davmor2 | cjwatson: I thought the default FS for karmic was going to be ext4 not 3? | 08:55 |
---|---|---|
cjwatson | davmor2: I believe we said something along the lines of "we may consider ext4 as the default filesystem after jaunty" | 09:15 |
cjwatson | pretty sure nobody ever committed to it | 09:15 |
davmor2 | cjwatson: Ah okay :) | 09:15 |
cjwatson | ext4 was a bit rockier than expected in jaunty, what with all the filesystem semantics debates | 09:16 |
cjwatson | so I'm not sure I want to take a unilateral decision about giving it a free pass as the default :) | 09:17 |
evand | surely we should just go straight to btrfs :) | 09:18 |
davmor2 | evand: hell yes we can be the kernel lab rats. Oh wait we are Ubuntu/Canonical and don't help the kernel right ;) | 09:24 |
evand | heh | 09:24 |
cjwatson | hmm, that's a point, we have btrfs in the kernel now | 09:32 |
cjwatson | should definitely add installer support for it | 09:32 |
cjwatson | Btrfs is highly experimental, and THE DISK FORMAT IS NOT YET | 09:33 |
cjwatson | FINALIZED. | 09:33 |
cjwatson | ... or maybe not | 09:33 |
davmor2 | cjwatson: :D | 09:34 |
davmor2 | kernel lab rats like I say :) | 09:34 |
evand | We could always do what Redhat did and enable it only when ICANTBELIEVEITSNOTBTR is in /proc/cmdline | 10:04 |
davmor2 | Lol :D | 10:10 |
=== dpm_ is now known as dpm | ||
DogWater | cjwatson: are there any plans for a new spin of the 9.04 netboot install? so far i've noticed several issues with it ranging from partition problems to it locking up, etc. | 13:46 |
cjwatson | err, that would depend on whether anyone knows how to fix those problems | 13:46 |
cjwatson | we respin the netboot installer for kernel ABI changes, not much else | 13:46 |
cjwatson | lockups sound like kernel bugs to me | 13:47 |
cjwatson | other than that ... too vague, sorry | 13:47 |
DogWater | Well, its still randomly saying that the partition table/filesystem is locked on brand new never used hard drives, stuff like that | 13:47 |
cjwatson | bug number? | 13:51 |
cjwatson | FWIW just asking for a respin is totally the wrong way round | 13:51 |
DogWater | Actually I was just trying to determine if there was a scheduled release for a newer version i could try, is all. I wasn't intending on offending you or anyone else. | 13:52 |
cjwatson | you didn't offend me, just saying that it's usually more productive to get the bugs fixed first :-) | 13:55 |
DogWater | Is there a list of the new options for preseeding listed somewhere? | 13:55 |
cjwatson | there is no *scheduled* release, but sometimes releases happen | 13:55 |
DogWater | such as use_weak_password ? | 13:55 |
cjwatson | there's no change summary for preseeding, perhaps unfortunately | 13:56 |
DogWater | ack | 13:56 |
cjwatson | I'm happy to look at incorrect locking claims given a bug report with the usual logs attached | 13:56 |
DogWater | Well, my point is, I dont understand how a partition/filesystem could possibly be locked if the OS isn't running | 13:57 |
DogWater | nothing is using the partition/filesystem until partman (or whatever mechanism you guys use) tries to partition it, so its kind of like chicken -> meet egg | 13:58 |
cjwatson | I need a bug report | 13:58 |
cjwatson | I can't make any even educated guesses without the logs | 13:58 |
DogWater | okay, didn't you tell me once there was a way to find out what option the installer stopped on in preseed? Its asking me to confirm a weak password and rather than outright ask you how to do it automatically i'd much rather see if you could tell me how to answer that question in the future | 13:59 |
cjwatson | I believe it's documented in the installer guide | 13:59 |
DogWater | oh, well thats easier. | 13:59 |
DogWater | thats what i meant by 'are preseed changes documented, etc' | 13:59 |
DogWater | it didn't do that on 8 or 7 so I assumed it was new in 9 | 14:00 |
cjwatson | the changes aren't, but the documentation should cover what you need to do with the current state | 14:00 |
