[12:56] <zul> hey
[01:51] <bddebian> Howdy folks
[01:53] <lifeless> who here is a total X guru ?
[01:53] <lifeless> I have a repeating issue after suspect with ice auth becoming non responsive :(
[01:53] <lifeless> And I'd like to know what info to get, to file a useful bug on it.
[02:00] <jdub> Keybuk: how are debian developers informed of totally new packages in our repos atm? (or not)
[02:10] <Keybuk> jdub: they're not
[02:13] <bddebian> That was going to be my guess but I didn't want to be wrong as always :-)
[02:16] <jdub> deb http://people.ubuntu.com/~jdub/edgy/ /
[02:16] <jdub> ^ beaglefs
[02:16] <jdub> wait, don't play with that yet
[02:19] <tseng> bddebian: woo!
[02:19] <tseng> bddebian: what are you weekend plans
[02:23] <jdub> okay
[02:23] <jdub> now you can play with it
[02:24] <tseng> ok.
[02:24] <jdub> it helps if your .deb contains the binary file it purports to
[02:24] <bddebian> tseng: Dunno for sure.  I know my wife is going to the American Idol concert Sat. night. (puke)
[02:24] <tseng> bddebian: double puke
[02:24] <tseng> I enjoy the American Idol auditions
[02:25] <tseng> where the guy is a major dick to everyone
[02:25] <bddebian> Yeah, and that's about it :-)
[02:25] <tseng> yep.
[02:25] <tseng> bddebian: Wachovia Center?
[02:26] <bddebian> I think so but I'm not sure
[02:26] <tseng> nice place
[02:29] <Fjodor> Erm, it would seem gksu doesn't see that it should default to sudo mode as per gconf in amd64. gksu <something> prompts for password, even though there is a NOPASSWD entry, gksu -S doesn't
[02:44] <whiprush_> anyone else notice significant battery life improvements with the last kernel in edgy?
[02:45] <Hobbsee> whiprush_: dont know about that - i've been doing a lot of building, but the laptop certainly seems to run better
[02:46] <whiprush_> I was having pretty horrible battery performance, but thought it was just my battery going bad, but it seems to be much better since the last few days of updates.
[04:05] <Hobbsee> Kamion: argh, almost assigned the archive again, but i remembered this time :P
[04:15] <fabbione> morning guys
[04:15] <Hobbsee> hi fabbione!
[04:15] <fabbione> hi Hobbsee 
[04:15] <bddebian> Hello fabbione
[06:04] <jdub> open("/var/lib/aptitude/pkgstates", O_RDONLY) = 3
[06:04] <jdub> fcntl64(3, F_SETFD, FD_CLOEXEC)         = 0
[06:04] <jdub> fstat64(3, {st_mode=S_IFREG|0644, st_size=140622, ...}) = 0
[06:04] <jdub> fstat64(3, {st_mode=S_IFREG|0644, st_size=140622, ...}) = 0
[06:04] <jdub> fstat64(3, {st_mode=S_IFREG|0644, st_size=140622, ...}) = 0
[06:04] <jdub> mmap2(NULL, 140622, PROT_READ, MAP_SHARED, 3, 0) = 0xb7073000
[06:04] <jdub> 
[06:04] <jdub> ^ stracing aptitude search
[06:04] <jdub> at this point it sits around using 100% cpu
[06:04] <Chipzz> ; Please register your domains at
[06:04] <Chipzz> ; http://www.cheapass.be
[06:31] <pitti> Good morning
[06:33] <Hobbsee> hi pitti 
[06:38] <bluefoxicy> holy crap, the networking dialog looks really dumbed down o_o
[06:41] <pitti> no wonder I couldn't connect for 10 minutes or so
[06:41] <Hobbsee> pitti: heh, it's borked
[06:42] <Amaranth> Get off freenode spec, anyone? :P
[06:44] <pitti> Amaranth: OFTCMigrationSpec? :)
[06:45] <fabbione> it's about time to implement that
[06:46] <FunnyLookinHat> Heh
[06:54] <whiprush> jdub: ping
[06:59] <jdub> whiprush: ppong
[07:00] <whiprush> jdub: I know you're on your way out the door, but my friend at google has confirmed that they intend on hosting the ubucon right after lwe.
[07:00] <jdub> whiprush: hrm - i wonder who will actually go, apart from you guys and google guys.
[07:00] <whiprush> he got a response from the guy and he's going to send a mail out this week ...
[07:00] <jdub> whiprush: and i wonder if anything is actually going to happen at it.
[07:00] <jdub> ugh!
[07:00] <whiprush> jdub: well, I'm thinking, it's in socal, surely a post on planet and a digg or two could possibly make it rock.
[07:00] <sharms> whiprush: pm
[07:01] <jdub> whiprush: yeah, possibly - 'specially san fran
[07:02] <whiprush> sharms: I'm responding to you but I seem to be getting freenoded'd, can you IM me at jorge.castro@gmail.com?
[07:02] <whiprush> fabbione: well, the guy in charge of LWE wanted to do an ubuntu thing after the show
[07:02] <whiprush> so he announced it, and then disappeared.
[07:02] <jdub> whiprush: 'in charge of lwe'?
[07:02] <jdub> i thought he was just some random dude
[07:02] <jdub> who had good links with google
[07:02] <whiprush> jdub: not afaik, let me find a link.
[07:05] <whiprush> jdub: I see him all over the linuxworld blog, john mark walker ...
[07:05] <whiprush> http://www.blogger.com/profile/18000952
[07:06] <whiprush> google says "director of the LinuxWorld Expo conference program"
[07:06] <jdub> ah, of the conf programme
[07:06] <jdub> ok
[07:07] <whiprush> So, I found out about the ubuncon on digg, but past the announcement, there has been zero activity on his site or on his list, and he doesn't respond to emails.
[07:07] <whiprush> but it's on things like the LWN schedule and whatnot
[07:07] <whiprush> and I was planning on going, so I was thinking if, people can show up, maybe we can make something happen.
[07:07] <whiprush> since apparently google has committed to hosting the event.
[07:08] <jdub> yeah
[07:08] <jdub> he hasn't replied to jane or i either
[07:09] <whiprush> but, I've been rolling up friends, and we have at least a day's worth of ubuntu programming we can do, including the guys who did Ubuntu hacks, so if someone promised google a con, then maybe we can rescue this mess.
[07:10] <jdub> rad
[07:10] <whiprush> jdub: to cut to the chase, I was leaning on "hey, you know chris, can you call and find out what I need to do?" :)
[07:10] <jdub> oh man, you can totally email him
[07:10] <jdub> he's very helpful
[07:10] <jdub> or
[07:10] <jdub> i could bug him at oscon
[07:11] <whiprush> it's a short notice, but I'm pretty sure we can pull it together, especially in such a geekladen environment like mountain view.
[07:11] <jdub> them thar mountains
[07:11] <jdub> laden with geek
[07:11] <whiprush> jdub: that would work out perfect. :D
[07:12] <whiprush> jdub: if chris is worried about content, then let him know I've got a pocketful of arsians ready to rock and roll wrt. ubuntu content.
[07:13] <whiprush> plus I got booth duty at LWE so I'll be making sure I get the word out.
[07:15] <whiprush> fabbione: if you want to make the trip I can schedule you to do some X and kernel talks. :)
[07:18] <fabbione> whiprush: when that would be?
[07:19] <whiprush> august 14-19
[07:19] <whiprush> ubucon is the last 2 days, apparently.
[07:20] <fabbione> no i can't sorry
[07:21] <whiprush> it's all a corporate thing anyway, LWE, it's best to save the funds for more hacker-friendly events.
[07:21] <fabbione> i have a wedding the 19 and the 20 we need to be in Germany for the hack sprint
[07:23] <whiprush> fabbione: to be honest, most shows here in the US suck, so you're not missing much.
[07:23] <whiprush> but jdub would be better off describing them than me. :D
[07:23] <fabbione> well let me decide that
[07:23] <fabbione> that as in what sucks and what not
[07:25] <whiprush> fabbione: there is a newer show here in the midwest US, the ohiolinuxfest, which rocks, it's a one day event, I can send you some info via mail if you're interest, sep 30.
[07:27] <fabbione> whiprush: nah thanks
[07:27] <fabbione> a one day event surrounded by 2 days travel is like kicking somebody in the testicles
[07:27] <fabbione> with the difference that the testicles will stop to hurt after 5 minutes
[07:28] <whiprush> fabbione: yeah, I know you're from another country, but I do my best to find content for the show, so I gave it a shot. :D
[07:29] <fabbione> eheh sure
[07:34] <jsgotangco> hahaha
[08:51] <G0SUB> pitti: hello
[08:51] <pitti> hey G0SUB, how are you?
[08:51] <G0SUB> pitti: i am fine, i mailed you just now. please check
[08:52] <pitti> G0SUB: if you want to lock into a small room with just a computer and pizza to hack for a week, that's fine :)
[08:53] <G0SUB> pitti: that's what I want. really
[08:53] <pitti> G0SUB: a meeting is not required, only if you want to discuss about something
[08:53] <G0SUB> pitti: i am badly behind schedule, so there is n other way.
[08:53] <G0SUB> pitti: i have already talked to pygi wrt the GUI. he will assist me in getting it right when I am ready
[08:54] <pitti> G0SUB: ok; the backend should really work in a week
[08:54] <G0SUB> pitti: yes, it will. with the command line frontend
[08:56] <G0SUB> pitti: ok, so I will leave now. see you.
[08:59] <pitti> G0SUB: see you, happy hacking!
[09:30] <dholbach> good morning
[09:30] <Mithrandir> 'morning, Daniel
[09:31] <dholbach> hey Tollef - how are you?
[09:31] <Mithrandir> good, I can feel the knot loosening up now.
[09:31] <Mithrandir> ;-)
[09:32] <sladen> makes it sound like constipation!
[09:32] <pitti> Mithrandir: 'Straight-Cord-1'? :)
[09:33] <dholbach> :)
[09:33] <Mithrandir> pitti: ;-)
[09:34] <Kamion> Mithrandir: shout if publish-release hates you
[09:34] <fabbione> hey Mith
[09:34] <fabbione> hey Kamion 
[09:34] <Kamion> (morning)
[09:34] <pitti> Hi Colin, how are you?
[09:34] <Kamion> hot
[09:34] <Kamion> and not in a good way
[09:34] <Mithrandir> Kamion: seems good so far.
