[01:05] <anir> #suse-es
[01:06] <anir> #suse
[01:06] <Burgwork> anir: hmm?
[01:13] <_ion> I believe she means we should start talking about Ubuntu development on those channels. I think this channel is more suitable, though.
[01:14] <Burgwork> right
[01:21] <LaserJock> I thought this was the opensuse devel channel? doh
[01:21] <LaserJock> ;-)
[01:23] <alex-weej> LOL did anyone get that email from "Shark Muddleworth" on ubuntu-devel?
[01:24] <infinity> No, you're the only one who got it.
[01:24] <infinity> (It's a mailing list, asking if people got the mail seems silly)
[01:25] <crimsun> I'm a bit befuddled at this FTBFS:  "Error: Package: and Architecture: do not alternate in debian/control"
[01:26] <infinity> crimsun: It's either a bug in pkg-create-dbgsyms (bug pitti), or you're actually generating broken control files.
[01:27] <crimsun> http://librarian.launchpad.net/5157781/buildlog_ubuntu-feisty-i386.realtime-lsm_0.8.7-2ubuntu1_FAILEDTOBUILD.txt.gz
[01:27] <crimsun> it's appearing right after dh_strip is called
[01:27] <infinity> Yes, which would be pkg-create-dbgsyms.
[01:27] <crimsun> ok, thanks.
[01:27] <alex-weej> *has anyone read it
[01:27] <infinity> But it doesn't mean your package isn't buggy anyway.
[01:27] <alex-weej> infinity: cock monkey :P
[01:28] <crimsun> infinity: the original sync (which built in Debian) directly from Debian Sid FTBFS, too, so I don't think it's a buggy package
[01:28] <infinity> crimsun: Hrm, the package looks okay at a glance.  But pitti about the error.
[01:28] <crimsun> ok, danke
[01:30] <infinity> Oh, wait.
[01:31] <mjg59> infinity: Dude, Saturday.
[01:31] <infinity> crimsun: The Source stanza has an Arch line.
[01:31] <infinity> crimsun: That's not allowed.
[01:31] <infinity> crimsun: Remove it, SVP.
[01:31] <crimsun> interesting, thanks
[01:31] <infinity> crimsun: (And file a bug in Debian on it, please)
[01:31] <crimsun> will do, with patch pushed.
[01:32] <infinity> mjg59: Saturwhat?
[02:41] <bluefoxicy> jdub:  still trying to get grsec working?
[05:32] <luisbg> anybody knows a dbus service made in python?
[05:49] <Fall2Hell> anyone here?
[05:49] <Hobbsee>  perhaps
[05:50] <Fall2Hell> i need help
[05:50] <Hobbsee> sounds like a #ubuntu type of question
[05:50] <Hobbsee> see the /topic
[05:50] <Fall2Hell> sorry
[05:50] <Fall2Hell> wrong chanel :)
[05:50] <Fall2Hell> thx
[05:51] <Hobbsee> if you're not actually saying what it is...we cant really help
[07:33] <StevenK> sfllaw: Hi. Have you got a sec to talk about wlassistant?
[07:43] <sfllaw> StevenK: Oh man, it's 1:43.
[07:43] <sfllaw> And I'm a bit tipsy.
[07:43] <sfllaw> So technical talk is a bit out.
[07:43] <sfllaw> But I can try.
[07:43] <StevenK> sfllaw: I'd suggest leaving it then.
[07:44] <sfllaw> StevenK: Is this about the SRU?  And my concerns about the usleep()?
[07:44] <sfllaw> StevenK: I can talk about that.
[07:44] <StevenK> sfllaw: Ah.
[07:45] <StevenK> sfllaw: I agree with you, that usleep() may be a bit racy on slower hardware, but what to do instead?
[07:46] <StevenK> sfllaw: My only thought is looping for a maximum of say 5 seconds, checking that an IP has been assigned.
[07:46] <sfllaw> By the time dhclient exits, hasn't it finished configuring things?
