[00:44] <cyphermox> slangasek: sad because of what I did or sad because of the initial state of it? it's initial state made me sad, too
[00:59] <Kano> whats the command used to create the efi binaries you find at http://de.archive.ubuntu.com/ubuntu/dists/raring/main/uefi/grub2-amd64/current/
[00:59] <Kano> signed is clear, i mean the other ones
[01:36] <slangasek> cyphermox: the initial state, of course :)
[01:38] <Kano> slangasek: do you work on secure boot?
[01:38] <slangasek> Kano: yes.  The unsigned binaries there are generated from the grub2 source package and handled as a custom upload type in .changes
[01:39] <Kano> in that _amd64.tar.gz
[01:39] <Kano> btw. did you notice that kubuntu does not support secure boot
[01:45] <infinity> Kano: We left it up to the flavours to add SB support, rather than doing it across the board.
[01:46] <Kano> its still weird when you get grub but can not start the kenrel
[01:48] <sarnold> hrm, how do the flavours get different sb support than the main distro? I'd have thought the same grub* would be used on all..
[01:48] <Kano> does the default grub installed in secure boot env support booting unsigned kernels?
[01:48] <infinity> sarnold: It's the same grub, not everyone has twiddles seeds and installers to have signed kernels.
[01:48] <Kano> the grub most likely and shim too but not the signed kernel
[01:48] <infinity> s/twiddles/twiddled/
[01:48] <sarnold> infinity: aha, thanks :)
[01:49] <Kano> some might want to use other kernels
[01:51] <Kano> grub2-signed (1.5) quantal; urgency=low
[01:51] <Kano> does not have this check it seems
[01:55] <Kano> launchpad has it...
[02:01] <Kano> whats the tool to create the iso images now?
[02:01] <Kano> it does not seem to work with the normal live config
[02:01] <Kano> i see no references to grub
[02:03] <Kano> for better bsd support: http://people.freebsd.org/~nox/tmp/grub2-paste_180121.patch
[02:04] <Kano> this patch should be added to grub2
[02:04] <Kano> was commited to grub2 a bit later
[02:05] <infinity> If it's committed upstream, raring or later will get it.
[02:05] <Kano> i just dont like bzr
[02:05] <infinity> I'm not sure how deep our concern for bsd support is, as far as SRUing such things back.
[02:05] <Kano> it was just a typo
[02:06] <Kano> i would prefer when grub would use git
[02:06] <sarnold> I'm a bit dissapointed the patch didn't #define the magic number
[02:07] <Kano> checking out bzr takes ages
[02:07] <infinity> Kano: No one's making you use bzr...
[02:08] <Kano> and where is git for grub?
[02:08] <infinity> Kano: If your goal is just to file a bug and provide a patch, you could just grab the source package and work a patch against it.
[02:08] <Kano> its already commited
[02:08] <Kano> i just want to search the commit
[02:12] <Kano> revno: 4556
[02:12] <Kano> must it be
[02:13] <infinity> http://bzr.savannah.gnu.org/lh/grub/trunk/grub/revision/4556
[02:13] <infinity> Looks like.
[02:15] <Kano> is there a better way to show the commit instead of
[02:15] <Kano> bzr diff -r4555..4556
[02:16] <infinity> That's what I use.  There doesn't seem to be a "bzr show".
[02:17] <infinity> Though I guess it would be trivial to alias "bzr show $n" to "bzr diff -r$((n-1)..$n"
[02:17] <stgraber> I've been using "bzr log -p" lately, which shows you the commit message, author and the diff
[02:17] <sarnold> "better" .. maybe bzr log -p ?
[02:17] <Kano> but with git you see the log entry
[02:18] <Kano> ah -pr
[02:19] <Kano> why does bzr always create patches with -p0?
[02:19] <infinity> Indeed, bzr log -pr is pretty much exactly git show.
[02:19] <infinity> Slightly less friendly to find, but whatever.
[02:19] <infinity> Now I know, at least.
[02:19] <Kano> and it is p0 not p1
[02:20] <infinity> Kano: And that, yeah.
