[09:15] good morning ev, cjwatson. How is the POT template of ubiquity created (the one for the installer, not the one for the .desktop file on the live installation)? I've had a quick look at the sources and I couldn't quite figure out how. I'm just looking at bug 533859 and I thought it might just be a matter of the template not being up to date and not yet containing those strings missing for translations [09:16] Launchpad bug 533859 in ubiquity "Missing translations in latest Ubuquity version from Lucid" [Undecided,New] https://launchpad.net/bugs/533859 [09:28] dpm: by running debconf-updatepo by hand [09:28] ok, thanks cjwatson [09:59] ubiquity: evand * r3897 ubiquity/ (debian/changelog ubiquity/frontend/gtk_ui.py): [09:59] ubiquity: Only try to set the focus on the next button when we've handed over [09:59] ubiquity: control to the user. Prior to this the quit button was getting the [09:59] ubiquity: focus because the next button was insensitive. [10:07] dpm: anyway, at least one of those strings has a 'TODO i18n' comment by it [10:09] so I think it's just not internationalised yet :( [10:11] on it now [10:11] ah, thanks cjwatson, ev :) [10:51] usb-creator: evand * r271 usb-creator/ (bin/usb-creator-helper debian/changelog): [10:51] usb-creator: Provide a quick workaround for LP 529366 while a better fix is [10:51] usb-creator: prepared. Ignore errors on setting the boot flag for the target [10:51] usb-creator: partition. [10:51] Launchpad bug 529366 in usb-creator "Regression: usb-creator-gtk doesn't work as of 0.2.16" [High,Triaged] https://launchpad.net/bugs/529366 [10:55] ubiquity: evand * r3898 ubiquity/ (4 files in 3 dirs): [10:55] ubiquity: Use translatable strings for the keyboard query widget [10:55] ubiquity: (LP: #533859). [11:00] usb-creator: evand * r272 usb-creator/debian/changelog: releasing version 0.2.18 [11:02] michaelforrest: http://people.canonical.com/~evand/tmp/usb-creator-common_0.2.18_all.deb http://people.canonical.com/~evand/tmp/usb-creator-gtk_0.2.18_all.deb [12:25] I've heard some reports (one in Millbank, others in bugs) of grub install failing and not going to the "grub install failed, choose a different device to install it on dialog". Just a heads up, I'm going to run through a test install and see if I can reproduce it. [12:26] also, I broke the resize progress bar, fixing nowish === mcasadevall is now known as Guest20793 === Guest20793 is now known as NCommander [14:09] ubiquity: evand * r3899 ubiquity/ (4 files in 3 dirs): [14:09] ubiquity: Bring back the resizing partition progress dialog that was removed [14:09] ubiquity: with the rest of the automatic partitioning progress dialogs. [14:09] cjwatson: your extra parameter to progress_stop ended up having a use after all. ^ [14:14] ev: can't we keep track of this some other way? we ought to be able to remember the template used for the topmost progress bar [14:14] cjwatson: sure [14:22] cjwatson: how's this: http://paste.ubuntu.com/391075/ [14:23] ah, nesting, damn. [14:27] ev: I thought ubiquity.progressposition already kept track of this [14:34] Okay. I'll take a look after I'm done hunting down the cause of an apt-setup bug with cruzer micros. [14:35] I'm resurrecting a machine to put some giant hard disks in for parted testing [14:38] Nice === robbiew_ is now known as robbiew [15:08] I really need to pick up a netbook. As much fun as playing roulette with my primary laptop and two USB keys is... [15:10] you must be coming up on three years at Canonical soon [15:10] laptop refresh benefit ... [15:10] ev: don't be tempted with the ssd ones ;) [15:11] davmor2: why not? I am :) [15:11] May :) [15:11] cjwatson: not enough space [15:11] I've certainly seen laptops with SSDs bigger than my current laptop's hard disk [15:12] Ubuntu takes under 3gb... [15:12] I'm happy to use fileservers elsewhere for serious storage [15:12] * persia has happily used a 4G netbook at UDS. [15:12] cjwatson, ev: yeah the issue comes when you want to try a side by side with windows :) [15:13] davmor2: you can get SSDs much bigger than that [15:13] 128GB is not totally unreasonably priced [15:13] cjwatson: yeah but at the time 8gb was the standard in netbooks :) [15:13] yeah, I'm not interested in netbooks :) [15:14] not made for people with big hands [15:14] cjwatson: ev said he was though :) [15:14] mm, yeah [15:15] Yeah, but my use case for it is only installer development and