[00:10] <CIA-17> ubiquity: evand * r3601 lucid/ (3 files in 2 dirs):
[00:10] <CIA-17> ubiquity: Provide human readable sizes in the partitions-too-small warning
[00:10] <CIA-17> ubiquity: (LP: #298974).
[01:01] <bdmurray> evand: it installed for me on one with <3GB of memory
[01:02] <bdmurray> evand: using an ubuntu server i386 cd
[01:02] <evand> bdmurray: what does /usr/lib/base-installer/dmi-available-memory print out?
[01:02] <bdmurray> evand: that should be on the live cd right?
[01:02] <evand> bdmurray: yes
[01:03] <bdmurray> evand: that directory doesn't exist only base-installer.d
[01:04] <bdmurray> evand: and I can't find dmi-available-memory
[01:04] <evand> what live CD is this?
[01:04] <bdmurray> ubuntu-server/karmic-server-i386.iso
[01:06] <bdmurray> maybe live CD is the wrong word...
[01:06] <evand> indeed
[01:06] <evand> can you boot a live CD with that machine and run the above command?
[01:07] <bdmurray> evand: sure, how does it work on the server install though?
[01:08] <cjwatson> you need to run through the installer a certain amount
[01:08] <cjwatson> (server)
[01:08] <evand> the same way, it's just easier to explain how to do it on a live CD
[01:08] <evand> well, roughly
[01:08] <cjwatson> run up to the hostname prompt, then you'll have /usr/lib/base-installer/dmi-available-memory
[01:08] <evand> or do that :)
[01:08] <cjwatson> when d-i boots, it only has the core of itself - it installs more bits of itself on the fly
[01:08] <bdmurray> ah, right
[01:09] <bdmurray> 524288
[01:10] <bdmurray> on both of them
[01:11] <evand> bdmurray: please attach the syslog after install to a new bug
[01:11] <evand> so /var/log/installer/syslog
[01:17] <bdmurray> evand: its bug 463027
[01:17] <ubottu> Launchpad bug 463027 in debian-installer "pae kernel installed on a system with ~512MB of RAM" [Undecided,New] https://launchpad.net/bugs/463027
[01:18] <evand> bdmurray: please also attach the output of dmidecode
[01:27] <cjwatson> bdmurray: baffling
[01:28]  * cjwatson pulls down the server iso
[01:28] <bdmurray> dmidecode added
[02:11] <isamar> hi folks
[02:12] <isamar> customizing Jaunty LiveCD and I get the following error mesg:
[02:12] <isamar> Signature verification failed for: /cdrom/dists/gutsy/Release.gpg
[02:12] <isamar> gusty=jaunty
[04:01] <JanC> <CompactDstrxion> hmm running wubi from this iso. at the top it says 'You are about to install Ubuntu-9.10ubuntu1"
[04:01] <JanC> (from #ubuntu-release-party)
[09:07] <CIA-17> wubi: Agostino Russo * r161 trunk/ (3 files in 3 dirs): Fix country detection (LP: #428575)
[09:17] <CIA-17> wubi: Agostino Russo * r162 trunk/ (debian/changelog src/wubi/backends/common/mappings.py): Add 'Nederland' to country name mappings (LP: #428575)
[09:29] <Omahn> When adding d-i apt-setup/local0/repository line in preseeded installs, is it possible to include a variable that resolves to the relevant release or do you specifically have to put the release name in?
[09:30] <Omahn> The example given shows:
[09:30] <Omahn> #d-i apt-setup/local0/repository string \
[09:30] <Omahn> #       http://local.server/ubuntu &releasename; main
[09:30] <Omahn> Yet that doesn't appear to resolve, it just inserts as is. :-(
[09:43] <Omahn> And to answer myself, a workaround is to use:
[09:43] <Omahn> d-i preseed/include_command \ string echo `grep CODENAME /etc/lsb-release | cut -f 2 -d '='`.cfg
[09:43] <Omahn> and then have a preseed for each release available.
[09:55] <Omahn> Or maybe not. The installer is failing to load the reference .cfg file, it doesn't even attempt to download it.