[09:34] <fabbione> Kamion: when you have time could you be so kind to NEW libopenais2 and move libvolumeid* to main (the latter being part of udev source)
[09:35] <fabbione> Kamion: they are both B-D of the redhat-cluster-suite
[09:35] <Kamion> ok
[09:35] <fabbione> Kamion: it's not urgent
[09:35] <fabbione> even after knot is fine
[09:35] <Kamion> I'll just do a NEW pass now, it's been a bit neglected
[09:35] <fabbione> i still need the new headers for the suite to be able to build
[09:41] <Mithrandir> hmm, did we want to do an ubuntu-server knot-1 release too?
[09:49] <Kamion> Mithrandir: why not
[09:49] <Kamion> oh, you know how to publish server, don't you?
[09:49] <Mithrandir> Kamion: yeah.  We do want to build it first, though.
[09:49] <Kamion> oh
[09:50] <Kamion> I'll do that
[09:50] <Mithrandir> thanks
[09:50] <Mithrandir> hmm
[09:52] <Kamion> fabbione: does that /etc/ld.so.conf.d/openais.conf thing actually work?
[09:52] <fabbione> Kamion: yes
[09:52] <Kamion> interesting
[09:52] <Kamion> weird, but interesting
[09:52] <Mithrandir> Kamion: ld.so.conf.d is something we've dragged in from Debian, iirc.
[09:52] <fabbione> cat  ld.so.conf
[09:52] <fabbione> include /etc/ld.so.conf.d/*.conf
[09:52] <pitti> /etc/ld.so.conf.d/$machine-$os ??
[09:52] <fabbione> Kamion: yeah i did check it too when i first saw it
[09:52] <pitti> ^ that looks like quoted once too much
[09:53] <pitti> s/much/often/
[09:54] <fabbione> pitti: i have seen a couple of different way to call those conf files
[09:54] <fabbione> pitti: also $source-$arch.conf
[09:54] <pitti> fabbione: yes, but a literal '$machine-$os' as filename looks wrong
[09:55] <pitti> looks like it was intended to be 'linux-amd64' or so
[09:55] <fabbione> pitti: yeps.. you really want $binary.conf or something
[09:56] <fabbione> other 9 accepted upstream
[09:57] <Kamion> fabbione: new/override changes done
[09:57] <fabbione> Kamion: thanks a lot
[09:58] <fabbione> Kamion: would you object a lot for a UVF exception for the redhat-cluster-suite? the only changes are our local patches that have made their way upstream and 2/3 bug fixes i did backport
[09:59] <fabbione> it's basically the same source with 2 different names...
[09:59] <fabbione> s/names/versions
[10:00] <Mithrandir> ogra: can you look at http://cdimage.ubuntu.com/edubuntu/daily-live/current/ ?  I don't think the header text is correct for edubuntu?
[10:05] <Kamion> hmm, yes, that should have the old live CD text really
[10:06] <Mithrandir> do you have that around somewhere?
[10:06] <Kamion> yes
[10:08] <Kamion> Mithrandir: make-web-indices again and you'll get it
[10:08] <Kamion> or, publish-release will do it
[10:08] <Mithrandir> well, I've already run publish-release for edubuntu
[10:11] <Kamion> ok, 'for-project edubuntu make-web-indices /srv/cdimage.no-name-yet.com/www/full/edubuntu/releases/edgy/knot-1 edgy' should fix it then
[10:11] <Mithrandir> yeah, just figured that out.
[10:12] <Mithrandir> uh, the edubuntu live cd doesn't actually give you the option of installing, does it?
[10:12] <Kamion> should do?
[10:12] <Kamion> it has ubiquity on it
[10:12] <Mithrandir> oh, ok.
[10:12] <Mithrandir> I thought it didn't.
[10:12] <Kamion> it'll only install a workstation, but shrug
[10:12] <Kamion> it's not promoted in the same way as the other derivatives because it doesn't let you install an LTSP server or whatever
[10:13] <Mithrandir> yeah
[10:21] <Mithrandir> Kamion: ubuntu-server ready to be published?
[10:22] <Kamion> sladen: xaralx binaries accepted
[10:23] <fabbione> Mithrandir: unlikely.. 
[10:23] <fabbione> Mithrandir: but you can try if you want
[10:23] <Kamion> Mithrandir: ports didn't build
[10:23] <Kamion> but the main build should be fine
[10:24] <Kamion> (cdimage-wise)
[10:24] <dholbach> morning mdz
[10:24] <mdz> morning
[10:25] <Mithrandir> Kamion: hmm, ok.  They're supposed to be published in their own tree or into the ubuntu tree now?
[10:26] <Kamion> into the ubuntu tree
[10:27] <Kamion> so publish-release daily ../ubuntu-server/daily/20060720 ...
[10:27] <Kamion> must simplify that at some point
[10:27] <Mithrandir> for-project ubuntu publish-release daily ../ubuntu-server/daily/20060720 server no should work, then
[10:28] <Mithrandir> (yes, I know for-project ubuntu is a no-op, but I still like to do it)
[10:28] <Kamion> nod
[10:28] <Kamion> add knot-1 to the end of that command
[10:28] <Mithrandir> uh, true.
[10:28] <Mithrandir> I wonder where that went.
[10:39] <pitti> G0SUB: still here?
[10:42] <G0SUB> pitti: yes?
[10:42] <pitti> G0SUB: nevermind, sorry (I had a python-apt question, but I found it out now)
[10:42] <G0SUB> ok
[11:02] <Mithrandir> maswan: any chance I could ask you to run a mirror sync?
[11:02] <pitti> hey knopper 
[11:04] <Kamion> infinity_,BenC,iwj,doko_,ogra,seb128,Mithrandir: you guys are all even later than me at filling out DistroTeamMeeting20060720 :-)
[11:04] <Kamion> (others too, but they aren't here)
[11:04] <Mithrandir> Kamion: yeah, I suck, I know.
[11:05] <seb128> ah, right
[11:06] <Mithrandir> seb128: does knot-1 have a new gnome?
[11:06] <torkel> Mithrandir: I don't think he is awake yet
[11:06] <dholbach> Mithrandir: it has 2.15.4 (minus some modules, which didn't make it)
[11:07] <dholbach> torkel: he is
[11:07] <Mithrandir> Riddell: does your knot-1 have anything particularly interesting you want to note?
[11:07] <Mithrandir> dholbach: ok, thanks.
[11:07] <dholbach> 2.15.90 next week - YAY!
[11:07] <Kamion> Mithrandir: https://wiki.ubuntu.com/Testing/Kubuntu/Edgy/Knot1
[11:08] <torkel> dholbach: ok. He has not showed up at the office yet though :-)
[11:08] <Mithrandir> Kamion: that was an odd URL to put such information at, but ok.
[11:08] <Kamion> Mithrandir: I agree, just happened to find it by a search a moment ago
[11:09] <seb128> Mithrandir: 2.15.4 yep, and GTK 2.10 :)
[11:09] <dholbach> torkel: seb128 talked to me an hour ago already
[11:09] <Kamion> whoa
[11:09] <Kamion> https://launchpad.net/distros/ubuntu/+spec/ubuntu-server-tasks
[11:09] <Kamion> when did that dependency tree appear?
[11:09] <seb128> torkel: speaking of me?
[11:09] <dholbach> whoa, nice
[11:10] <torkel> dholbach: I was speaking of maswan...
[11:10] <dholbach> torkel: ok, sorry then :)
[11:10] <Mithrandir> torkel: ah.  Can you trigger the mirror for us?
[11:10] <torkel> dholbach: np :-)
[11:11] <torkel> Mithrandir: nope. sorry. I don't have super cow powers at ACC...
[11:11] <maswan> Ok, guys, I just woke up. The mirror run is at :13 though, so I should rather rush off to work than make the sync go 2 minutes earlier. :)
[11:11] <Mithrandir> maswan: ah, ok.  You're mirroring hourly now?  (Including cdimage?)
[11:12] <maswan> Mithrandir: Oh, no, just archive.
[11:12] <maswan> Mithrandir: You want a cdimage sync? And/or release sync?
[11:12] <torkel> maswan: ah, you are awake. Not need to call you then... :-)
[11:12] <pitti> maswan: any chance to run at :43 instead of :13? lp_archive usually finishes around :30
[11:12] <Mithrandir> maswan: cdimage, please.
[11:13] <pitti> maswan: :43 would reduce the lag of security updates
[11:13] <Mithrandir> maswan: no need for release syncs, unless I've seriously fucked up. :-P
[11:13] <maswan> pitti: Ok, done. We're likely to move security back to the DC soon though. I just need to wake up, get to work, do work stuff, then talk to Znarl a bit.
[11:13] <Kamion> pitti: no it doesn't
[11:13] <Kamion> takes much longer than :30
[11:13] <pitti> maswan: thanks
[11:13] <pitti> Kamion: oh, since when?
[11:13] <Kamion> :50 would be a better bet
[11:13] <Kamion> for ages
[11:14] <pitti> *grump* some weeks ago :30 to :35 was fine
[11:14] <maswan> :53 now
[11:14] <Kamion> maswan: sounds good, thanks
[11:14] <pitti> Kamion: thanks
[11:14] <pitti> maswan: merci
[11:15] <Kamion> cron.daily mails usually arrive between :44 and :47 at the moment, but of course you then have to allow a bit of time for internal mirroring
[11:15] <Kamion> I'm not sure exactly when that finishes
[11:18] <seb128> Kamion, mdz: do I need an UVF mail to update xchat-gnome to a new version? It's mainly bug fixes, it adds a "unread line" as xchat and support passwords to join a chan from the UI now
[11:20] <Kamion> does xchat-gnome follow the GNOME release schedule?
[11:20] <seb128> no
[11:20] <seb128> they tend to roll a tarball every month
[11:20] <seb128> but they have no freeze, etc
[11:23] <Mithrandir> http://err.no/tmp/knot-1.txt ; please proofread.
[11:23] <seb128> Kamion: I'm fine sending an UVF mail to make your job easier if you want to have a look on changes for it
[11:24] <Kamion> seb128: no, looks ok to upload
[11:24] <seb128> ok, thank you
[11:24] <Kamion> Mithrandir: s/caldron/cauldron/ normally although I suppose it might have been an old spelling ...
[11:24] <StevenK> Mithrandir: Kernel version is not 2.17
[11:25] <Kamion> Mithrandir: s/primary changes from Dapper has been/primary changes from Dapper have been/
[11:25] <StevenK> Mithrandir: Maybe a comma after Ubuntu in "In Ubuntu GNOME has been updated to 2.15.4 ..."
