/srv/irclogs.ubuntu.com/2009/01/08/#ubuntu-installer.txt

saispohi10:59
saispocjwatson: i don't understand something... when i deactivate proposed the kernel on my installer are 2.6.24-19 and not 2.6.24-22 and i have the debian-installer 0.46... where is the mistake ? i don't want to distribute a cd with proposed activated :x11:00
cjwatsonsaispo: I assume you mean 20070308ubuntu40.6?11:06
saispoyep11:06
cjwatsonsaispo: I'm sorry, though, I can't help you. The fault is very likely in your mirror setup somewhere11:06
saispoin hardy-updates11:06
saispoPackages.gz are right :x11:06
cjwatsonsaispo: check that you have the new dists/hardy-updates/main/installer-*/ directories as well as pool11:06
cjwatsonyou may need to rsync them up by hand11:06
saispowill see that, thanks11:07
davmor2cjwatson: I only had time to do a couple of quick tests on live desktop last night all seems well.  Is encrypted home working on d-i yet and do you need it testing?11:08
cjwatsondavmor2: kirkland told me yesterday that it was working, but the real test should be today since the kernel has changed about a bit, wouldn't hurt to give it a quick steam through11:09
davmor2cjwatson: No Problems I got a bit more time throughout today so I can test a bit better :)11:10
cjwatsondavmor2: actually don't bother there was a udev NEW glitch11:11
cjwatsonso today's CDs are probably busted11:11
saispocjwatson: i have exactly the same11:12
davmor2cjwatson: np's I should be around most of the week so I'll keep trying it :)11:13
cjwatsonsaispo: I've given you all the help I can; since now you know the path where it fetches the kernel, you should have enough information to trace through what's happening11:16
saispoyep11:16
saispothks11:16
saispocjwatson: ubuntu/dists/hardy-updates/main/installer-i386/current/images/MANIFEST.udebs -> tell about -19 and not -22 :x11:20
cjwatsonblink11:24
cjwatsonyes, you're right, let me fix that in the archive11:24
saispo;)11:24
cjwatsonI guess it wasn't copied properly, sorry about that11:24
saispono problem11:28
cjwatsonsaispo: fixed, will be visible on archive.ubuntu.com in about half an hour11:28
cjwatsonlp_publish@cocoplum:/srv/launchpad.net/ubuntu-archive/ubuntu/dists.new$ grep 2.6.24 hardy-updates/main/installer-i386/current/images/MANIFEST.udebs | head -n111:29
saispook, thanks, will wait :)11:29
cjwatson        acpi-modules-2.6.24-22-generic-di 2.6.24-22.45 i38611:29
CIA-3apt-setup: cjwatson * r150 ubuntu/debian/changelog: correct misleading changelog comment11:45
CIA-3apt-setup: cjwatson * r151 ubuntu/ (debian/changelog generators/50mirror.ubuntu):11:46
CIA-3apt-setup: Honour apt-setup/restricted, apt-setup/universe, and11:46
CIA-3apt-setup: apt-setup/multiverse when writing -backports lines (LP: #314462).11:46
CIA-3apt-setup: cjwatson * r152 ubuntu/debian/changelog: releasing version 1:0.37ubuntu711:48
* kirkland is pulling today's iso to test encrypted home14:19
davmor2kirkland: I didn't see the option15:03
kirklanddavmor2: server/alternate installer?15:04
davmor2kirkland: d-i alternate but from what cjwatson said it's a bit borked with a boat load of updates15:05
davmor2once installed15:05
davmor2server might be different though15:05
davmor2kirkland:  (11:11:30) cjwatson: davmor2: actually don't bother there was a udev NEW glitch15:06
davmor2(11:11:34) cjwatson: so today's CDs are probably busted15:06
kirklandbummer15:06
cr3cjwatson: how can I determine whether the netboot files on an alternate image are for the same version as the kernel and modules on the same image?15:06
cjwatsoncr3: they should always be the same15:07
