[01:26] <calc> BenC: i noticed something a bit off with my laptop, may be common though, it works fine for resume in i386 but won't in amd64 when it tries to resume it looks like it almost comes back but the xserver doesn't come up and i can't get to different tty's, i see the resume kernel(?) messages though on the screen
[01:31] <crimsun> BenC: could we build the OSS/Free opl3{,sa{,2}} sound drivers?  Because libasound2-plugins (containing the OSS alsa-lib plugin) is in main, we can easily let users with that hardware use Ubuntu apps
[01:32] <crimsun> no one's stepping forward to fix snd-opl3sa2, so the sabdfl's "have sound work out of the box" may be resolved most easily in this manner
[01:33] <BenC> crimsun: ok, can you file a bug please?
[01:33] <BenC> calc: odd, haven't seen that before
[01:53] <calc> BenC: perhaps i'll let you look at it in London :)
[02:00] <BenC> calc: ok
[04:37] <xipietotec> hey, is there a known issue with suspend and hibernate with the low-latency-kernel?
[06:43] <bremenbeck> hiya... daft question that causes me a whole pile of grief
[06:43] <bremenbeck> I think I'm up to date with the patches and all.. and I know how to edit boot code
[06:43] <bremenbeck> root@harold:~# uname -a
[06:43] <bremenbeck> Linux harold 2.6.20-16-386 #2 Thu Jun 7 20:16:13 UTC 2007 i686 GNU/Linux
[06:44] <bremenbeck> so then this gets in the boot...
[06:44] <bremenbeck> root@harold:~# grep -i default /boot/grub/menu.lst | egrep 'kernel|Ubun' | grep -v \# | head -1
[06:44] <bremenbeck> title           Ubuntu, kernel 2.6.18-028test007 Default
[06:44] <bremenbeck> and not the  2.6.20-16-386
[06:45] <bremenbeck>   /boot/grub/menu.lst seems to always break
[06:46] <bremenbeck> I have to have missed something... but every time I apt-get update, the kernel comes up on that pesky 2.6.18-028test007
[06:46] <bremenbeck> no matter what I do
[06:51] <bremenbeck> I'll try and delete it *again*...  wish me luck
[01:29] <\sh> hmm...why is hugetablefs in feisties kernel not enabled
[01:29] <\sh> ?
[01:31] <\sh> it could be important, if ubuntu server wants to be a good java server platform :) (new feature of java6 largepages support)
[01:42] <\sh> #122800
[02:56] <kraut> moin
[03:04] <bje> good day!
[03:04] <bje> I'm trying to compile linux-kernel-di-i386
[03:05] <bje> But I get "missing module ide-core"
[03:05] <bje> when I do a "fakeroot debian/rules binary"
[03:05] <bje> any hints?
[03:33] <bje> cjwatson: you about?
[04:05] <cjwatson> bje: linux-kernel-di-i386 doesn't work with Ubuntu kernels; the Ubuntu kernels build the udebs themselves
[04:06] <cjwatson> (besides, even linux-kernel-di-i386 in upstream d-i svn was designed for 2.4 kernels; there's a -2.6 variant for 2.6 kernels; but you don't want that either on Ubuntu for the reason above)
[04:07] <bje> cjwatson: I build my kernel like this: 'fakeroot make-kpkg --append-to-version="-claranet" kernel_image kernel_headers', I do not get any udebs.  What am I doing wrong then?
[04:09] <bje> cjwatson: this is the actual problem I'm trying to solve: https://lists.ubuntu.com/archives/ubuntu-server/2007-June/000493.html :)
[04:10] <cjwatson> bje: the udeb building is done by debian/rules, not make-kpkg
[04:10] <cjwatson> so you would need to use the Ubuntu kernel source with 'debian/rules build-images' or similar
[04:10] <cjwatson> (or whatever the rune is on dapper; but it won't involve make-kpkg)
[04:11] <cjwatson> 2.6.20 is going to be extremely hard to make work with dapper; IIRC there were incompatible changes affecting things like udev
[04:11] <kylem> right.
[04:13] <bje> cjwatson: suggestions then for making my RAID controller work with Dapper?
[04:13] <cjwatson> sorry, I am but a humble installer guy; if the kernel doesn't work I tend to punt to these guys
[04:13] <bje> Ah ok, cool. I'm in the right channel, though?
[04:14] <kylem> for the installer, you're kind of boned.
