|
[00:19] <persia> ScottK, Excellent. Thanks. |
|
[02:53] <LimCore> elkbuntu: well I responded to all arguments presented in a logic way, if you fell otherwise, fell free to say which ones you think where not clear and we can find a solution :) |
|
[02:54] * LimCore hopes this channel is correct for discussion of developing apps for ubuntu |
|
[02:55] <Hobbsee> LimCore: how much data corruption actually occurs, on ext3, in ubuntu? |
|
[02:55] <Hobbsee> that you've found bug reports for? |
|
[02:55] <LimCore> ok, I will sacrafice my brand new interpid install to test it |
|
[02:56] * LimCore notices that laptop is not the best test tool for power-outage... but searches for battery slot |
|
[02:56] <Hobbsee> i thought you said you were using XFS? |
|
[02:56] <LimCore> as I said, form what I heared the problem would be most visible of XFS; I also said that I seen such corruption in ext3, but long ago. Currently I use ext3 and jfs |
|
[02:57] <Hobbsee> i didn't ask about xfs. I asked about the one we ship by default, and almost all users use. |
|
[02:57] <LimCore> yes, therefore Im going to test ext3, right? |
|
[02:57] <Hobbsee> good |
|
[02:57] <jdong> it is almost entirely an XFS problem |
|
[02:58] <jdong> XFS journals too tersely and takes the disk too seriously when it says something is written |
|
[02:58] <LimCore> jdong: as I written 3 times so far in last 15 minutes, I had ext3 corruptions (serious one) on ext3 |
|
[02:58] <jdong> furthermore I should point out that XFS zeroes incomplete transaction datablocks rather than keeping stale (and most of the time probably correct) data, which can make corruption more apparent |
|
[02:58] <jdong> LimCore: what kind of corruption? |
|
[02:59] <LimCore> jdong: I will test again now |
|
[02:59] <jdong> I asked what kind of corruption. |
|
[02:59] <LimCore> lost files |
|
[02:59] <LimCore> *afair* even from dirs that where not at all written to like /usr (but perhaps atime was written) |
|
[03:00] <ion_> Why is everybody badmouthing XF\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0 |
|
[03:01] <LimCore> any existing tool for this? |
|
[03:01] <LimCore> write lots of files with known content, then start swapping them around etc |
|
[03:01] <jdong> ion_: :D |
|
[03:02] <jdong> well you can use bonnie++ or mongrel (whatever Reiser's metadata thrasher is called) |
|
[03:03] <LimCore> but I found it strange to prove to you that pulling out power cable can indeed corrupt the FS (not just the file that was actually written to, because this is obvious) |
|
[03:04] <LimCore> Hobbsee: kid using computer is the only reason to not like the power off button? |
|
[03:05] <Hobbsee> LimCore: well, the general complaint about "lock screen" doesn't actually do that anymore is. |
|
[03:05] <LimCore> that is another of my bugs |
|
[03:05] * LimCore have a bug for every ocassion \o/ |
|
[03:05] <Hobbsee> yeah, it's closer to lock session, i guess. |
|
[03:05] <LimCore> Hobbsee: for about 1.5+ year lockscreen is broken: in some cases it shows what is on the desktop |
|
[03:06] <jdong> LimCore: that bug has been fixed as far as I know of |
|
[03:06] <Hobbsee> LimCore: ah yes, that's a propriatory driver triggered bug, no? |
|
[03:06] <Hobbsee> or has been fixed. |
|
[03:06] <LimCore> Hobbsee: mostly nvidia |
|
[03:06] <LimCore> but! |
|
[03:06] <jdong> there was one related to compiz too. |
|
[03:06] <Hobbsee> jdong: yeah, i think i saw that one for a bit. That definetly got fixed. |
|
[03:06] <LimCore> not only. other bug was on intell, when one changed screen resolution. no compiz. the black rectangle blocked only the small part of screen, if resolution was switched in some way like grandr |
|
[03:07] <LimCore> 1 year ago. dunno if fixed. |
|
[03:07] <jdong> if the screen resolution was changed... while the screensaver is active? |