[02:20] <Kano> for diff you can use -p1 i see
[02:21] <Kano> whatever you use, please add it
[02:23] <stgraber> if you have bzr-git installed, you can also do "bzr diff --format=git" which will give you an output identical to that of "git diff" (so -p1 by default)
[02:37] <Kano> btw. that was the original bug report: http://people.freebsd.org/~nox/tmp/grub-bugmail.mbox.txt
[02:38] <Kano> but it seems it never appeared on the mailing list
[02:49] <Kano> (bzr log -r 4556;echo;bzr diff -r4555..4556 --format=git)
[02:49] <Kano> maybe something like that
[13:39] <bdrung> tumbleweed: will you take care of the testing for bug #1068019?
[19:06] <gkcn> #ubuntu-x
[19:21] <rawplayer> hello
[19:21] <rawplayer> how can i remove the black line after removing the top bar from unity-greeter?
[19:22] <rawplayer> maybe its an idea to provide an option for the gsettings schema related to unity-greeter to hide the bar or disable it entirly
[20:10] <Bluefoxicy> sigh this is ridiculous.
[20:10] <Bluefoxicy> we need a system like dkms for friggin' nginx
[21:16] <TheLordOfTime> Bluefoxicy, uh... what?
[21:16]  * TheLordOfTime only responded because 'nginx' is on highlight for him
[21:48] <Bluefoxicy> TheLordOfTime:  there is no dynamic module loader for nginx
[21:48] <Bluefoxicy> want to get a module into Debian/Ubuntu?  You have to convince them to include it and build nginx with the module.  Possibly as an additional package.
[21:49] <Bluefoxicy> TheLordOfTime: if you want to generally include modules with nginx, you need an nginx package that brings in the source code; and you need a package that knows about all the module packages brought in and rebuilds nginx with all that stuff as a trigger when you install or update any of them.
[21:58] <TheLordOfTime> Bluefoxicy, err, yeah, that's how they designed it...
[21:59] <TheLordOfTime> Bluefoxicy, and afaik there's no dynamic module loading slated for any release in the near future
[21:59]  * TheLordOfTime double checks upstream trac
[21:59] <TheLordOfTime> Bluefoxicy, that, and if such a system is created, it has to be created upstream first.
[21:59] <TheLordOfTime> which afaik will result in a complete redesign
[22:00] <Bluefoxicy> yes
[22:00] <TheLordOfTime> that, and since nginx is universe here...
[22:00] <Bluefoxicy> which is why we need something like dkms if we want to include outstream modules
[22:00] <Bluefoxicy> it's ridiculous.
[22:00] <TheLordOfTime> Bluefoxicy, let me reiterate nginx is universe.
[22:01] <Bluefoxicy> Yes I saw that part.
[22:01] <TheLordOfTime> so unless upstream redesigns nginx (which is unlikely to happen) there's not going to be a dynamic nginx module loader
[22:01] <Bluefoxicy> I'm not entirely sure why.
[22:01] <Bluefoxicy> ...
[22:01] <TheLordOfTime> because the way nginx is designed does not currently support such things.
[22:01] <Bluefoxicy> were you not paying attention?
[22:01] <TheLordOfTime> and nginx is in universe because its community maintained
[22:02] <TheLordOfTime>  if you want to generally include modules with nginx, you need an nginx package that brings in the source code; and you need a package that knows about all the module packages brought in and rebuilds nginx with all that stuff as a trigger when you install or update any of them.  <-- aka "do it by hand"
[22:02] <TheLordOfTime> Bluefoxicy, such a package as well i *think* would need a conflicts: nginx-*
[22:02] <Bluefoxicy> yes
[22:02] <Bluefoxicy> that's not a redesign
[22:02] <TheLordOfTime> or at lleast nginx, nginx-full, nginx-light, nginx-...
[22:02] <TheLordOfTime> Bluefoxicy, if you make a package, i'll test it
[22:02] <Bluefoxicy> that's a pile of stupid crap to work around the non-existent dynamic loader
[22:02] <TheLordOfTime> but until you make such a package... its an upstream-resolve issue.