[07:46] <StevenK> It forks and daemonizes
[07:46] <sfllaw> Does it daemonize instantly?
[07:46] <sfllaw> Doesn't it write stuff?
[07:47] <StevenK> You're suggesting don't usleep() at all?
[07:47] <sfllaw> It daemonizes _after_ assigning an IP.
[07:47] <sfllaw> Lemme look at the source.
[07:47] <StevenK> I'm happy to upload 3.2 that dumps usleep() or does something else.
[07:52] <sfllaw> StevenK: Hmm.  Looking at WirelessAssistant::runCommand, it looks like it exits after the command it runs does.
[07:52] <sfllaw> Is that true?
[07:54] <StevenK> sfllaw: It also has a timeout.
[07:55] <sfllaw> StevenK: It will return ::ERR::killed if the timeout hits, right?
[07:56] <sfllaw> And then you return in that case.
[07:56] <sfllaw> Due to CONNECTION FAILED.
[07:56] <sfllaw> So I guess I'm not concerned anymore about that usleep.
[07:57] <sfllaw> It's just superfluous.
[07:57] <sfllaw> By the time dhclient exits, it should have set things and run all its exithooks.
[07:57] <sfllaw> OK.  Well, I can test it on Monday and let you know.
[07:57] <StevenK> Right. Thanks.
[07:57] <sfllaw> Did you uncomment that usleep line because it looked like it shouldn't have been commented out?
[07:58] <StevenK> And also I wasn't certain of dhclient's behaviour.
[07:59] <sfllaw> StevenK: UTSL!
[07:59] <StevenK> Yeah, well. :-/
[07:59] <sfllaw> StevenK: Also, always be weary of arbitrary sleeps.
[07:59] <StevenK> Good advice.
[08:02] <infinity> s/weary/wary/, I assume, though "weary of sleep" is cute. :)
[08:02] <sfllaw> infinity: I love bad puns.
[08:03] <infinity> Especially when they can be used as a cover for poor spelling? :P
[08:03] <sfllaw> Indeed.
[08:03] <sfllaw> The puns go up as the wine goes down.
[08:04] <StevenK> archive.ubuntu.com has address 192.168.7.9
[08:04] <StevenK> Let's see it download from the internet now.
[09:36] <dade`> someone using a macbook here ?
[10:09] <raphink> o_O
[10:09] <raphink> linda wishes itself a happy birthday today
[10:09] <raphink> when you launch it
[10:14] <ajmitch> raphink: look in /usr/share/python-support/linda/linda/eggs.py
[10:15] <raphink> hi ajmitch
[10:15] <ajmitch> hello
[10:15] <raphink> :)
[10:17] <raphink> ajmitch: lyricue is on REVU. Do you think you could have a look and complete what I already said about it?
[10:20] <raphink> thanks http://revu.tauware.de/details.py?upid=3461
[02:12] <cjwatson> hmm, oops, we broke USB stick installs in Edgy :-(
[02:12] <cjwatson> bug 66220
[02:12] <Ubugtu> Malone bug 66220 in busybox "ubuntu edgy hd-media image couldn't mount iso file." [High,Confirmed]  http://launchpad.net/bugs/66220
[02:12] <Hobbsee> cjwatson: with the hal update?
[02:12] <cjwatson> Hobbsee: the clue's in "... in busybox"
[02:13] <Hobbsee> ah, that's different, sorry
[02:13] <cjwatson> i.e. no, nothing to do with hal. hal doesn't run in the installer
[02:13] <Hobbsee> yeah, hadnt actually read the report yet.  didnt even see installs in there
[02:18] <Mithrandir> Hobbsee: when you're doing merges, please do actually list the changes still present, not just "All changes preserved"
[02:18] <Hobbsee> Mithrandir: okay.  usually i do.  which was this for?