testing [15:16] ev: not enough room for side by side or wubi testing. on 8gb drive windows takes up loads [15:16] Sure, I'd plan accordingly. [15:17] ev: almost certainly I'm just saying :) [15:17] Assuming the pound ever rebounds, I can always pick up a big Intel SSD on the cheap in the states [16:56] shtylman: can you please provide a changelog message for the outstanding items in your branch? Also, why did you comment out the progress window code? [16:57] does d-i and/or casper support installation where the installation media is available over an iscsi drive? [16:59] cr3: casper doesn't, d-i mostly does [16:59] oh, *installation media* available over iscsi [16:59] hmm, probably only if your BIOS sets up the iscsi drive to look like scsi [16:59] cjwatson: as opposed to installing onto an iscsi drive [17:00] I have to admit that my knowledge of iscsi is flaky at best [17:00] ev: I commented it out cause I put back the new style progress bar which was commented out for a quick release right before alpha 3 [17:01] ev: yea...I think I can put some poetry in the changelog [17:01] thanks [17:02] can you simply delete it then? We don't need to comment things out when we have a VCS, unless you expect it to be reverted again. [17:02] ev: yea... I agree... and no.. I hope it doesn't need to be reverted again [17:02] the goal is for it not to be at least [17:05] okay, cool === dpm is now known as dpm-afk [18:31] cjwatson: is grub2 taking 3-5 minutes to install the bootloader a bug, or expected behavior? [18:32] cjwatson: i ask because my automated cloud installations take about 10 minutes per machine, with 3-5 minutes of that being the grub installation step (at least as reported by d-i's status bar) [19:24] kirkland: bug [19:26] cjwatson: okay, i'll reproduce and attach logs [19:54] cjwatson, i grabbed parted 2.2-1ubuntu2 out of NEW and manually installed those binaries (and put symlinks in place so parted_server worked still), but same thing and even calling partprobe still is returning an error, along with new device nodes not being created [21:22] cjwatson: is there a way to run a script only on the first install of a deb package? [21:22] and not on subsequent updates? [21:23] yes, check debian-policy for the version argument passed to the postinst [21:27] um... how does one do that... [21:34] shtylman, http://www.debian.org/doc/debian-policy/ [21:34] cody-somerville: yea...I found that document... to say that it helped... well... [22:04] shtylman: it's in the section about how maintainer scripts are called [22:13] cjwatson: yea... I found the order of the parameters and whatnot.. I guess I wan't clear on just how you would check after that... if a simple query against dpkg would do... or if you meant something more complex [22:17] you normally use dpkg --compare-versions [22:17] although actually, first-install doesn't get a version passed to it [22:17] so in fact you'd just test [ -z "$2" ] [22:18] heh... clever :) [22:18] thanks [22:19] my small dabble into the world of deb packages [22:19] will hopefully end well [22:20] what does get me ... is that I can't seem to find a way to tell it where to put the generated debs and other files it makes... (when running debuild I mean) [22:20] maybe I am just not looking in the right places... [22:25] they always go in the parent directory [22:26] and no way to influence that? [22:26] it's not configurable, no [22:26] strange [22:26] some things shouldn't be configurable [22:26] heh [22:26] location of generated files should be :) [22:26] you can always move them with a script if you really care [22:27] there's absolutely no point bloating the core development programs with that [22:27] there's a 'dcmd' program you can use to get the list of files from the .changes file if you want [22:27] heh [22:27] but many tools work better with them in the parent, so honestly I suggest you get used to it :) [22:27] will have to do that... I suppose .. [22:27] for example if they're in the parent directory then you can invoke debc without arguments to list the contents of the thing just built [22:28] or debi [22:28] thats kinda cool [22:44] cjwatson: on a different note... what does bzr builddeb do then? [22:44] cause it puts things in a different directory... [22:44] voodoo magic? [22:44] not especially, it copies the source tree to a different directory and changes to that directory before starting the build [22:45] :(