[09:56] <cjwatson> JanC: we know, and it's corrected in bzr, but was too late for the final images
[09:56] <cjwatson> Omahn: no variables, sorry
[09:57] <Omahn> cjwatson: No problem, any idea why the preseed/include_command doesn't appear to work?
[09:57] <cjwatson> Omahn: the example is intended to be processed by a program that turns &releasename; into karmic; if you're seeing &releasename;, you might be reading something too high up in the processing chain?
[09:57] <cjwatson> Omahn: dunno, anything in syslog?
[09:57] <Omahn> Nothing at all.
[09:57] <cjwatson> that's a hint to please show me the syslog :)
[09:58] <Omahn> :-)
[09:58] <Omahn> Two tics..
[09:59] <Omahn> http://ubuntu.pastebin.com/d7acc32bf
[09:59] <cjwatson> "server returned error 503: HTTP/1.0 503 Service Unavailable"
[10:00] <cjwatson> so check the server logs?
[10:00] <Omahn> That's several minutes after the failed attempt.
[10:00] <Omahn> (see the shell I start just before)
[10:00] <cjwatson> hm, ok
[10:01] <Omahn> I'm just having a look at a packet capture on the http server, I suspect it's not making any request at all.
[10:02] <cjwatson> before it gets to that point (you might have to e.g. not preseed the locale question), edit /var/lib/dpkg/info/network-preseed.postinst and change the second line from 'set -e' to 'set -ex'
[10:02] <cjwatson> then try again, and you should get a trace in the syslog
[10:03] <Omahn> I'll do that now. The packet capture shows that the only incoming request is for the original preseed, not the included one.
=== mterry is now known as partyanimal10
[10:13] <Omahn> Does the preseed/include_command need to appear anywhere in particular in the original preseed file? (I currently have it a the bottom)
[10:15] <CIA-17> partman-iscsi: cjwatson * r36 ubuntu/debian/ (di-numbers install-rc changelog control rules): Use dh-di.
[10:15] <cjwatson> order shouldn't matter
[10:19] <Omahn> cjwatson: Thanks, i'll get back to you. I'm starting to suspect my web server is serving up old copies of the original preseed.
[11:10] <CIA-17> ubiquity: cjwatson * r3602 lucid/.pycheckrc: [pychecker] ignore unused args/kwargs
=== partyanimal10 is now known as mterry
[11:19] <CIA-17> ubiquity: cjwatson * r3603 lucid/ubiquity/frontend/base.py: [pychecker] suppress argsused warnings in ubiquity/frontend/base.py, as there are lots of abstract methods here
[11:34] <CIA-17> ubiquity: cjwatson * r3604 lucid/ubiquity/frontend/gtk_ui.py: [pychecker] remove datetime and math imports; suppress classattr warnings; more usual syntax for boolean comparisons
[11:37] <CIA-17> ubiquity: cjwatson * r3605 lucid/ubiquity/plugin_manager.py: [pychecker] using lists as default arguments can be confusing
[11:39] <CIA-17> ubiquity: cjwatson * r3606 lucid/ubiquity/ (segmented_bar.py timezone_map.py): [pychecker] use gtk.gdk explicitly; note unused arguments; sum -> total
[11:49] <CIA-17> ubiquity: cjwatson * r3607 lucid/run-pychecker: pychecker has a habit of emitting extraneous warnings ignoring __pychecker__ unless we process files one at a time
[11:50] <CIA-17> ubiquity: cjwatson * r3608 lucid/ubiquity/components/ubi-language.py: [pychecker] remove unused re and locale imports
[11:50] <CIA-17> ubiquity: cjwatson * r3609 lucid/ubiquity/wrap_label.py: [pychecker] note unused arguments
[11:52] <CIA-17> ubiquity: cjwatson * r3610 lucid/.pycheckrc: [pychecker] neater way to avoid limit
[12:03] <CIA-17> ubiquity: cjwatson * r3611 lucid/ (debian/changelog debian/control ubiquity/tz.py):
[12:03] <CIA-17> ubiquity: Require Python 2.5, so we can now use hashlib rather than md5 and avoid
[12:03] <CIA-17> ubiquity: a slew of warnings.