[11:25] <Kamion> yeah, I was about to say tat
[11:25] <Kamion> that
[11:26] <Kamion> Mithrandir: "Notable Kubuntu changes are noted" sounds bad. "Notable Kubuntu changes are listed" perhaps
[11:26] <Kamion> Mithrandir: "malone" => "Malone:"
[11:27] <Kamion> Mithrandir: has the default theme been fixed? I haven't tried the most recent round of images
[11:28] <Riddell> looks good with Kamion's change, thanks Mithrandir 
[11:31] <iwj> LaserJock: pong
[11:33] <Kamion> ogra: I've renamed server to server-ship. When you merge it, be careful; you want to end up with a server-ship seed that has the same contents and bzr id as the server-ship seed in Ubuntu, and a server seed that has the same contents as what you have in Edubuntu now
[11:34] <Kamion> ogra: do you want me to do that merge? getting it such that future merges are clean might be delicate
[11:37] <Kamion> ogra: actually, I've got a merge done locally anyway, I'll just commit that ...
[11:37] <Mithrandir> StevenK: fixed.
[11:37] <Riddell> Kamion: does that apply to kubuntu too?
[11:38] <Mithrandir> Kamion: fixed.
[11:38] <Kamion> Riddell: no
[11:38] <Kamion> Riddell: you can just do an ordinary merge and it'll DTRT
[11:38] <StevenK> Mithrandir: What about the second change I suggested?
[11:38] <Kamion> Riddell: but Edubuntu already has its own server seed which I'm now divorcing from the Ubuntu server seed, since they have different meanings
[11:39] <Riddell> ok
[11:39] <Mithrandir> StevenK: I noticed it myself.
[11:40] <StevenK> Mithrandir: s/\(Malone\)/\1:/
[11:41] <Mithrandir> thanks, fixed.
[11:42] <StevenK> Mithrandir: Looks fine to me.
[11:42] <Mithrandir> thanks.
[11:42] <StevenK> And that Shakespeare quote. Ouch.
[11:43] <Mithrandir> well, I'm not going to mangle Shakespeare. :-)
[11:43] <Kamion> StevenK: ouch?
[11:43] <Mithrandir> Kamion: caldron, I think.
[11:43] <StevenK> It's been a while since I read Macbeth, I seem to recall that quote not being so brutal-sounding.
[11:44] <Kamion> Mithrandir: my version has cauldron. :-)
[11:44] <Kamion> Mithrandir: oh, "howlet" -> "owlet"; that's what my version has and it makes a lot more sense to modern readers
[11:44] <Mithrandir> Kamion: URL so I can C&P?
[11:45] <Kamion> Mithrandir: http://www.shakespeare-online.com/plays/macbeth_4_1.html
[11:45] <StevenK> I was half tempted to grab the dead tree copy my wife more than likely has around here.
[11:45] <Kamion> I'm sure we have one but it would take a while to dig it out
[11:46] <Mithrandir> Kamion: ok, better now?
[11:46] <Kamion> Mithrandir: much, thanks
[11:55] <ogra> Kamion, thanks for the merge
[12:04] <mdke> so bugs should be filed on edgy now, rather than on ubuntu?
[12:04] <mdke> "There are currently no open bugs."
[12:04] <dholbach> filing on ubuntu is fine
[12:05] <mdke> dholbach: which is preferred?
[12:05] <dholbach> if you have a bug that needs fixing in edgy and dapper, you'd file it on ubuntu and add a dapper task
[12:06] <mdke> right now it seems that only 18 bugs have a dapper task
[12:06] <seb128> mdke: no need to add a dapper task if we don't intend to backport a fix
[12:07] <dholbach> mdke: we only fix serious stuff in dapper
[12:07] <mdke> yes.
[12:07] <mdke> so bugs in edgy should have an Ubuntu task, or an Ubuntu Edgy task?
[12:07] <seb128> ubuntu
[12:07] <mdke> Mithrandir: ^^
[12:07] <seb128> ?
[12:08] <mdke> the bug report link on the knot release announcement needs to be adjusted
[12:08] <Kamion> ur. bugger
[12:08] <Mithrandir> mdke: hmm?
[12:08] <Kamion> Mithrandir: you haven't sent that yet have you?
[12:08] <Mithrandir> Kamion: not yet, waiting for the nl mirror to sync.
[12:09] <Mithrandir> Kamion: why?
[12:09] <Kamion> Mithrandir: https://launchpad.net/distros/ubuntu/edgy/+filebug -> https://launchpad.net/distros/ubuntu/+filebug
[12:09] <Kamion> the /edgy/ there does obsolete things
[12:09] <Mithrandir> hm, ok.
[12:09] <Mithrandir> changed now
[12:10] <mdke> I wonder if +bugs would be a better link
[12:10] <Mithrandir> *shrug*; either is fine with me.
[12:10] <mdke> it might encourage people to search for dups
[12:11] <Mithrandir> that's a nice theory, if nothing else. :-P
[12:11] <Mithrandir> (changed)
[12:11] <pygi> ...so he doesn't have to look :P
[12:11] <seb128> pygi: there is none
[12:12] <seb128> pygi: if you have a question ask to dholbach or me 
[12:12] <pygi> seb128: there is a new patch which removes 40-60MB of memory usage,...but you probably already know that
[12:12] <seb128> mmap?
[12:12] <seb128> no way
[12:13] <pygi> :P
[12:13] <seb128> we have enough bugs without using an alpha patch with many issue changing the way evolution stocks its datas and which might be an issue on NFS by example
[12:13] <seb128> you deserve some good kicking for thinking about it :p
[12:14] <seb128> you really suggest using an alpha patch against upstream for something like that?
[12:14] <pygi> no, I was just wondering, don't eat me pls :)
[12:14] <seb128> so no need to wonder
[12:15] <seb128> we are not going to use that patch before upstream
[12:19] <seb128> pygi: BTW I'm not sure it spares that much on memory usage, those numbers seems to be on loading, not on constant use 
[12:19] <pygi> seb128: indeed
[12:19] <mdke> iwj: around?
[12:28] <iwj> mdke: Y
[12:47] <simira> sladen: working on it, at least. Installing Dapper final first, I am already stuck on partman. So... Kamion, where do you want the bug reports?
[12:49] <ogra> mdz, mind to give an opinion on a ltsp change i plan ? 
[12:57] <Kamion> simira: debian-installer
[12:57] <Kamion> if you don't know more accurately
[12:57] <simira> Kamion: not directly on partman? currently it's the partitioning that goes wild.
[01:02] <Kamion> simira: no, there's no partman package at present
[01:03] <Kamion> (you might find one in LP, but it's obsolete)
[01:03] <Kamion> partman is made up of a number of components
[01:04] <Mithrandir> (that is, we're no longer frozen)
[01:05] <Mithrandir> just UVF.
[01:05] <ogra> ya
[01:05] <ogra> y
[01:05] <Mithrandir> so, go wild.
[01:08] <simira> Mithrandir: my installation already did
[01:08] <ogra> *goes too
[01:11] <rodarvus> Hooray \o/
[01:15] <dholbach> hey rodarvus
[01:15] <dholbach> ROCK ON :)
[01:15] <rodarvus> dholbach: hi there! :)
[01:27] <mdz> ogra: what's the change?
[01:27] <mdz> I'm leaving for lunch in about 2 minutes
[01:28] <ogra> mdz, i just wanted a second opinion on genererating the sshkeys for ldm through if-up.d
[01:29] <ogra> but rodarvus already gave his opinion :)
[01:29] <ogra> s/generating/copying/
[01:30] <ogra> mdz, enjoy your tapas :)
[01:49] <thom> pitti: how is cryptdisks-early supposed to work? currently it just bitches vigorously that it doesn't have the right devmapper, doesn't have the right cyphers, and then forces me to hit enter $check times to get to the real cryptdisks startup by which point everything works
[02:00] <Kamion> simira: bug 53511, is that the desktop CD installer or the text-mode installer?
[02:00] <Ubugtu> Malone bug 53511 in debian-installer "Partitions doesn't appear correctly in the "Prepare partitions" part (Dapper Final)" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/53511
[02:00] <pygi> sivang: poke poke
[02:01] <simira> Kamion: the cd, yes. I'll add that
[02:01] <Kamion> simira: which CD? :-)
[02:02] <Kamion> if you mean the desktop CD, sorry, I gave the wrong answer earlier - bugs on that installer go against ubiquity
[02:02] <simira> Kamion: ok, I guess I should have made that clear. I'll move the bug.
[02:03] <Kamion> simira: thanks
[02:11] <simira> Kamion: bugs for edgy coming up soon ;p
[02:14] <Kamion> simira: good :)
[02:18] <zul> hi
[02:28] <simira> what is this kind of madness???
[02:29] <simira> (booting edgy desktop cd)
[02:30] <Kamion> huh?
[02:31] <simira> Kamion: just the background pic on boot. A bit ... messy.
[02:32] <tseng> simira: the usplash test card?
[02:32] <simira> yup
[02:32] <tseng> multi colored boxes with gradings
[02:32] <tseng> yeah, thats for testing.
[02:41] <Hobbsee> is the freeze really over?
[02:43] <Kamion> Hobbsee: yes
[02:43] <Hobbsee> Kamion: yay!
[02:48] <doko> fabbione: is there some documentation about snakeoil certificates?
[02:48] <doko> dholbach: is there some documentation about dh_iconcache?
[02:49] <Hobbsee> doko: what did you want to know about dh_iconcache?
[02:50] <dholbach> doko: hm, with the comments in /usr/bin/dh_iconcache, i thought that it'd produce a manpage, somehow   man dh_iconcache  does not work, hrm
[02:51] <Hobbsee> doko: i think there was originally a page on "this is why we're making this dh_iconcache change", if you were looking for that
[02:52] <dholbach> doko: http://wiki.ubuntu.com/DhIconCacheChanges
[02:54] <seb128> doko: what do you need to know about it?
[02:55] <doko> dholbach, seb128: I was looking for something that I could use for https://wiki.ubuntu.com/UbuntuPackagingChanges
[02:55] <dholbach> doko: ah ok
[02:56] <Hobbsee> oh yeah, guess new packages still need that, dont they...
[02:56] <seb128> doko: what is the wiki page about?
[02:56] <seb128> doko: explaining changes we have over Debian or something else?