[22:02] <Bluefoxicy> nod
[22:03] <TheLordOfTime> so... *goes back to standard bug triaging*
[22:03] <Bluefoxicy> Poor design considerations aside, I'm not sure why nginx isn't considered a main package
[22:04] <Bluefoxicy> (though I may be biased, since I found if I replace Apache on a production Web server with nginx it only needs 2GB of RAM instead of 8GB ... Apache itself uses 5.8GB of RAM for 250 concurrent connections serving static files with SSI)
[22:11] <TheLordOfTime> trust me, its good where it is in universe.
[22:12] <TheLordOfTime> Bluefoxicy, the same question could be made about lighttpd
[22:12] <TheLordOfTime> since its also widely used
[22:12] <TheLordOfTime> i just gave up the arguing.  its irrelevant - since most NGINX bugs end up being SRU'd by me when upstream has patches/fixes
[22:13] <TheLordOfTime> :P
[22:13] <TheLordOfTime> (at least nowadays they're sru'd by me... ;P)
[22:42] <Bluefoxicy> TheLordOfTime:  you know if there's any truth to the rumor that MySQL is being replaced?
[22:43] <Bluefoxicy> I have been hearing that it's being seriously discussed to throw it completely out of Debian, Ubuntu, Fedora, etc.
[22:52] <TheLordOfTime> Bluefoxicy, why don't you ask the release team?
[22:52] <TheLordOfTime> or debian
[22:52] <TheLordOfTime> :P
[22:53] <Bluefoxicy> heh
[22:53] <Bluefoxicy> well the MySQL release team says there are no replacements for MySQL and that all distros who are using MySQL will continue to use and supply it because there is nothing better.
[22:54] <Bluefoxicy> :P
[22:54] <Bluefoxicy> But of course Oracle would say that.
[22:54] <jtaylor> given that mariadb is a recent fork it is certainly able to replace it
[22:55] <TheLordOfTime> jtaylor, is the sql syntax still the same?
[22:56] <jtaylor> of course
[22:57] <Bluefoxicy> http://www.zdnet.com/opensuse-also-considers-switching-from-mysql-to-mariadb-7000010223/ hmm latest news i can find
[22:58] <TheLordOfTime> i ask because CREATE DATABASE dbname; works in mysql and mssql, but not oracledb unless you have a lot of other stuff with it
[22:59] <Bluefoxicy> TheLordOfTime: i in-place replaced mysql with percona so...
[23:22] <TheLordOfTime> are any of the forks in raring yet?
[23:28] <Bluefoxicy> which folks?
[23:28] <Bluefoxicy> Us folks?
[23:28] <Bluefoxicy> oh, forks.  Can't read.
[23:28] <TheLordOfTime> :P
[23:28] <Bluefoxicy> TheLordOfTime:  apparently Debian doesn't have any of them.  Which is strange to my senses
[23:28] <Bluefoxicy> I mean it's like, you know.
[23:28] <TheLordOfTime> might be stuck in the freeze
[23:28] <TheLordOfTime> or in the "We can't process because frozen" queue
[23:29] <Bluefoxicy> they supply repositories which I'm pretty sure have a deb-src
[23:29] <Bluefoxicy> yeah debian is weird about freezing sid
[23:29] <TheLordOfTime> Bluefoxicy, they don't have to
[23:29] <TheLordOfTime> :P
[23:29]  * TheLordOfTime points at his local private debian repositories which only have the binaries
[23:29] <Bluefoxicy> Well I mean, you can just pull the debs from somewhere else and build them yourself, and import them
[23:30] <TheLordOfTime> note two things: (1) private repositories, and (2) they're private and binary-only because i'm paranoid about my source code.
[23:30] <Bluefoxicy> it's not like there's licensing problems
[23:30] <TheLordOfTime> true
[23:30] <Bluefoxicy> define 'paranoid about your source code'
[23:30] <Bluefoxicy> you're a crappy programmer?
[23:30] <TheLordOfTime> no, just paranoid.
[23:31] <Bluefoxicy> that seems strange in this crowd
[23:31] <TheLordOfTime> :P
[23:31] <Bluefoxicy> i'm not sure your source code can be used against you effectively
[23:35] <Bluefoxicy> man i still don't understand what is going on with upgrading from 2.7 to 3.0 puppet on ubuntu.
[23:36] <Bluefoxicy> every time I move up, apt breaks everything.