[04:15] <bje> So what exactly is the point of running a LTS then? :)
[04:15] <kylem> custom kernel for your hw + empty udebs is probably the easiest thing.
[04:15] <cjwatson> bje: the only reason we can manage to support dapper for 5 years is that we *aren't* trying to upgrade the kernel constantly ...
[04:16] <bje> I see.
[04:16] <cjwatson> so yes, support is sometimes confined to hardware where it worked at time of release
[04:16] <cjwatson> there has been some talk of possibly backporting a newer kernel to dapper in future
[04:16] <cjwatson> because obviously a number of people have this kind of issue
[04:16] <cjwatson> but that's not in place yet
[04:17] <cjwatson> bje: if you have to have 2.6.20 on dapper, it would probably be less painful to work via the live CD
[04:17] <cjwatson> that way you can experiment more easily with upgrading other components like udev, and you won't have to mess around with udebs
[04:17] <cjwatson> and yes, that would involve a process other than PXE boot / kickstart
[04:17] <cjwatson> (more like install once and clone the disk)
[04:26] <shawarma> kylem: We're getting more and more people in #ubuntu-server asking for backported drivers for various new raid controllers and that sort of thing.. What happened to the the 2.6.15-50.whatever kernel?
[04:27] <kylem> shawarma, i have Real Work(tm) to do...
[04:27] <kylem> that and it became clear after the first time i did it that it was a Really Really Bad Idea.
[04:28] <shawarma> kylem: Erm? Oh, I didn't know it was kind of a personal experiment. I thought it was a serious stab at trying to provide updated drivers.
[04:28] <kylem> it was the first time we'd ever done it.
[04:29] <shawarma> kylem: Right. Let me rephrase: I thought that it itself constituted "real work(tm)"..
[04:29] <shawarma> kylem: but no?
[04:29] <kylem> well, nobody has assigned it to me to do again, so...
[04:32] <kylem> besides, if we don't respin the installers, what's the point?
[04:32] <shawarma> None.
[04:32] <shawarma> I would expect new installer CD's, too.
[04:32] <shawarma> I'll take it up at the meeting today.
[04:34] <kylem> yes, well, you can do the backports then. :P
[04:34] <kylem> you learn quick.
[04:34] <zul> lol
[04:36] <shawarma> What's the idea going forward, then? The issue is only going to get worse. More frequent LTS's?
[04:36] <shawarma> That's going to cost us even more time.
[04:40] <kylem> not my place to say; i'm just relaying my experience.
[04:51] <BenC> shawarma: backporting drivers is tedious and usually not very helpful without CDs...unfortunate as well that we don't have driver-updates feature in dapper
[04:51] <BenC> shawarma: as cjwatson said, we are seriously considering a kernel backport to dapper, but it's very tedious as well
[04:51] <BenC> we have to look at what userspace backporting we'll also need to do
[04:52] <kylem> "lots"
[04:52] <kylem> i think even python updates end up in the dep chain unless we fork packages.
[04:53] <fabbione> never mind the installer bits to cope with the other user land changes
[04:53] <kylem> somehow hal needs the new python version...
[04:53] <fabbione> it's probably easier to take feisty and turn it LTS
[04:54] <shawarma> How does RedHat cope with this?
[04:54] <kylem> shawarma, uh, hundreds of engineers.
[04:54] <fabbione> with big $$$
[04:54] <fabbione> :)
[04:54] <shawarma> kylem: I see.
[04:55] <kylem> well, certainly more than 5.
[04:55] <zul> heh need more community people then ;)
[04:55] <kylem> heh. while i certainly value the community, i don't think we could convince them to work on something so tedious and menial.
[04:55] <zul> kylem: been there done that
[04:56] <kylem> especially not when all the fun is to be had on ubuntu+1
[04:56] <shawarma> kylem: Yeah. Ubuntu-3 is nowhere near as fun :)
[06:24] <Daviey> Can a ninja help me with a kernel bug?
[07:53] <bdmurray> kylem: ping
[07:54] <kylem> yo
[07:55] <bdmurray> Is it just me or is bug 121601 unrelated to the upstream bug?
[07:57] <kylem> dunno
[07:57] <kylem> looks unrelated at first glance
[08:01] <bdmurray> They don't seem to be the same ata driver do they?
[11:25] <xivulo1> mjg59, did you have a go at wubi by any chance?