[02:56] <doko> seb128: https://wiki.ubuntu.com/UbuntuPackagingChangesSpec
[02:57] <seb128> doko: basically dh_iconcache is "update GTK icon cache" and is useful for "any package shipping an icon to /usr/share/icons/some_theme"
[02:57] <Hobbsee> doko: shouldnt be hard.  rationale is that mailing list thing, affected packages are anything that installs files into those two directories
[02:58] <doko> seb128: in /usr/share/icons/some_theme or /usr/share/icons/ ?
[02:58] <dholbach> doko: they all ship to <somet_theme>
[02:58] <seb128> doko: is there anything dropping an icon directly to /usr/share/icons? that would be a bug
[02:59] <seb128> doko: the icons are meant to be part of a theme, which is a why the some_theme directory
[02:59] <dholbach> and the some_theme/index.theme gives instructions on how the theme "works"
[03:00] <seb128> dholbach: that's not revelant for dh_iconcache
[03:00] <dholbach> without that dh_iconcache skips the directory
[03:00] <dholbach> OOo ships to locolor which has no index.theme
[03:00] <dholbach> which is ... a bit weird :)
[03:00] <seb128> OOo is bugged 
[03:00] <seb128> icons are not used in that case
[03:00] <seb128> nothing to do with the cache 
[03:00] <dholbach> yeah
[03:01] <seb128> keep it easy
 doko: basically dh_iconcache is "update GTK icon cache" and is useful for "any package shipping an icon to /usr/share/icons/some_theme"
[03:01] <seb128> that is good enough as a summary probably
[03:01] <doko> dholbach: can you fix  the missing index.theme?
[03:01] <seb128> dholbach maintains OO.o :)
[03:02] <dholbach> doko: i'd suggest to use the hicolor index.theme
[03:02] <dholbach> forget it
[03:02] <dholbach> i'm off - bye
[03:02] <dholbach> :-p
[03:02] <seb128> dholbach: come on, it's a funny package I'm sure :p
[03:02] <seb128> doko: nice try, didn't work apparently though :p
[03:02] <dholbach> seb128: you had the sun shining on your head for too long, if you ask me
[03:02] <seb128> dholbach is not that easy to trick into OOo maintainship :p
[03:03] <dholbach> seb128: what do you think why doko makes me drink cocktails all the time... apparently i never got drunk enough to take over :)
[03:03] <seb128> dholbach: possible, it's too hot here :)
[03:03] <seb128> hehe
[03:03] <Kamion> dholbach: is there any chance dh_iconcache will ever go back to Debian?
[03:03] <seb128> there is a bug to the BTS about it
[03:04] <Kamion> ah good, I think last time I looked there wasn't
[03:04] <seb128> but Debian guys don't like that all packages have to be transitioned together ...
[03:04] <seb128> so there is some open discussion
[03:04] <dholbach> Kamion: what seb128 said. the gtk-gnome team has a proposal but didn't take it forward yet
[03:04] <dholbach> they need to take the transition slower than we did
[03:05] <Hobbsee> dholbach: i hate the idea of updating all their packages like we did.  assuming they have way more packages.  ouch.
[03:05] <maswan> Mithrandir: sync finally done, don't know why it took so long, but for some reason it didn't want to pull more than 1-2M/s
[03:06] <seb128> Hobbsee: they don't
[03:06] <dholbach> Hobbsee: they don't have more packages, it's just there are more maintainers involved, migration to testing, etc
[03:06] <seb128> Hobbsee: you know, we do sync on Debian, so we have the package they have
[03:06] <seb128> packages they have
[03:06] <Hobbsee> seb128: yeah, i do know that.  for some reason, my brain was expecting that there was a section of debian we didnt sync from.
[03:07] <seb128> Hobbsee: the issue is that it requires the different maintainers to upload their packages, you can get dholbach uploading everything that needs to be transitioned :p
[03:07] <Hobbsee> more architectures to build on too, although i guess they have more build machine
[03:07] <Hobbsee> seb128: yeah, true.  :P  or me too
[03:07] <Hobbsee> *to
[03:07] <dholbach> seb128: is still angry that i touched kde packages
[03:07] <dholbach> he doesn't get over it easily
[03:07] <Hobbsee> haha
[03:07] <dholbach> ;-)
[03:07] <dholbach> Hobbsee: and it felt good, didn't it?
[03:08] <ogra> haha
[03:08] <Hobbsee> dholbach: no, it felt horrible, and it had so many icky dependancies to install.
[03:08] <Hobbsee> hi ogra 
[03:08] <ogra> hey
[03:08] <dholbach> haha
[03:08] <dholbach> see you later
[03:08] <Hobbsee> bye dholbach 
[03:08] <seb128> dholbach: have fun
[03:08] <ogra> enjoy the sun
[03:10] <Mithrandir> maswan: thanks
[03:10] <Hobbsee> hi Mithrandir!
[03:10] <Hobbsee> whiprush: you may well be right, w.r.t. laptop power
[03:11] <Keybuk> meh, do I upgrade my desktop to edgy or leave it on dapper?
[03:11] <Hobbsee> Keybuk: upgrade to edgy.  breakage is fun.
[03:11] <Keybuk> Hobbsee: heh, breakage of one's primary machine isn't
[03:11] <Hobbsee> Keybuk: sure it is.
[03:14] <Kamion> Hobbsee: we don't autosync from contrib/non-free, only on request
[03:14] <Kamion> (at present)
[03:14] <Kamion> but there's really not a huge amount in there
[03:14] <Hobbsee> heh.  so i was right :P  true
[03:14] <Hobbsee> Kamion: did you see/process my sync?
[03:14] <Kamion> I saw it, haven't done a sync run since though
[03:14] <Hobbsee> Kamion: cool, okay.
[03:54] <simira> and live in from Edgy Knot 1 we've got Simira!!!
[03:54] <Hobbsee> simira: yay!
[03:55] <simira> *coughs*
[03:55] <Hobbsee> heh :P
[04:10] <Keybuk> mdz: you so should be in the UK this week
[04:10] <Keybuk> it'd cure ALL your bitching about how the UK weather is never hot enough
[04:11] <mdz> Keybuk: I will be there next week
[04:11] <thom> it'll probably rain all next week
[04:11] <Keybuk> oh, any special occasion?
[04:11] <mdz> and for the record, I don't complain so much about the cold as about the rain
[04:11] <mdz> Keybuk: no, just stopping in for a few days on my way home
[04:11] <Hobbsee> hi bddebian 
[04:11] <thom> Hobbsee: aussies have no understanding of what weather actually is
[04:11] <Hobbsee> thom: :P
[04:11] <Keybuk> mdz: ah, how has spain been?
[04:11] <mdz> hot
[04:11] <Mithrandir> Hobbsee: it's just +25C?
[04:12] <Hobbsee> Mithrandir: more like 10C outside, more like 20C in here
[04:12] <thom> like i said...
[04:12] <ogra> mdz, i thought you complain about the clouds 
[04:12] <Keybuk> man, LWN is heavy going today ... it's huge with all these kernel summit summaries
[04:12] <bddebian> Heya folks
[04:12] <bddebian> Hi Hobbsee
[04:12] <mdz> ogra: that too
[04:13] <mdz> ok, back to the mines
[04:16] <Mithrandir> Kamion: hmm, Simira just did an install here and seems to be bitten by the apt-setup bug.
[04:17] <simira> (it didn't hurt very much, just a small itch)
[04:18] <simira> should I report/comment it somewhere?
[04:20] <Kamion> simira: details?
[04:20] <Kamion> simira: anyway, yes please, bug on apt-setup with /var/log/installer/syslog attached
[04:22] <Mithrandir> Kamion: just security repos enabled in sources.list, nothing else.
[04:22] <Kamion> ok
[04:23] <Kamion> oh also /var/log/syslog (before reboot) please if this was a ubiquity install
[04:23] <Kamion> unless you've already rebooted in which case I guess I might lose, but we'll see
[04:23] <Kamion> need to get round to unifying those
[04:25] <simira> Kamion: sorry, I already bootet. I seldom check sources.list before the boot ;)
[04:26] <Kamion> well, do the best you can log-wise then
[04:30] <simira> Kamion: it won't add empty files (syslog)
[04:30] <Kamion> check whether it's really empty or just only readable by root
[04:31] <simira> weird
[04:33] <fabbione> doko: what kind of documentation are you looking for'+
[04:33] <fabbione> ?
[04:33] <simira> Kamion: do you want the version file also?
[04:34] <Kamion> simira: no
[04:36] <doko> fabbione: just what consists of a patch for a snakeoil-cert change
[04:36] <fabbione> doko: you mean how a package need to be changed to use the snakeoil-cert?
[04:40] <Gloubiboulga> Mithrandir, hello, is there a problem with the xubuntu Desktop CDs? 
[04:40] <Kamion> Gloubiboulga: very oversized
[04:40] <Gloubiboulga> Kamion, oh, ok
[04:41] <Kamion> live fs build logs are here if you want to investigate: http://people.ubuntu.com/~cjwatson/livefs-build-logs/edgy/xubuntu/current/
[04:41] <Gloubiboulga> thanks
[04:41] <doko> fabbione: yes
[04:43] <ogra> Gloubiboulga, http://cdimage.ubuntu.com/xubuntu/daily/current/report.html is a rough indicator for oversizedness
[04:44] <Kamion> daily-live not daily
[04:44] <Kamion> (there is no report.html for daily-live)
[04:44] <ogra> oh, right 
[04:44] <ogra> but .OVERSIZED files (sometimes :) )
[04:44] <fabbione> doko: there isn't much of an howto. snakeoil gives you a (fake) cert in standard path. Your package can Depends: and use that certificate as it please (modulo remove it).
[04:45] <Kamion> since the alternate CD is apparently not oversized, its report.html isn't too useful in this case
[04:45] <ogra> yup
[04:46] <fabbione> doko: you want to notice that the secret part of the cert is also special group readable. so if your daemon or wahtever does not run as root, can still gain read access if you in postinst make that daemon part of the group
[04:46] <fabbione> doko: you can see how pitti did pqsql
[04:46] <Gloubiboulga> I guess we'll have to remove some language packs
[04:47] <rodarvus> guys, just for a quick update on X.Org -> UVF excpeption has just been authorised by mdz, x11proto-* are on their way to the archive right now. As usual, further updates on this topic will occur on #ubuntu-x
[04:47] <Kamion> jesus, *how* many language-support-*?
[04:48] <Kamion> Gloubiboulga: I'm sure I've commented before on the unwisdom of having lots of language-support-* on the CD
[04:48] <Kamion> the other CDs just have language-support-en
[04:48] <rodarvus> I apologize in advance for any brokeness that might happen during the process :D
[04:48] <fabbione> rodarvus: you are a good guy with good hopes.. 