|
[03:07] <LimCore> btw... black rectangle... imho this is really braindead approach to locking the screen O__o |
|
[03:07] <Hobbsee> never saw that one. |
|
[03:07] <LimCore> jdong: no, before it was activated |
|
[03:07] <Hobbsee> LimCore: surely you'll have to talk to upstream about that. |
|
[03:07] <jdong> LimCore: that is how every operating system locks the screen. |
|
[03:07] <jdong> by drawing a fullscreen overlay on top of it |
|
[03:07] <LimCore> well, then they are all braindead |
|
[03:08] <LimCore> obvious appraoch would be, at least to me |
|
[03:08] <jdong> except Windows, which uses its output redirection capabilities of terminal services to connect to the LSASS login prompt |
|
[03:08] <Hobbsee> would be to go talk to upstream about it. |
|
[03:08] <LimCore> 1) clear entire screen / "video memory" 2) NOT draw anything |
|
[03:08] <jdong> LimCore: that's complete nonsense |
|
[03:08] <Hobbsee> again, this is not on topic for here - we do not create X, etc, here. |
|
[03:09] <LimCore> heh Hobbsee you really keep interrupting constructive discussions today :) ok then, jdong talk to me in ##linux or ##c++ if you want about this |
|
[03:09] <jdong> I am not really interested |
|
[03:09] <jdong> I don't feel such a conversation would be more productive than what I am currently working on |
|
[03:10] <Hobbsee> LimCore: not constructive for packaging things already made, sorry. |
|
[03:10] <LimCore> ok, for another time then |
|
[03:10] <Hobbsee> LimCore: and things that happen only sometimes on XFS aren't release critical. That's the way it is, i'm afraid. |
|
[03:10] <Hobbsee> But fix ti for jaunty. That'd actually be helpful. |
|
[03:12] <ion_> I fail to understand how users randomly pulling the power plug is more important a use case than e.g. users randomly throwing the computer out of the window. Power supply problems? Get a UPS. Kids meddling with the computer? Restrict their access. |
|
[03:12] <LimCore> ion_: you didnt apparently read the bug report we are refering to? |
|
[03:12] <jdong> well not to say that filesystem integrity isn't an area that we can improve on as an operating system... |
|
[03:13] <jdong> but it's not something that's logical to waste the time discussing while we have more than enough work on our plate to push intrepid out in one piece! |
|
[03:13] <LimCore> ok Im totally dissapointed in this. I will just make my tests and my patch. |
|
[03:13] <Hobbsee> good. |
|
[03:13] <slangasek> I propose that we push intrepid out in three pieces that reassemble themselves once they're out |
|
[03:13] <jdong> slangasek: doesn't that .torrent file thing already do that? |
|
[03:13] <Hobbsee> slangasek: before or after defenestrating themselves? |
|
[03:14] <slangasek> jdong: oh, then I guess my idea is unoriginal |
|
[03:14] <LimCore> btw, you are still aiming at 2008-10-30 ? O_o |
|
[03:14] <Hobbsee> LimCore: yes... |
|
[03:14] <jdong> slangasek: but I think another distro beat you at what you actually meant: http://wiki.mandriva.com/en/2009.0_Errata |
|
[03:14] <LimCore> as software manager I would rate it as impossible, looking at KDE state... but. Can one test easly? |
|
[03:14] <jdong> that is either the most brutally honest errata list or one rushed cycle :) |
|
[03:15] <LimCore> I didnt seen a clear easy to use wiki page on how to run 100% automated smoke test? |
|
[03:15] <Hobbsee> LimCore: which is precisely *why* these channels arekept more or less on topic, closer to release, and we don't allow random rants in either channel. |
|
[03:16] <slangasek> jdong: hmm? how is mandriva what I meant? :) |
|
[03:16] <jdong> slangasek: the whole push it out the door in a couple pieces and post assembly instructions later idea :) |
|
