you know [18:44] cjwatson: By running it line by line I'm sure it's that line that's causing it, but I don't want to waste your time on an non installer based issue [18:45] cjwatson: thanks for answering my question though [19:14] hi! [19:14] hi! [19:14] hi [19:14] sorry about my really bad english :P but we are from #EdulinuxChile a Chilean Edubuntu based distribution [19:14] we are trying to reduce de ubuntu instalator steps [19:15] and...we are using preseed [19:15] the CD was created, but it's not bootable [19:16] mmmm..... I don't know [19:17] maybe this isn't the best tool to do this...but we don't know any other that works [19:17] anyone?? [19:21] jarias: don't be so impatient, please, IRC is asynchronous [19:21] jarias: what did you use to regenerate the .iso image file? [19:25] this command ----> mkisofs -pad -l -r -J -V "CD Preseed" -o /tmp/cdpreseed.iso /var/cdpreseed [19:27] cjwatson this is the command, cosf is part of the EdulinuxChile staff [19:27] you haven't given any of the options that would make it bootable [19:27] the command I would use (given your other options) is: mkisofs -r -V 'CD Preseed' -o /tmp/cdpreseed.iso -cache-inodes -J -l -b isolinux/isolinux.bin -c isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table /var/cdpreseed [19:31] excelent! thanks cjwatson [19:35] persia: did you ever think of allowing passwordless sudo when passwd/allow-password-empty is set? === Guest22585 is now known as Basti_dash [21:10] what is the best option to modify the ubiquity steps??? [21:12] cjwatson, unfortunately the incomplete language support is still coming up with that daily image from yesterday. i've put a bug together about it to track (bug 290398), but i anticipate not enough time to fix in time for release now unless it's something silly about how i'm preseeding or what i'm preseeding [21:12] Launchpad bug 290398 in ubiquity "Incomplete language support even with all languages preseeded" [Undecided,New] https://launchpad.net/bugs/290398 [21:17] thanks [21:22] superm1: ok, thanks [21:23] cjwatson, also you were looking for a log regarding stuff that was getting removed consistently, i put that on bug 290400 [21:23] Launchpad bug 290400 in ubuntu-meta "DVD livefs always removes java packages" [Undecided,New] https://launchpad.net/bugs/290400 [21:26] superm1: ok, thanks, the reason is as predicted judging from that so we'll deal with it in jaunty [21:27] cjwatson, it didn't appear so glancing through install.py , but is there any preseed key I could put to prevent them from getting removed? [21:28] such as preseeding those packages as to stay installed? if not, i'm thinking I might populate some hackery into an early command to make /var/lib/ubiquity/apt-installed with them listed [21:29] I don't think you can do this from an early_command; that's run from casper and ubiquity clears out /var/lib/ubiquity/apt-installed when it starts [21:30] that's too bad. [21:30] superm1: horrible hack alert [21:31] d-i pkgsel/language-pack-patterns string language-pack-gnome-$LL icedtea6-plugin [21:31] oh geez. [21:31] it's not designed for that but would work :) [21:31] yeah [21:32] as long as it's temporary for intrepid, and works, it'll save some factory burn time. i'll give it a shot === ubott2 is now known as ubottu [23:30] hum. this probably isn't a good sign. choosing s-chi in oem-config isn't setting the user's account into s-chi at all w/ oem-config 1.52 or 1.53. going into language support, no default language is getting set