[04:48] <doko> fabbione: ok, I that should suffice
[04:48] <fabbione> rodarvus: people will still hunt you down :PO
[04:49] <ogra> rodarvus, yippie !
[04:49] <Gloubiboulga> Kamion, Jani added language-packs for dapper because we had some free space, he hasn't changed this for Edgy I guess
[04:49] <ogra> we know where to send angry users, dont worry :)
[04:49] <rodarvus> :)
[04:50] <Kamion> Gloubiboulga: nod
[04:50] <jsgotangco> expect a death threat or two lol
[04:56] <simira> Kamion: everything seems to work too well, though, so don't you guys break too much!
[04:56] <ogra> simira, we havent even started yet :)
[04:57] <ogra> feature development *ends* on sept 7, expect some funny weeks ahead :)
[04:57] <simira> ogra: *sigh* It works NOW
[04:57] <Hobbsee> heh
[04:57] <pitti> ogra: actually, I think the debian sync rave should have accounted for the majority of breakage (package-wise, at least)
[04:58] <ogra> pitti, do you have any features that involve packaging ? 
[04:58] <ogra> i dont :) 
[04:58] <pitti> ogra: I have some that involve all packages, okay :)
[04:58] <ogra> i'll just break code :)
[04:58] <pitti> ogra: but packag*ing*, only two new packages
[04:59] <pitti> ogra: so I'm better off, I add new broken code :)
[04:59] <ogra> hehe
[05:03] <Kamion> infinity,BenC,doko,sfllaw: ping, #ubuntu-meeting
[05:04] <Kamion> missing heno
[05:13] <doko> Kamion: already there
[05:14] <Kamion> doko: ok, thanks, I was a bit lagged working out who was/wasn't there
[05:14] <Kamion> and irssi's tab completion was confusing me
[05:22] <seb128> pitti, ogra: better to discuss here or after the meeting
[05:22] <pitti> seb128: ack
[05:23] <seb128> vuntz: around?
[05:23] <ogra> yup
[05:23] <vuntz> seb128: yes
[05:25] <seb128> vuntz: we are annoyed by gnome-power-manager 2.15, it apparently requires hal policykit which a CVS feature still being worked and discussed ... do you know what if that's a concern for GNOME too?
[05:25] <seb128> vuntz: like 2.16 will not depends on hal CVS, will it?
[05:25] <vuntz> seb128: we don't know that
[05:25] <vuntz> seb128: can you send a mail to release-team?
[05:26] <vuntz> we have a meeting tomorrow to discuss this
[05:26] <vuntz> s/this/modules inclusion/
[05:26] <seb128> ok, will do, thank you
[05:26] <seb128> I'm not subscribed to the list, will somebody moderates it? :p
[05:26] <seb128> or is the list moderated every n weeks and I should ping somebody about it after sending? :)
[05:27] <vuntz> seb128: it should work
[05:27] <vuntz> if you need moderation, ping me :-)
[05:27] <seb128> ok
[05:28] <bddebian> WTF should the GLwM library be?
[05:28] <Kamion> something we removed
[05:29] <Kamion> if it's anything to do with GLw
[05:29] <Kamion> which is mesa's widget library thing that needed lesstif in main
[05:29] <bddebian> Ahh, so we don't have it?
[05:30] <vuntz> seb128: btw, did you discuss this with the g-p-m maintainer?
[05:30] <seb128> vuntz: like upstream maintainer or ubuntu one?
[05:31] <bddebian> Kamion: libgl1-mesa-swx11-dev seems to provide the headers.  Are you saying we aren't providing the libraries themselves?
[05:31] <seb128> vuntz: ogra pointed he's blocked to package g-p-m 2.15 due to that
[05:31] <seb128> vuntz: he's going to mail upstream about it but has not done yet
[05:31] <vuntz> upstream
[05:31] <vuntz> cool
[05:31] <fabbione> bddebian: we don't provide the libraries no
[05:31] <ogra> vuntz, hughsie just moved houses and was only online via gprs ...
[05:31] <Kamion> bddebian: right, we killed GLw
[05:31] <vuntz> ogra: can you cc the release-team? :-)
[05:31] <ogra> vuntz, will do
[05:31] <bddebian> Eeks
[05:31] <Kamion> it had like four rdepends, several of which were removable anyway
[05:31] <seb128> ogra: today? :)
[05:32] <ogra> seb128, yes
[05:32] <seb128> ok, thank you
[05:32] <bddebian> Kamion or fabbione: So any idea what I should do about grass?
[05:32] <fabbione> bddebian: what's grass?
[05:32] <ogra> bddebian, whats wrong ? 
[05:32] <bddebian> Besides smoke it ;-P
[05:32] <Mithrandir> bddebian: smoke it! :-)
[05:32] <dholbach> haha
[05:32] <fabbione> (other than the green thing you smoke)
[05:32] <bddebian> Mithrandir: ;-P
[05:32] <Mithrandir> jinx
[05:32] <ogra> fabbione, GIS system 
[05:33] <ogra> editing/creating maps etc
[05:33] <bddebian> ogra: It build-deps libgl1-mesa-swx11-dev and looks for Glw libs
[05:33] <fabbione> ogra: ah ok
[05:33] <bddebian> http://pastebin.us/1702
[05:33] <Mithrandir> bddebian: port it to SDL or write an API-compatible replacement for libGLw which uses SDL or GTK or something.
[05:33] <ogra> bddebian, fix the build dep ? 
[05:34] <fabbione> ask for green removal from archive
[05:40] <seb128> vuntz: looks like g-p-m has a --disable-policykit configure option, sorry for the noise, I'll look next time before pinging you about what orgra said
[05:40] <seb128> pitti, ogra: g-p-m has a --disable-policykit
[05:41] <seb128> fedora is packaging it using that flag
[05:41] <seb128> looks like it's no issue
[05:41] <ogra> seb128, yes, ten it doesnt suspend/hibernate anymore :P
[05:41] <pitti> seb128: \o/
[05:41] <ogra> i already tried it :)
[05:41] <pitti> ogra: teaching it to use pmi is too hard?
[05:41] <seb128> ogra: are you sure?
[05:41] <ogra> yes i am
[05:41] <tseng> ogra: it doesnt support if you build with with policykit and pk isnt there, too
[05:41] <tseng> support suspend
[05:41] <ogra> (i spent two days with this package ...)
[05:42] <seb128> ogra: I'm surprised that fedora guys just dropped suspend and hibernate like that
[05:42] <ogra> seb128, they run suid scripts 
[05:43] <ogra> they can just use the gconf key and set it to /sbin/hibernate
[05:43] <pitti> ogra: eek
[05:43] <pitti> ogra: why the heck don't they use hal 0.5.7's power backends?
[05:43] <pitti> hi mdz
[05:43] <seb128> ogra: ok, better to mail GNOME list and upstream then 
[05:43] <ogra> pitti, i must admit i havent looked into their 2.15 package yet ... its what they did in 2.14
[05:44] <ogra> seb128, yep
[05:44] <pitti> ogra: right, for this purpose we needed hal 0.5.7 in dapper, AFAIR
[05:44] <pitti> ogra: that worked reasonably well, I think
[05:44] <ogra> hughsie should also be available on irc again  ... i'll see if i can catch him after meeting
[05:44] <pitti> ogra: (hey, don't get me wrong, I'm not blaming you :) )
[05:51] <pitti> iwj: the changelog doesn't really look promising, though
[05:51] <iwj> I didn't look at it ...
[05:52] <iwj> You mean the `patches to fix regressions' ?
[05:52] <pitti> iwj: I just looked at the .changes (a signed one, yay... :) )
[05:52] <iwj> The latest in ftp.debian.org is sarge5 so there were quite a few attempts and you only see the last one there (lack of -v).
[05:53] <pitti> iwj: oh, you mean sarge8 had the actual fixes, and sarge9 only some fixes of the fixes?
[05:53] <iwj> I think so (that's what I guess - I didn't look at sarge8).
[05:53] <iwj> Presumably sarge6 were the fixes and sarge7 the fixes to the fixes, which makes sarge9 the fixes to the fixes to the fixes to the fixes or something.
[05:54] <pitti> iwj: ah, I just looked at the sarge8 changelog, that has quite a bunch of fixes
[05:54] <pitti> iwj: but sarge8 changelog is from mid-june
[05:54] <iwj> Yes, it's taken them a while.
[05:55] <pitti> iwj: erm, wait, I'm confused. we already have 1.0.8
[05:55] <iwj> I mean 1.0.9.
[05:55] <pitti> iwj: it's a common 1.0.9 tarball that we are looking for
[05:56] <pitti> iwj: yep, ok, so we want the 1.0.4-2sarge8 + 9 changes
[06:06] <fabbione> BenC, ogra: the setup is simple
[06:06] <fabbione> you need machine A to export a block device via AOE for example
[06:06] <fabbione> (doesn't need to be x86)
[06:06] <fabbione> then you need machine B and C that will play cluster
[06:06] <fabbione> the cluster setup is easy
[06:07] <fabbione> you import the device from A to B and C
[06:07] <fabbione> format with GFS and/or GFS2
[06:07] <fabbione> start the cluster in minimal setup
[06:07] <fabbione> mount GFS/GFS2 read/write to it
[06:07] <fabbione> no need to play dbench with a million instances
[06:07] <fabbione> right now the bug we need to hunt down doesn't allow you to get that far
[06:08] <fabbione> BenC: unfortunatly the cluster stuff still doesn't run properly on sparc. i am working with upstream to get that fixed
[06:08] <pitti> dholbach: I'll make some time for the motu school
[06:08] <fabbione> so who wants to take it?
[06:09] <dholbach> pitti: you rock - i think i'll do the "package update" session soon, now that I watched seb128 in awe for a long time ;)
[06:09] <BenC> fabbione: I have a xeon/k8/ia64/hppa "cluster"
[06:09] <fabbione> xeon/k8/ia64/ppc should be good
[06:09] <fabbione> you can use ppc to export the AOE device
[06:10] <ogra> fabbione, i only have a bunch of laptops, but after my move is done i'll hapily test for you
[06:10] <fabbione> it can just be a 2GB file loopback mounted and exported with vblade
[06:10] <fabbione> that i know it works
[06:10] <fabbione> BenC: xeon-k8 cluster is perfectly fine
[06:10] <Petaris> ajmitch: ping
[06:11] <Petaris> ajmitch: Are you still working on AD integration?