[03:17] <LimCore> wouldn't it be better to make wait +1 month |
|
[03:17] <slangasek> ah |
|
[03:17] <jdong> LimCore: no |
|
[03:17] <jdong> you can make that argument for an endless stream of bugs. |
|
[03:18] * LimCore buys 1 container of coffee |
|
[03:24] <LimCore> when is deadline of accepting patches? |
|
[03:24] <jdong> depends on the nature of the patch |
|
[03:24] <jdong> but extremely soon. |
|
[03:25] <LimCore> like, 2 days? |
|
[03:25] <Hobbsee> you've probably got slightly longer than that. |
|
[03:25] <Hobbsee> but, the longer you pushit, the less chance you've got of it being accepted. |
|
[03:28] <nixternal> so what is on the agenda this evening? |
|
[03:28] <Hobbsee> nixternal: bug fixing! |
|
[03:28] <jdong> world domination |
|
[03:28] <nixternal> eww boogs :) |
|
[03:28] <Hobbsee> nixternal: https://bugs.edge.launchpad.net/ubuntu/intrepid/+bugs?field.searchtext=&orderby=-importance&search=Search&field.importance%3Alist=CRITICAL&field.importance%3Alist=HIGH and such |
|
[03:29] <nixternal> holy smokes, tinyurl rocks! |
|
[03:29] <nixternal> ;p |
|
[03:29] <Hobbsee> nixternal: https://bugs.edge.launchpad.net/ubuntu/intrepid/+bugs?field.searchtext=&orderby=-importance&search=Search too |
|
[03:29] <Hobbsee> (the first is a subset of the second) |
|
[03:31] <YokoZar> Wine just made a very minor release today (1.0.1), designed to have just minimal patches and remain stable (most of the patches are translation fixes)...would it be appropriate to upload it to Intrepid now, rather than SRUing it in a couple weeks anyway? |
|
[03:32] <jdong> meh it's Wine, I don't personally see how it would hurt to follow upstream's latest releases available |
|
[03:32] <Hobbsee> YokoZar: er...have you had this in a PPA / tested it yourself / etc? |
|
[03:32] <jdong> (I don't have any release management capacities though) |
|
[03:32] <Hobbsee> YokoZar: assuming that the answer to the above is "yes", then i'd assume it should be fine. |
|
[03:33] <Hobbsee> YokoZar: if it's features only, it won't even need a ffw |
|
[03:33] <Hobbsee> er, ffe |
|
[03:56] <Hobbsee> emgent: when are you planning on fixing https://bugs.edge.launchpad.net/ubuntu/intrepid/+source/fetchmail/+bug/240549 ? |
|
[03:56] <ubottu> Launchpad bug 240549 in fetchmail "fetchmail denial of service CVE-2008-2711" [Unknown,Fix released] |
|
[03:56] <Hobbsee> emgent: it's probably easier to fix it before release than after. |
|
[04:00] <YokoZar> jdong: Wine's been in a "stable/unstable" branch mode since 1.0 came out. Shipping latest beta would cause regressions...but latest stable shouldn't |
|
[04:01] <ajmitch> many people that rely on wine probably use your other packages anyway :) |
|
[04:10] <nixternal> I rely on beer, not wine! |
|
=== LimCore_ is now known as LimCore |
|
[04:30] * Hobbsee wonders how the sponsorship queue is going. |
|
[04:39] <Hobbsee> !ffe |
|
[04:39] <ubottu> uvf is Upstream Version Freeze. For an exception, see https://wiki.ubuntu.com/FreezeExceptionProcess#head-9523bc4076ff011324d67cddc97969ec609618d6 |
|
[04:49] <Hobbsee> !5-a-day |
|
[04:49] <ubottu> 5-a-day is a community event where each person will take 5 bugs a day and work on them. Everyone is invited to help no matter your abilities! More information available at https://wiki.ubuntu.com/5-A-Day |
|
[04:55] <LimCore> is there something to *report* N-bugs-a-day? I would totally win |
|
[04:56] <Hobbsee> LimCore: no. you have to actually fix them. |
|
[04:56] <Hobbsee> LimCore: actually, i think there is. it's called brainstorm. |
|
[04:56] <wgrant> I think I can win at N-launchpad-bugs-a-day |
|
[04:56] <wgrant> I filed 15 once. |
|
[04:56] <wgrant> Hobbsee: ROFL |
|
[04:56] <Hobbsee> but that's more for ranting, rather than actual bugs that you want to get fixed |
|