[12:11] <CIA-17> ubiquity: cjwatson * r3612 lucid/ubiquity/frontend/kde_components/Timezone.py: [pychecker] unused variables
[12:29] <CIA-17> ubiquity: evand * r3613 lucid/ (3 files in 2 dirs):
[12:29] <CIA-17> ubiquity: Mark the "Installation Complete" window as always on top
[12:29] <CIA-17> ubiquity: (LP: #462178).
[12:29] <cjwatson> shtylman: what's the 'selected_zone = -1' in ubiquity/frontend/kde_components/Timezone.py:mouseReleaseEvent() for? pychecker correctly points out that that's never used after setting it, and I was wondering if it was meant to be a change to self.selected_zone although that seems to have a different format
[12:29] <cjwatson> shtylman: also x and y in the same method have the same problem, and I don't know what to do about those
[12:31] <CIA-17> ubiquity: cjwatson * r3614 lucid/ubiquity/segmented_bar.py: [pychecker] remove unused remove_segment methods
[12:33] <CIA-17> ubiquity: cjwatson * r3615 lucid/ubiquity/segmented_bar.py: [pychecker] none -> no_corners
=== ogra_ is now known as ogra
[12:41] <CIA-17> ubiquity: cjwatson * r3616 lucid/ubiquity/frontend/noninteractive.py: [pychecker] note unused variables
[12:42] <CIA-17> ubiquity: cjwatson * r3617 lucid/ubiquity/components/ubi-language.py: [pychecker] unused import
[12:44] <CIA-17> ubiquity: cjwatson * r3618 lucid/ubiquity/frontend/kde_components/Keyboard.py: [pychecker] type shadows builtin and is unused anyway
[12:46] <shtylman> cjwatson: I will take a look at that in a few moments (about to change locations) ... it may be that it was there at one point for something but eventually the code got changed and those never got cleaned up
[12:46] <CIA-17> ubiquity: cjwatson * r3619 lucid/ubiquity/__init__.py: [pychecker] revert extra imports from r3591, which turned out to be unhelpful
[12:47] <cjwatson> shtylman: right, pychecker is showing up a lot of stuff like that
[12:53] <CIA-17> ubiquity: cjwatson * r3620 lucid/ubiquity/ (3 files in 2 dirs): [pychecker] unused variables and imports
[12:57] <kim0> cjwatson: man, I want to stop the installer from installing updates online. it should only install what's on the DVD. You mentioned I should use "d-i apt-setup/use_mirror boolean false"
[12:58] <kim0> that correctly removes sources from sources.list, *except* the security updates .. can I remove that also somehow ?
[12:58] <kim0> thanks in advance
[13:06] <kim0> I think I need "d-i apt-setup/security_host string" ...
[13:24] <cjwatson> kim0: you've got it
[13:25] <cjwatson> or I think 'd-i apt-setup/services-select multiselect' will do it too (default is 'security')
[15:50] <CIA-17> ubiquity: evand * r3621 lucid/ (debian/changelog ubiquity/frontend/kde_ui.py):
[15:50] <CIA-17> ubiquity: * Fixes from Pychecker for the KDE frontend (kde_ui):
[15:50] <CIA-17> ubiquity:  - Don't import datetime or math. The timezone code is in a separate
[15:50] <CIA-17> ubiquity:  module now.
[15:50] <CIA-17> ubiquity:  - Remove some unused variables.
[15:51] <CIA-17> ubiquity:  - Don't assign to a variable that's going to be immediately discarded.
=== robbiew is now known as robbiew-afk
[22:59] <DanielRichman> Ok. Being a fan of the minimal installer/mini.iso I'd like to use that to install karmic. The links arn't up on the wiki, but I'm guessing that http://archive.ubuntu.com/ubuntu/dists/karmic/main/installer-amd64/current/images/netboot/ is the right place. The date modified is showing 17th of October - are these images good to use to install karmic release now?
[23:14] <choman2> is this a good place for wubi questions
[23:32] <DanielRichman> (question has been answered on #ubuntu; no worries)