[02:19] <Mithrandir> Hobbsee: a bunch you did two or three days ago
[02:19] <Hobbsee> hrm
[02:21] <Mithrandir> https://lists.ubuntu.com/archives/feisty-changes/2006-November/001191.html for instance
[02:21] <Hobbsee> ah
[02:21] <Mithrandir> Hobbsee: if you tend to do it, that's fine, I just noticed two or three in a row and thought I should poke you about it.
[02:22] <Hobbsee> i probably did
[02:23] <Hobbsee> Mithrandir: when will heno be back here? office hours, presumably
[02:24] <Mithrandir> Hobbsee: monday, 0800 UTC or thereabout, I'd imagine
[02:24] <Hobbsee> ok
[02:46] <bhale> take that :)
[02:46] <bhale> not BenC long, mind you
[02:49] <Hobbsee> bhale: i wonder if there's a limit on how long they can be.  guess not
[03:52] <bhale> hi pitti 
[03:52] <bhale> someone marked lp #65886 as "security vulnerability", it is a normal segfault
[03:52] <bhale> can you debunk?
[03:52] <bhale> "This bug may be a security problem since segfaults in unmanaged native code are often exploitable."
[03:53] <Mithrandir> bhale: but you can run code as yourself!
[03:53] <bhale> Mithrandir: oh nose!
[03:53] <pradeep> bug 65886
[03:53] <Ubugtu> Malone bug 65886 in beagle "beagle segfaults in libwv native code" [Undecided,Unconfirmed]  http://launchpad.net/bugs/65886
[03:53] <bhale> ah that is where I can unmark it
[03:57] <lamont> pitti: any word from those logfiles?
[04:01] <pitti> hi bhale
[04:02] <pitti> lamont: I looked at them, but they gave me no clue
[04:02] <lamont> pitti: figures
[04:02] <lamont> note that nss has ldap involved, although the user in question resolves completely within 'files'
[04:02] <pitti> lamont: a hal debug log *might*
[04:03] <lamont> pitti: ok.  same instructions as edgy?>
[04:03] <lamont> er, dapper
[04:03] <pitti> yes
 bhale: but you can run code as yourself!
[04:08] <bluefoxicy> Mithrandir:  actually a hole in libwv would more likely mean someone could dcc you a word document, your auto-accept would stick it in ~/dcc/, and then beagle would index it and libwv would run code from it
[04:08] <bluefoxicy> but you have to actually have an exploitable hole; some segfaults are, but some fall short.
[04:09] <Mithrandir> bluefoxicy: don't take candy from stangers.
[04:09] <Ng> xchat-gnome in edgy defaults to auto-accepting dcc :/
[04:09] <bluefoxicy> Mithrandir:  I know I know, people fill my disk up once in a while
[04:23] <sladen> mpt: you'll love http://www.joelonsoftware.com/items/2006/11/21.html
[04:26] <Mithrandir> sladen: as somebody in here commented - we can't suspend and hibernate reliably enough yet and it also doesn't give the user the ability to differentiate between "I'm going away" and "I'm going away and would like my programs to keep running".
[04:27] <sladen> Mithrandir: what I found scary was that all the continion we'd had over the "big logout dialogue" and to find out that MS had come up with the same "show them all"
[04:43] <cjwatson> bhale,Mithrandir: given that it's while indexing mailboxes, surely "somebody can send you a mail that can run code as you" is closer to the mark than "you can run code as yourself"
[04:44] <cjwatson> Joel employs the disgraceful debating tactic of pre-emptively discarding any argument that disagrees with him
[04:44] <bhale> cjwatson: one of the bigger problems with beagle (or anything like it)
[04:48] <sladen> Mithrandir: while you're around, did you have further ideas to add at the end of https://wiki.ubuntu.com/LiveCDStackedFileSystem
[04:49] <Mithrandir> cjwatson: it's harder to argue your mail server shouldn't take candy from strangers, I guess.
[04:49] <Mithrandir> sladen: yes, but not on a Saturday early evening.