cr3cjwatson: ok, I've been getting the error: No root filesystem is defined. I'll reproduce and upload the syslog and partman files15:08
cr3cjwatson: by the way, I have changed my preseed to use "d-i partman-auto/choose_recipe select atomic" instead of "... All files in one partition", although I think I observed that the installer was backward compatible15:09
davmor2cjwatson: What is it exactly that Udev does15:09
cjwatsoncr3: yes, either should work but your new value is preferred15:09
cjwatsondavmor2: err, apt-cache show udev?15:10
kirklanddavmor2: yeah, the cd does look busted to me :-/15:10
cjwatsonI'm going to build a new CD in a bit; lunch first15:10
davmor2cjwatson: ping me when :)15:10
cjwatsonwe really must get RSS feeds set up or something15:11
cjwatsonI hate having to ping people15:11
davmor2cjwatson: that would be cool :)15:11
cr3cjwatson: http://people.ubuntu.com/~cr3/(syslog|partman|preseed), I can see "INPUT critical partman-target/no_root" but not the reason for it15:13
cjwatsonpresumably because it hasn't detected any disks15:14
cr3cjwatson: I'm an idiot, that was the reason for my original question about the netboot files and the kernel, I was getting an error that they don't match.15:15
cjwatsonoh, you didn't say it was hardy15:15
cjwatsonI'll look at the hardy-proposed images in a bit and see what's up with them15:15
cr3cjwatson: I'm strictly using files from the alternate image though, I'm not going out to archive.ubuntu.com15:16
cjwatsonoh, the seeds are wrong15:17
cjwatsonfixed, will rebuild in a bit15:18
cr3cjwatson: strings ./install/netboot/ubuntu-installer/i386/linux | grep 2.6: 2.6.24-23-generic (buildd@palmer) #1 SMP Thu Nov 27 18:44:42 UTC 200815:18
cjwatsonthe CD would have been screwed if booted normally too15:19
cr3that looks alright to me15:19
cjwatsonit is, the problem is that the CD wrongly contained -22 module udebs15:19
tjaaltonbtw, any news if pgksel will be able to install "Recommends" too? now it only does what is on the list15:19
cjwatsondue to the seeds saying that15:19
cr3cjwatson: ah, thanks for the info, this was confusing because I just did a major refactoring of one part of code and I was worried the problem might be on my side15:19
cjwatsontjaalton: well, it installs Depends. I guess you're talking about pkgsel/include? We should probably make Recommends optional15:19
cjwatsontjaalton: wishlist bug on pkgsel please?15:20
tjaaltoncjwatson: ok, will do15:20
tjaalton*file15:20
davmor2Meh system not restarting on clicking the restart button after update :(15:22
CIA-3partman-ext3: cjwatson * r747 ubuntu/ (15 files in 11 dirs):16:48
CIA-3partman-ext3: Add basic ext4 support (LP: #293465). Some translatable strings are16:48
CIA-3partman-ext3: still incorrect, but that will be fixed once this gets into Debian.16:48
CIA-3partman-auto: cjwatson * r277 ubuntu/ (4 files in 3 dirs): Add ext4 support (LP: #293465).16:50
CIA-3partman-ext3: cjwatson * r748 ubuntu/debian/changelog: not so basic, actually :-)16:50
CIA-3partman-partitioning: cjwatson * r692 ubuntu/ (4 files in 3 dirs): Add ext4 support (LP: #293465).16:51
CIA-3partman-ext3: cjwatson * r749 ubuntu/debian/changelog: releasing version 55ubuntu216:53
CIA-3partman-auto: cjwatson * r278 ubuntu/debian/changelog: releasing version 83ubuntu216:55
CIA-3partman-partitioning: cjwatson * r693 ubuntu/debian/changelog: releasing version 64ubuntu216:57
CIA-3ubiquity: cjwatson * r2962 ubiquity/ (4 files in 2 dirs): Add ext4 support (LP: #293465).16:59
superm1evand, how feasible would it be to have an option to remove extra language packages at the end of OEM config if extra ones were installed?17:25