[06:11] <fabbione> BenC: does it sound to scary?
[06:11] <Hobbsee> Petaris: not a chance you'll catch him now
[06:11] <seb128> pitti: how do you manage to crack on so many things in the same time :)
[06:12] <BenC> fabbione: Scary yes...too scary, never :)
[06:12] <Petaris> Hobbsee: Ahh, ok
[06:12] <fabbione> BenC: hehe ok, should we arrange a session tomorrow around 2am UTC so i can help you setup the environment?
[06:12] <pitti> dholbach: oh, it's *today*? argh, I have a concert this evening
[06:12] <Hobbsee> Petaris: 4am there
[06:12] <dholbach> pitti: today?
[06:12] <Petaris> When is he usually around?
[06:12] <fabbione> BenC: it won't take too long
[06:12] <Petaris> ahh
[06:12] <dholbach> pitti: it's whenever you have time and we can announce it
[06:12] <BenC> fabbione: Yeah, I can do that
[06:12] <fabbione> BenC: ok sounds good.
[06:13] <dholbach> pitti: and it just needs to be a 20 min session
[06:13] <BenC> fabbione: that's 32 hours from now, right?
[06:13] <seb128> pitti: thank you :p
[06:13] <dholbach> pitti: with questions and examples afterwards... or something
[06:13] <BenC> well, 34 hours from now, or do you mean 10 hours?
[06:13] <fabbione> BenC: 10 hours.... i think
[06:13] <fabbione> 10 hours from now
[06:14] <fabbione> 34 hours would be saturday for me
[06:14] <pitti> dholbach: that sounds manageable
[06:14] <fabbione> and that's like asking for my wife to kick me out of the door :)
[06:14] <pitti> dholbach: as I said, today is baad, tomorrow or Monday is better
[06:14] <dholbach> pitti: sure, it's good if we can announce it in advance
[06:15] <pitti> dholbach: anything particular I should talk about?
[06:15] <dholbach> pitti: the good thing is: somebody will make wiki pages and documentation from it, so it really *should* make a change
[06:15] <dholbach> pitti: whatever you like... but you can take a look at http://wiki.ubuntu.com/MOTU/School/Requests
[06:15] <BenC> fabbione: ok, so tonight for me :)
[06:15] <fabbione> BenC: yes i guess so
[06:16] <fabbione> BenC: but i just read on -meeting you are going vac.. how many weeked?
[06:16] <fabbione> weeks
[06:16] <pitti> dholbach: oh 'how to patch sources' would suit me fine
[06:16] <pygi> sivang: poke poke?
[06:16] <pitti> dholbach: after getting to know all the different weird patch systems out there while doing security updates, I feel I have some experience in that :)
[06:16] <BenC> fabbione: all of next week
[06:16] <dholbach> hehe :)
[06:17] <dholbach> super, pitti
[06:17] <fabbione> BenC: ok.
[06:17] <fabbione> BenC: have fun :)
[06:17] <BenC> "and make money" :)
[06:17] <BenC> the two sort of go together though
[06:17] <simira> fabbione: how's Ulla? Did she deliver yet?
[06:17] <zul> oh yes...poker tournament
[06:17] <fabbione> simira: nope...
[06:17] <fabbione> BenC: ehehhehe
[06:18] <fabbione> BenC: money->females.fun();
[06:18] <pitti> dholbach: do these dates mean the date of the request or the date of the session? :)
[06:18] <dholbach> pitti: request ;)
[06:18] <fabbione> simira: she is big like a UFO now...
[06:18] <dholbach> pitti: i added suggestions after i wrote the spec
[06:19] <ogra> fabbione, she has little blinking lights ? o_O
[06:19] <Hobbsee> ogra: yes, for mind control.
[06:19] <fabbione> ogra: i did place them around the baloon to make sure she is 100% visible at night
[06:19] <pitti> dholbach: alright, sign me up for the patch session
[06:19] <simira> fabbione: she's a week overdue or something?
[06:19] <ogra> LOL
[06:20] <ogra> simira, two or three rather ? 
[06:20] <pitti> dholbach: shall I just pick a date and announce it to the -motu list? or do you already have a process for that?
[06:20] <fabbione> simira: yeah...
[06:20] <dholbach> pitti: just add it to the wiki and write to the list that'd be great
[06:23] <Hobbsee> pitti: :)...but i'm passed out, so...
[06:23] <Hobbsee> oh grr. make me even colder then.
[06:23] <Hobbsee> and soaking wet.
[06:24] <pitti> Hobbsee: colder? ah, sorry, I forgot - take this nice herb tea instead
[06:24] <Hobbsee> heh
[06:24] <Hobbsee> fabbione: you want to keep those, surely.
[06:24] <fabbione> Hobbsee: with this heat? yes
[06:25] <Hobbsee> fabbione: heh.  heat. how hot's it for you at the moment?
[06:25] <fabbione> around 30
[06:25] <fabbione> C
[06:25] <pitti> Hobbsee: 38 degrees Celsius here...
[06:25] <Hobbsee> pitti: bah.  nice and warm :)
[06:26] <pitti> Hobbsee: (in the shadow, I don't want to know how hot it is in the sun)
[06:26] <Hobbsee> pitti: yeah, okay...if that's in the shaddow, then i'm a bit more compassionate
[06:26] <Keybuk> Mithrandir: my problem with the 24 hour thing is the 0800 meeting ... I 
[06:26] <Keybuk> I'd have to have the text in the wiki by 0800 on Wednesday
[06:26] <Hobbsee> 30C should be nice, fabbione, although not for your wife, i expect.  no airconditioning?
[06:26] <Keybuk> which means that I really need to write it on Tuesday
[06:26] <Keybuk> so, effectively, we end up having a "what we did LAST WEEK" meeting, over half way into the next week
[06:27] <Mithrandir> Keybuk: make 0800 special, then
[06:27] <Keybuk> given the meeting is supposed to be about resolving blockers, and getting help, I don't see the point in that
[06:27] <fabbione> Hobbsee: no a/c but 30 is not nice when you add on top the temperature of a rack heat :)
[06:27] <Keybuk> if the meeting were just as a progress report, we could just do the wiki and not meet at all
[06:27] <fabbione> anywa
[06:28] <Hobbsee> fabbione: well...what can i say... :P
[06:28] <fabbione> does anybody need anything from me? otherwise i am off for BBQ
[06:28] <Keybuk> the fact it's supposed to be a "I'm blocked here, I need help here, etc." type of meeting means that the information we're discussing needs to be current
[06:28] <Hobbsee> fabbione: stick it in the cellar or something?
[06:28] <fabbione> Hobbsee: ehehe
[06:29] <Hobbsee> fabbione: no, wait, that's where all the bodies are, that wont work.
[06:29] <fabbione> Hobbsee: that too :)
[06:29] <Hobbsee> :P
[06:30] <Hobbsee> (sway sway thud bang ouch sway)...hmmm...i dont think i should handle a large knife tonight.
[06:30] <pitti> dholbach: ok, I scheduled the lesson for next Tuesday 1600 UTC, that should suit both the Americans and the Europeans
[06:31] <Hobbsee> pitti: what time is that, euro time?
[06:31] <Hobbsee> oh hang on, dont worry.
[06:31] <Hobbsee> i though that was a sane meeting time, for a second
[06:31] <pitti> Hobbsee: UTC
[06:31] <Hobbsee> pitti: oh bleh.  so it is, london's only 1 hour ahead.
[06:32] <Hobbsee> pitti: dholbach you'll log that, i take it?
[06:32] <dholbach> Hobbsee: me? we had so many ubuntu motu doc team people around, didnt we? :)
[06:32] <Hobbsee> dholbach: well, whoever.
[06:33] <ogra> Hobbsee, sudo apt-get install gworldclock :)
[06:33] <Hobbsee> ogra: meet kclock, installed by default.
[06:33] <pygi> dholbach: you have a sec for me? :)
[06:33] <dholbach> pygi: yes
[06:33] <ogra> Hobbsee, pfft ... by default ... pfft 
[06:34] <Hobbsee> ogra: unforunatly, i cant apt-get reinstall brain.
[06:34] <pygi> dholbach: how's your student coming along? any progress?
[06:34] <Hobbsee> whichis the problem
[06:34] <ogra> heh ... 
[06:34] <sladen> Hobbsee: excessive bling and bloat I'll tell yer
[06:34] <ogra> i'd prefer a apt-get fridge brain
[06:34] <Hobbsee> sladen: :P
[06:35] <dholbach> pygi: i pinged him for 2-3 days now and i didn't hear back - i'll write him a mail later and see what's goingon
[06:35] <dholbach> pygi: i hope he's alright and all
[06:35] <pygi> dholbach: I mailed him
[06:35] <dholbach> pygi: ahhh good
[06:35] <pygi> It's end of his semester now, and he said he'll work harder now to catch up on things
[06:35] <dholbach> yeah, exactly
[06:36] <pygi> nice, just wanted to inform you :)
[06:36] <ogra> dholbach, whats he doing ? was that the epiphany stuff ? 
[06:36] <dholbach> ogra: yup
[06:36] <pygi> ogra, aha
[06:36] <ogra> ah
[06:36] <pitti> Hobbsee: I'm fine with another time, but I have observed that European evenings/American mornings work well so far
[06:36] <Hobbsee> pitti: indeed :)
[06:36] <dholbach> 35,0 C
[06:36] <pitti> Hobbsee: and I don't want a huge discussion about the time :)
[06:36] <ogra> i'm lagging a bit with Amaranth's willowng package, but it should hit universe the next days
[06:37] <Hobbsee> pitti: had no intention of having one.  for a minute there, i thought you'd found a good time on *all* timezones.
[06:37] <ogra> dholbach, lucky you, my thremometer hangs in the bright sun ... i cant even guess the temp :)
[06:37] <pygi> ogra, I could probably also package the Olive backend, but I think I'll wait for UI to be done
[06:38] <Amaranth> ogra: oh, did i ever actually give you the package i made?
[06:38] <Amaranth> ogra: right as i finished you disappeared
[06:38] <ogra> Amaranth, i bzr puled the changes :)
[06:38] <ogra> *pulled
[06:38] <Amaranth> ah
[06:38] <Amaranth> i wonder if i pushed the latest stuff :P
[06:39] <ogra> i made some comments ... but you were afk or something (and actually i forgot what comments that were, need to look again)
[06:39] <Hobbsee> night all
[06:39] <ogra> night Hobbsee 
[06:39] <Amaranth> whoops
[06:39] <Amaranth> looks like the latest stuff wasn't pushed
[06:39] <ogra> Amaranth, great, likely my comments are adressed then ;)
[06:39] <ogra> i
[06:40] <ogra> 'll look later 
[06:41] <doko> getting some ice cream, floating away ...