[05:05] <siretart> is nov 30 still targeted for herd1? there seem to be no dailies yet
[05:56] <cjwatson> siretart: hard to say yet - I haven't finished merging the installer, which is why there are no dailies
[05:58] <cjwatson> working on that as fast as possible
[05:58] <cjwatson> ubiquity will need a half-day's work to deal with partman-auto changes
[05:58] <cjwatson> if the first CDs Just Work then the 30th may be possible, but if not then it may be optimistic ;-)
[06:07] <siretart> cjwatson: no problem 'bout that. I just looked at FeistyReleaseSchedule and wondered that herd1 is scheduled that early from now
[06:24] <MatthewG> fsck.ext3 -nv /dev/sda1 reports this ... http://paste.ubuntu-nl.org/33925/ ... this is a second box, and it shows the exact same things. What gives? It's ubuntu server install with LAMP option. Is ubuntu installing a borked filesystem or something? 
[06:38] <revoltee> hi
[06:38] <revoltee> how is ubuntu comparatively tyo deian
[06:38] <revoltee> debian
[06:38] <revoltee> very different?
[06:38] <revoltee> i mean kernel wise
[06:40] <Lathiat> kernels are quite different yes
[06:40] <Lathiat> notably ubuntu adds drivers and makes some modifitions etc, AIUI debian packages pure stock kernels?
[06:42] <siretart> Lathiat: the debian kernels are patched quite heavily as well. e.g. they have kernel images with linux vserver and xen and both
[06:42] <sladen> the debian kernels are not 'stock'---they too have patches, as required
[06:42] <Lathiat> oh hrm ok, i thougth they didnt at all
[06:42] <sladen> but not quite the number of driver/latest crack patches that Ubuntu attempts to ship
[06:42] <Lathiat> apparently i was wrong :)
[06:42] <revoltee> so in short
[06:43] <revoltee> debian kernels are way better
[06:43] <siretart> revoltee: who says this?
[06:43] <Lathiat> "better" is subjective
[06:43] <revoltee> siretart: im asking
[06:43] <siretart> revoltee: the debian kernels support less hardware. what do you mean with 'better'?
[06:43] <revoltee> better as in performance
[06:43] <revoltee> optimizatioin
[06:44] <neuralis> revoltee: how did you infer anything at all about performance from the conversation above?
[06:45] <neuralis> revoltee: we ship a desktop kernel, a server kenrel, a supercomputer kernel, and a xen kernel. all are tuned to their respective uses.
[06:45] <siretart> IIRC ubuntu's kernel is compiled with SSP. so it is potentially more 'secure' 
[06:45] <sladen> ..looking carefully at your userspace is far more likely to create gains than tinkering with the kernel
[07:46] <ssam> is there anyone around that i could talk to about a possible security bug in ubuntu (i think it might be specific to powerpc)
[07:47] <Laser_away> ssam: it seems pitti isn't here at the moment
[07:48] <Laser_away> ssam: you might just want to file the bug and make sure to mark it as a security bug
[07:49] <ssam> Laser_away, ok, i just thought i should check that i was not being silly first
[08:28] <revoltee> i test all the live installations of linux .. in vmware before really trying to install or run it live with  a reboot ..
[08:28] <revoltee> o how does it actually partition the RAM and stuff and run an OS inside an OS
[08:28] <revoltee> so
[08:28] <revoltee> just need a basic idea of how vmware does this
[08:28] <revoltee>  question is how is the bridging done ..?
[08:36] <siretart> revoltee: this channel is about ubuntu development. please ask in #ubuntu for end user support
[08:44] <ajmitch> hi siretart 
[09:11] <ogra> BenC, around by chance ?
[09:41] <siretart> hey ajmitch 
[10:25] <JB[away] > Any news about this bug ? -> https://bugs.launchpad.net/distros/ubuntu/+source/apache2/+bug/62820
[10:25] <Ubugtu> Malone bug 62820 in apache2 "RPC over HTTP" [Medium,Unconfirmed]