cjwatson | hmm, but apparently doesn't. blah | 14:00 |
cjwatson | DEBCONF_DEBUG=developer as a boot parameter will get you full debconf logs | 14:00 |
cjwatson | FWIW 8.04 and 8.10 are just as different as 8.10 and 9.04 - the 9 is just a year not a series | 14:01 |
DogWater | just as a general suggestion it would be cool if you guys could maybe have a page somewhere that lists new options/changes to kickstart/preseed. It would be cleaner if i could tell it to ignore weak passwords in kickstart but i'm not sure how i'd do that (without using the preseed/user-config, etc option | 14:02 |
DogWater | maybe a little -ignore-weak flag on the password directive, or something in KS | 14:02 |
cjwatson | I'd like to but it needs to be done in Debian | 14:02 |
cjwatson | no point just listing the ones that change in Ubuntu | 14:03 |
DogWater | well, debian doesn't have KS so its even more important that the KS docs are updated, no? | 14:03 |
cjwatson | docs? kickstart doesn't have this *facility* (except using the general preseed command) | 14:04 |
DogWater | okay, thats cool that is what i was suggesting, maybe trying to keep KS updated to match new stuff in preseed, etc | 14:04 |
CIA-4 | installation-guide: cjwatson * r459 ubuntu/ (debian/changelog en/appendix/preseed.xml): Document how to override the weak password check. | 14:04 |
cjwatson | I'm reluctant to add options to kickstart because it could easily conflict with later changes in Red Hat | 14:04 |
cjwatson | Kickstart's syntax is not owned by us | 14:04 |
cjwatson | so we're constrained | 14:04 |
DogWater | right, I see. | 14:05 |
cjwatson | I reckoned we were OK with the new preseed command, but I don't want to stretch it too far | 14:05 |
DogWater | is there a URL to that thing you just updated? :D | 14:05 |
cjwatson | yeah, I'm just getting it for you | 14:05 |
DogWater | thanks sir | 14:05 |
cjwatson | http://bazaar.launchpad.net/~ubuntu-core-dev/installation-guide/ubuntu/revision/459 | 14:05 |
DogWater | With the rate of development I'm surprised a bit that more people aren't using Ubuntu in our datacenter. | 14:07 |
DogWater | People love their CentOS I guess... | 14:08 |
soren | DogWater: IME, they don't actually. They're just stuck on it for some reason. | 14:14 |
DogWater | Its probably that whole "we used redhat until it cost money (rh9) then we scrambled to find something, and CentOS pretty much worked for us" thing | 14:15 |
DogWater | What really surprises me is how many people put fedora on servers | 14:15 |
DogWater | thats nuts to me | 14:15 |
DogWater | People sometimes request Fedora 3 or The old Redhat 9 on machines and we just have to put our foot down somewhere. that is just irresponsible. | 14:19 |
evand | cjwatson: Do you think (https://wiki.ubuntu.com/MergeOEMConfig) is worth scheduling for this UDS? I think this was discussed earlier this month but I was offline for most of the context. I'm not sure if either of us will have the time to take it on, but perhaps someone else would be interested. | 15:51 |
evand | ah: "(07:52:08 PM) shtylman: evand: cjwatson_: ive started moving things around and trying to merge oem-config and ubiquity :)... man... I think I need a shield and broadsword to fight this beast..." | 16:00 |
=== cjwatson_ is now known as cjwatson | ||
shtylman | evand1: yea..I have been working on some major re-org stuff, but nothing yet finished or buildable | 18:09 |
=== CarlFK2 is now known as CarlFK | ||
geiseri | hi, im having problems with a custom package repo on my CD, when i try to run the debootstrap step i get the error: "Invalid Release file: no entry for custom/binary-i386/Packages" but the entry is there in my release file as well as the file is actually there. | 19:00 |
geiseri | is there some part i am missing with respects to my release file for multiple components? | 19:13 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!