[06:43] <Amaranth> it looks like my init script doesn't actually work :P
[07:07] <LaserJock> Kamion (or others): is the devlopment team meeting wiki publicly available somewhere?
[07:08] <Kamion> LaserJock: https://wiki.ubuntu.com/DistroTeamMeeting20060720
[07:08] <LaserJock> Kamion: oh, thanks
[07:11] <zul> Kamion: ping...
[07:11] <rodarvus> Kamion: have a good night!
[07:12] <zul> Kamion: btw...the xen kernel boots now
[07:14] <ogra> zul, awesme+1
[07:22] <Tonio_> hi
[07:32] <Gloubiboulga> Is it possible to run a new Xubuntu Desktop iso build with updated seeds?
[07:33] <fschoep> dholbach: ping
[07:33] <bddebian> Oh shite, qgis is broken because of grass :-(
[07:33] <dholbach> fschoep: pong
[07:33] <dholbach> fschoep: where do we meet?
[07:34] <fschoep> PM?
[07:34] <dholbach> ok
[07:35] <ogra> Gloubiboulga, i could trigger an iso build, but you will need someone with access to the livefs build system before else i can only build from an old livefs which gains you nothing
[07:36] <Gloubiboulga> ogra, thanks, I don't actually know the processes for all these build things. Do we have a doc somewhere?
[07:37] <sharms> mako: ping
[07:38] <ogra> Gloubiboulga, i dont think so ...
[07:39] <ogra> Gloubiboulga, the process is: pester Kamion, infinity or Mithrandir to trigger a livefs build ... if thats done, poke either of these three, Riddell or me to make an iso from that 
[07:39] <ogra> for install CDs the livefs step isnt needed indeed
[07:41] <Gloubiboulga> ogra, ok, I'll bug everyone tomorrow ;)
[07:42] <ogra> :)
[07:52] <bddebian> doko: ping?
[08:05] <sharms> My laptop, running edgy, had a reiserfs partition on it that apparently only has 1 bad sector and now reiserfs is reporting it is a hardware issue.  I am now reinstalling dapper as a ext3 system, but I have the feeling it is not a hardware issue but a reiserfs issue.  I don't have any logs or anything, although I didn't shutdown the system properly, but should I be thinking about filing a bug report or write it off to freak hap
[08:05] <sharms> pening?
[08:06] <sharms> That bad sector was apparently in /boot which makes it more odd that it would just be in my boot loader
[08:06] <dholbach> can somebody promote cairomm to main?
[08:07] <Petaris> sharms: reiserfs is sort of know for not being overly reliable
[08:08] <sharms> Ah see I always got the impression it was better, but just like the bastard of linux
[08:08] <Petaris> Don't get me wrong its a great fs if you need the speed and work with small files but there is always a payoff
[08:08] <Petaris> s/payoff/tradeoff
[08:13] <dholbach> Kamion, Keybuk: can somebody of your promote cairomm to main? :)
[08:14] <Keybuk> dholbach: it doesn't appear to be depended on by anything or seeded?
[08:14] <dholbach> Keybuk: gtkmm2.4 will need it
[08:15] <dholbach> gtkmm2.4 upload first?
[08:15] <Keybuk> ok
[08:19] <dholbach> Keybuk: shall i do the gtkmm2.4 upload now?
[08:20] <Keybuk> if you like
[08:21] <dholbach> yeah, very much so :)
[08:26] <sharms> so I run /sbin/badblocks on the said drive with reisferfs, and the bad block is near the grub location.  I format, put ext3 on, and run /sbin/badblocks and I have no bad blocks
[08:26] <sharms> but I really have no diagnostics to report a bug with
[08:31] <Petaris> sharms: I would guess it was a reiser issue then
[08:32] <dholbach> Keybuk: thanks a lot
[08:38] <Gloubiboulga> No desktop cd for xubuntu but at least the alternate iso works just fine :)
[08:45] <Burgwork> jdub, take a look at the front page of eweek --> http://eweek.com/
[08:46] <sladen> Burgwork: excellent find!
[09:12] <rodarvus> Keybuk: ping
[09:14] <Keybuk> rodarvus: yup?
[09:15] <rodarvus> may I ask you to sync three rather urgent X libraries (their sync was requested a few minutes ago)
[09:15] <Keybuk> sure
[09:15] <rodarvus> thanks :)
[09:15] <Keybuk> xtrans?
[09:15] <rodarvus> no, xtrans is built already
[09:15] <Keybuk> hmm
[09:15] <Keybuk> which three?
[09:15] <rodarvus> just a sec
[09:15] <Keybuk> you're faster than Malone! :p
[09:15] <rodarvus> libxau, libxdmcp and libfontenc
[09:16] <rodarvus> the sync requests were made on Malone (ubuntu-archive subscribed to them, too)
[09:17] <Keybuk> right, but Malone hasn't actually mailed them out yet <g>
[09:17] <rodarvus> oh
[09:17] <rodarvus> haha
[09:17] <Keybuk> meh, you've changed your lp id!
[09:17] <rodarvus> I apologize for asking you here - its just that it would be great to have the base of X already published by the end of the day today :)
[09:18] <rodarvus> Keybuk: right, is rodarvus for a few weeks, now
[09:18] <Keybuk> no problem
[09:18] <Keybuk> it isn't a few weeks! :p  it's a week at most
[09:18] <rodarvus> internet time ;)
[09:19] <Keybuk> uhh, are you sure these are updated in Debian?
[09:21] <rodarvus> Keybuk: debian experimental
[09:21] <Keybuk> heh
[09:21] <rodarvus> :D
[09:22] <pygi> sivang: poke? :)
[09:23] <Keybuk> [NOT Updating - Modified]  libfontenc_1:1.0.1-6ubuntu2 (vs 1:1.0.2-1)
[09:23] <Keybuk> [NOT Updating - Modified]  libxau_1:1.0.0-3ubuntu1 (vs 1:1.0.1-1)
[09:23] <Keybuk> ok to override?
[09:24] <rodarvus> sure
[09:24] <rodarvus> (its marked as 'ok to override' on the sync requests, for reference)
[09:24] <Keybuk> they still haven't turned up yet <g>
[09:25] <rodarvus> hahaha
[09:25] <Keybuk> done
[09:26] <rodarvus> thanks a lot!
[09:26] <Keybuk> can you close the bugs?
[09:26] <rodarvus> sure
[09:33] <ogra> rodarvus, wow, my ltsp change today gives me a lot of feedback ...
[09:33] <rodarvus> Keybuk: bugs closed - thanks again!
[09:33] <rodarvus> ogra: oh, good feedback? :)
[09:33] <ogra> well ... 
[09:34] <ogra> i guess once the users use it i will get good feedback from less support questions ...
[09:34] <ogra> but debain seems not happy ...
[09:34] <ogra> *debian
[09:34] <ogra> ARGH
[09:34] <ogra> *d e b i a n
[09:35] <LaserJock> why?
[09:36] <ogra> becuse they dont like the idea to copy the sshkeys into the ltsp chroot on every ifup  
[09:37] <rodarvus> can't we check if they are different before copying?
[09:42] <ogra> rodarvus, sure, but that would slow down significanty ... 
[09:42] <ogra> adding grep and stuff
[09:42] <ogra> i podered it ... but i think just copying them is way faster
[09:43] <ogra> mdz asked as well about it :)
[09:43] <lfittl> ogra: do you plan to update blender to 2.42 for edgy?
[09:44] <ogra> lfittl, is that the version with all the movie stuff added ? 
[09:44] <lfittl> yep
[09:44] <rodarvus> ogra: I'm reading the backlog on #ltsp now - didn't noticed the conversation you had with vagrantc was there
[09:44] <ogra> i'll look into it ... i think we should ... but i have to find some time to examine
[09:45] <lfittl> ogra: can I help with that somehow?
[09:45] <ogra> rodarvus, well vagrantc is a hardliner ... usually the compromises we find after some discussion are te best solution :)
[09:45] <ogra> lfittl, testbuild it run it ? 
[09:46] <lfittl> ogra: debian has 2.42 and I rebuilt it on edgy, basic stuff works without a problem
[09:46] <ogra> oki
[09:46] <ogra> then i'll request an UVF exception 
[09:46] <lfittl> perfect :)
[09:53] <Amaranth> ogra: if you pull from my branch the packaging should all be good now
[09:53] <ogra> ok
[09:53] <Amaranth> afaik the init script works now
[10:22] <smokeyd> Hey all, is openoffice for dapper 64bit compiled against the same libraries as included in the 32bit emul in dapper 64bit?
[10:22] <smokeyd> Openoffice says it is not with me
[10:22] <smokeyd> On the normal ubuntu channel I only get the repluy: run 32bit
[10:23] <smokeyd> I hope one of you has a more satisfactory solution
[10:23] <smokeyd> see http://www.copypot.com/200
[10:23] <smokeyd> fir the errors I get
[10:24] <smokeyd> I don't mind if a solution takes some work on my part, but I would like to get it to work
[10:25] <ogra> smokeyd, please see the topic ... especially the first sentence
[10:29] <smokeyd> yeah, I know, but I don;t know where support ends and devel starts. This is on the edge and I had the feeling the general ubuntu channel was more, "my printer doesn't work" stugg
[10:29] <smokeyd> stuff
[10:31] <Keybuk> "my openoffice doesn't work" is just as supporty
[10:31] <tseng> there isnt a difference between "my printer doesnt work" and "open office doesnt work" until you are telling us exactly why
[10:31] <tseng> and working towards fixing it
[10:31] <tseng> instead of throwing out a line
[10:31] <rodarvus> Keybuk: are you able to promote universe packages to main?
[10:31] <smokeyd> no, it's a compile problem
[10:31] <smokeyd> I told you
[10:31] <tseng> goodness.
[10:31] <rodarvus> (or rephrasing my question) I need x11proto-print to be promoted to main (since it has become a build-dep for libxaw)
[10:32] <rodarvus> do I need to fill a MainPromotionRequest for it?
[10:32] <smokeyd> The problem is that the glibc version libstdc++.so.6 expects is not the same as provided in the 32bit emul
[10:32] <tseng> smokeyd: now you're getting somewhere.
[10:33] <smokeyd> (22:22:12) smokeyd: Hey all, is openoffice for dapper 64bit compiled against the same libraries as included in the 32bit emul in dapper 64bit?