superm1i was just doing updates for a factory install, and having 716 updates because of all the language packs seems a bit excessive17:26
evandSeems reasonable.  cjwatson, any objections?17:28
cjwatsononly difficulty is that we'd have to port all the tedious python-apt code across to oem-config, but it's possible17:29
=== robbiew1 is now known as robbiew
cjwatsoncr3: new hardy images up, should be happier now18:00
cjwatsondavmor2,kirkland: new jaunty images up, should fix the udev dependency glicth18:00
cjwatsonglitch18:00
kirklandcjwatson: nice, i'll pull18:00
cr3cjwatson: thanks, I'll try it out18:32
CIA-3ubiquity: cjwatson * r2963 ubiquity/ (d-i/manifest debian/changelog):18:41
CIA-3ubiquity: Automatic update of included source packages: apt-setup 1:0.37ubuntu7,18:41
CIA-3ubiquity: hw-detect 1.71ubuntu3, partconf 1.30build1, partman-auto 83ubuntu2,18:41
CIA-3ubiquity: partman-base 128ubuntu4, partman-ext3 55ubuntu2, partman-partitioning18:41
CIA-3ubiquity: 64ubuntu2, partman-target 58ubuntu2, tzsetup 1:0.24ubuntu1, user-setup18:41
CIA-3ubiquity: 1.23ubuntu6.18:41
CIA-3ubiquity: cjwatson * r2964 ubiquity/debian/changelog: releasing version 1.11.218:57
davmor2cjwatson: pulling now :)19:01
davmor2thanks19:01
NCommandercjwatson, hola19:11
kirklandcjwatson: worked like a champ!!!19:41
kirklandevand: I have validated today's server build -- encrypted home option working as expected.  that's the 'go-ahead' from me that the encrypt-home bootstrapping backend is ready for integration into the desktop installer19:42
evandwonderful19:43
cjwatsonNCommander: hi19:58
cjwatsonkirkland: yay19:58
NCommandercjwatson, how goes it?19:58
cjwatsonfine, just spent the day doing ext4 support19:59
kirklandcjwatson: yeah, thanks a lot for re-spinning20:00
CIA-3ubiquity: cjwatson * r2965 ubiquity/debian/ (changelog copyright): Correct Bazaar link in debian/copyright, really this time.20:07
cr3I have a dhcp timeout in my preseed, d-i netcfg/dhcp_timeout string 120, but it doesn't seem like it's waiting for 2 minutes though20:23
cjwatsonhm, I have a feeling that we need to configure the DHCP client to spend longer trying, for very long timeouts20:32
cjwatsontime it with a stopwatch, and file a bug on netcfg if you confirm it20:32
cr3cjwatson: reported bug #315231 against netcfg regarding the dhcp timeout21:31
ubottuLaunchpad bug 315231 in netcfg "netcfg/dhcp_timeout doesn't take effect when installing" [Undecided,New] https://launchpad.net/bugs/31523121:32
davmor2kirkland: which arch did you check?22:28
CIA-3ubiquity: evand * r2966 ubiquity/ (debian/changelog ubiquity/components/partman.py):22:35
CIA-3ubiquity: Fix the edit partition dialog by properly preseeding22:35
CIA-3ubiquity: partman/active_partition.22:35
kirklanddavmor2: i tested server amd6423:08
kirklanddavmor2: it might be interesting to test the alternate cd, and i38623:08
davmor2kirkland: I don't tend to touch server so I did 32bit alternate and it seems to work okay although initial startup of apps that are slow FF OO.o seem slower other than that seems okay :)23:11
kirklanddavmor2: interesting, okay23:11
kirklanddavmor2: i'm working with michael larabel of phoronix for performance benchmarking of the encrypted home stuff23:11
davmor2kirkland: only the initial startup though after that seems relatively comparable with a normal system23:12
kirklanddavmor2: have you tested the same setup, without encrypted home?23:13
davmor2I'm guessing the need to insert .ooo .ff directories23:13
davmor2kirkland: yes live install on yesterdays image23:13
kirklanddavmor2: cool, well, i'll have a look23:13

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!