[10:34] <Keybuk> rodarvus: yes
[10:34] <smokeyd> libstcd++.so.6 is not present in the lib32 folder on dapper
[10:34] <Keybuk> rodarvus: you need an MIR for it though
[10:34] <smokeyd> if I softlink libstdc++.so.5 to libstdc++.so.6
[10:35] <Keybuk> https://wiki.ubuntu.com/UbuntuMainInclusionQueue
[10:35] <rodarvus> Keybuk: I will fill it now - thanks!
[10:35] <smokeyd> I get the error /usr/lib32/libstdc++.so.6: version `GLIBCXX_3.4' not found (required by /usr/lib/openoffice/program/javaldx
[10:43] <Keybuk> smokeyd: see, that's definitely a support question
[10:49] <ogra> Keybuk, is ipconfig a busybox builtin or is there any source package ?
[10:50] <Keybuk> it's a klibc utility
[10:50] <ogra> ah, thanks
[10:55] <Keybuk> Accepted makedev 2.3.1-83 (source)
[10:55] <Keybuk> gnargh!
[10:55] <Keybuk> I hate it when I do that
[10:55] <Keybuk> (used -i and didn't add ubuntuX)
[10:57] <ogra> meh
[10:57] <ogra> but you rejected it early enough ?

[10:58] <ogra> so that looks like "tear down instead of shut down" what you are doing there :)
[10:58] <Keybuk> right
[10:58] <Keybuk> wiki.ubuntu.com/Teardown
[10:58] <ogra> YAY!
[10:58] <ogra> \o/
[10:58] <ogra> yep
[10:58] <ogra> you explained it to me in mataro i think ;)
[10:59] <ogra> i dont forget the good stuff ;)
[10:59] <Keybuk> hehe
[11:00] <Keybuk> dpkg-checkbuilddeps: Unmet build dependencies: bison flex libpam0g-dev mail-transport-agent
[11:00] <Keybuk> WHY THE FUCK DOES THIS NEED AN MTA TO BUILD ?!
[11:00] <ogra> ugh
[11:01] <ogra> to mail you the build errors ?
[11:08] <lucas> how often does the mirror push happen in ubuntu ?
[11:08] <Keybuk> hourly
[11:09] <lucas> ty
[11:12] <wasabi> Is there any other suspend to disk tech besides to a swap file?
[11:15] <Keybuk> wasabi: such as?
[11:16] <wasabi> Eh. Something that can suspend to something without relying on a static partition that must be >= ram size.
[11:16] <wasabi> Ya know. More flexible.
[11:17] <Keybuk> like what? :p
[11:17] <Mithrandir> Keybuk: uh, what build-deps on mta?
[11:17] <wasabi> Eh. Well how does windows do it? I suspect they suspend to a file on the FS.
[11:17] <Keybuk> Mithrandir: "at"
[11:17] <wasabi> And resume, I have no idea.
[11:17] <Keybuk> Build-Depends: bison, flex, libpam0g-dev, mail-transport-agent
[11:18] <Mithrandir> Keybuk: probably lazy configure script or something, I'd guess
[11:18] <wasabi> But if you upgrade your ram in a windows box, hibernate doesn't break.
[11:18] <Keybuk> wasabi: you need something that you can write to without affecting anything in memory
[11:19] <wasabi> Sure. I understand the reasons it's done the way it's done now.
[11:19] <Mithrandir> wasabi: I have an interesting idea for how to do suspend to swap files and with the automatic swap manager rodarvus has been working on, I think we should be able to just put swap on / or something and let it grow automatically.
[11:19] <wasabi> Yeah. MS puts a file in C:\System32\
[11:19] <wasabi> Err, WINDOWS\SYstem32
[11:19] <wasabi> Suspect it's probably static on the FS in some way.
[11:19] <wasabi> No clue there though.
[11:20] <wasabi> Mithrandir: ahh. cool.
[11:20] <wasabi> Mithrandir: So there isn't anything now, but the problem is being worked on.
[11:20] <wasabi> Thanks. :0
[11:20] <Mithrandir> wasabi: I have an idea on how to do it.  I've not yet implemented it, but I was going to tell Scott about it at the distro sprint so he doesn't think my crack level has gone down.
[11:20] <Mithrandir> ;-)
[11:21] <rodarvus> :D
[11:21] <Keybuk> suspend to usb key?
[11:21] <Mithrandir> Keybuk: nah, rewrite initramfs on suspend.  Will work great with runtime assembly of same.
[11:22] <ogra> suspend to pocket ...
[11:22] <Keybuk> why not just give grub a different initramfs option on resume?
[11:22] <Mithrandir> suspend to USB is trivial. :-P
[11:22] <wasabi> I bet MS does something like store the physical disk locations of the swap file in something related to the bootloader
[11:22] <Keybuk> rewriting menu.lst is probably cheaper
[11:22] <Mithrandir> because that would require grub to know if it resumes or not?
[11:22] <Mithrandir> wasabi: that's basically what I'm suggesting.
[11:22] <wasabi> For instance, append resume=/dev/hdc#1234:1234 into the menu.list
[11:22] <wasabi> Yeah.
[11:23] <Mithrandir> you can just tell the kernel "resume from those blocks" or something like it.
[11:23] <wasabi> Yeah.
[11:23] <wasabi> Would have to create a non fragmented file.
[11:23] <wasabi> Or... handle that complexity.
[11:24] <Mithrandir> just give the kernel a list of blocks.
[11:24] <Mithrandir> not hard.
[11:24] <Keybuk> makes a lot of sense
[11:24] <lucas> what's the best place to fetch changes for ubuntu packages from ?
[11:24] <wasabi> Yeah. Right now if our user's upgrade their ram, suspend breaks.
[11:24] <wasabi> Basically.
[11:24] <lucas> http://changelogs.ubuntu.com/ ?
[11:24] <Keybuk> lucas: define changes
[11:24] <lucas> s/changes/changelogs
[11:24] <wasabi> I'm working on swapd right now.
[11:24] <lucas> sorry
[11:25] <Mithrandir> anyway, 'night.
[11:25] <wasabi> Hey Mithrandir, you aware of any way for a user space program to be notified before the OOM killer ... kills?
[11:25] <Mithrandir> see you around.
[11:25] <wasabi> ahh. night. ;)
[11:27] <Keybuk> lucas: right, I think you answered your own question then
[11:28] <lucas> okey
[11:29] <elmo> wasabi: you mean, like, the kernel announcing in a booming voice "WIZARD IS ABOUT TO DIE"?
[11:29] <wasabi> heh.
[11:32] <Keybuk> wasabi: surely  you wouldn't have the memory to start or run such a program? :p
[11:32] <wasabi> Yeah. That's an issue. Basically what I'm doing is fixing swapd.
[11:32] <wasabi> To support a) a minimum number of swap files. So you can always make it manage/create one.
[11:33] <wasabi> b) would like it to have a chance to beat the OOM killer, by adding more swap, before the OOM killer kills.
[11:33] <wasabi> ie, some sort of notification of "memory is this low, and is about to get lower."
[11:33] <Keybuk> MORE SWAP!!  MORE SWAP!!
[11:34] <wasabi> with an additional SIGSTOP to the process that is about to make the system run the OOM killer.
[11:34] <wasabi> ie, it freezes, more swap is created, it proceeds.
[11:35] <Keybuk> yeah, that can be really bad you know
[11:35] <Keybuk> what if it's init that made the system run the OOM killer?
[11:35] <Keybuk> BYE BYE SYSTEM
[11:37] <wasabi> yeah. doesn't usually happen in practice though. the OOM killer basically kills the program using hte most memory.
[11:37] <wasabi> and init isn't usually that.
[11:38] <wasabi> on my system it consistantly takes out vmware, though.
[11:38] <wasabi> which is super annoying.
[11:39] <Keybuk> you are confusing two things though
[11:39] <Keybuk> "the process that is about to make the system run the OOM killer" is the process that asked for more RAM
[11:40] <Keybuk> that is not "the process that the OOM killer kills first"
[11:40] <wasabi> Oh yeah. I know.
[11:40] <Keybuk> clunits.h:88: error: extra qualification 'CLDB::' on member 'get_file_join_coefs'
[11:40] <Keybuk> ^ someone who talks C++ese tell me what that means
[11:40] <wasabi> The OOM killer though doesn't just randomlly choose a process. It chooses the biggest one.
[11:41] <wasabi> That's not likely going to be init.
[11:41] <Keybuk> wasabi: right, but that's not the process you claimed you were going to send SIGSTOP to
[11:41] <wasabi> You're right. I'd stop the one that is about to spawn the OOM killer.
[11:41] <wasabi> So that it doesn't.
[11:41] <Keybuk> erm?
[11:41] <Keybuk> like I said, the process that _spawns_ the OOM killer is not the process that the OOM killer, once spawned, chooses to target
[11:41] <wasabi> RAM gets increased before the kernel code (the thing allocating the page basically) runs the OOM killer.
[11:42] <wasabi> Yeah? I'm just preventing the OOM killer by running.
[11:42] <wasabi> s/by/from/
[11:46] <Keybuk> as an experiment, to demonstrate what I mean, kill -STOP 1 ;p
[11:46] <wasabi> heh. I'm unsure what that will do. I suspect it will stop init, but no children.
[11:46] <wasabi> Am I right or wrong?
[11:46] <wasabi> I'd rather not test right now. ;)
[11:47] <Keybuk> like I said, sending STOP to the process that just needed more RAM may actually have other consequences than you think :p
[11:47] <wasabi> Well, does it stop children too?
[11:48] <Keybuk> actually, I believe it does nothing in this case
[11:48] <wasabi> If so I'd probably do it another way.
[11:48] <Keybuk> (ie. doesn't even stop)
[11:48] <wasabi> Heh.
[11:49] <wasabi> There is going to be a function in the kernel which handles lazy-allocating a page for a process that is trying to write in it. That line of code will basically say if (no pages left!) { run oom killer; }
[11:49] <wasabi> I'm just going to replace it with, { pause calling process; tell swapd to make more ram; wait for swapd to finish; are there free pages yet? no? Oom killer. =( }
[11:50] <Keybuk> how you going to deal with SMP?
[11:50] <wasabi> nfi. =)
[11:50] <wasabi> I'll figure that out when I get there.
[11:50] <Keybuk> or even the basic fact that the kernel is pre-emptive?
[11:50] <Keybuk> so another process might ask for pages while you're in that routine