[04:56] <LimCore> brainstorm: where you can fill in brillaint ideas! vote for them \o/ ! and finally.... do nothing about them. |
|
[04:56] <Hobbsee> yeah, pretty much |
|
[04:57] <Hobbsee> and hope tha tsomeone else does |
|
[04:57] * ajmitch votes that Hobbsee replace the archive with checkinstall & alien |
|
[04:58] <Hobbsee> ajmitch: argh. Now you *have* to help fix bugs. |
|
[04:58] <NCommander> wgrant, I had a 12 task SRU :-) |
|
[04:58] <ajmitch> Hobbsee: or else...? |
|
[04:58] <wgrant> NCommander: gnat-4.2? |
|
[04:58] <NCommander> wgrant, yup |
|
[04:58] <Hobbsee> ajmitch: the Long Pointy Stick of DOOM!!!!!!!!!!!!!!!™ might come out |
|
[04:58] <NCommander> !hobbsee Hobbsee |
|
[04:58] <ubottu> Sorry, I don't know anything about hobbsee hobbsee |
|
[04:58] <NCommander> argh |
|
[04:58] <NCommander> !hobbsee |
|
[04:58] <ubottu> I phear the stick so shhhhh |
|
[04:58] <ajmitch> sure, if you feel like visiting NZ |
|
[04:58] <Hobbsee> right. 10 bugs done today. |
|
[04:58] <LimCore> canonical should introduce 10 usd donation-licence thingy, and use that to pay devels |
|
[04:58] <wgrant> !ncommander |
|
[04:58] <ubottu> Sorry, I don't know anything about ncommander |
|
[04:58] <wgrant> Damn. |
|
[04:58] <NCommander> I know :-/ |
|
[04:58] <StevenK> LimCore: Um. Ow. |
|
[04:58] <wgrant> You have no factoid yet :( |
|
[04:58] <NCommander> I feel so unloved |
|
[04:59] <ajmitch> !wgrant |
|
[04:59] <ubottu> Sorry, I don't know anything about wgrant |
|
[04:59] <NCommander> !wgrant |
|
[04:59] <wgrant> Bah. |
|
[04:59] <NCommander> Oh |
|
[04:59] <Hobbsee> !msgthebot |
|
[04:59] <ubottu> Please investigate with me only with "/msg ubottu Bot" or in #ubuntu-bots. Search for factoids with "/msg ubottu !search factoid". |
|
[04:59] <NCommander> lol |
|
[04:59] <StevenK> I don't think I have one either |
|
[04:59] <ajmitch> you're not the only unloved one |
|
[04:59] <wgrant> Mine used to be an alias for !nixternal |
|
[04:59] <StevenK> Haha |
|
[04:59] <NCommander> !nixternal |
|
[04:59] <ubottu> Oh no! The pointy-clicky Vista lover has arrived! He's rumoured to be giving out free money, too! |
|
[04:59] <nixternal> NOOOOOOOOOOO!!!!!!!! |
|
[04:59] <NCommander> O_O; |
|
[04:59] <wgrant> (I've never actually used Vista) |
|
[04:59] * NCommander has |
|
[04:59] <Hobbsee> nixternal: how many bugs have you fixed today? |
|
[04:59] * StevenK has :-/ |
|
[04:59] <nixternal> I have..you aren't missing a damn thing |
|
[04:59] <LimCore> wgrant: did you use Mojave? |
|
[04:59] <nixternal> Hobbsee: to many, but they have been for work and not here |
|
[05:00] <wgrant> LimCore: No, but I hear it's very good. |
|
[05:00] <Hobbsee> nixternal: oh well, get to it. |
|
[05:00] <nixternal> I am setting up my development environment now so I can do some work here |
|
[05:00] * nixternal wonders how many hard drives have to fail before dell gets it through their thick skull |
|
[05:00] <Hobbsee> interesting. soyuz borks if there's not a legitimate email address in debian/changelog |
|
[05:00] <StevenK> Sounds right |
|
[05:01] <wgrant> Hobbsee: Of course. It mails it, so it needs it to be valid. |
|
[05:01] <StevenK> Like, correct behaviour |
|
[05:01] <LimCore> 1. brainstorm |
|
[05:01] <LimCore> 2. community voting |
|
[05:01] <LimCore> 3. ..voting with donations (like vim) |
|
[05:01] <LimCore> 4. PROFIT |
|
[05:01] <nixternal> 5. WORLD DOMINATION |
|
[05:01] <LimCore> and use that for x2 bigger team and make 8.10 on time \o/ |
|
[05:01] <nixternal> though #4 is good |
|
[05:01] <wgrant> 8.10 will be on time... |
|
[05:01] <Hobbsee> wgrant: i'd not tried it before... |
|
[05:02] <wgrant> I'd seen somebody do it once. |
|
[05:02] <wgrant> It used to silently reject. |
|
[05:02] <LimCore> to make 8.10 on time easly |
|
[05:02] <wgrant> Not even email the signer. |
|
[05:02] <wgrant> LimCore: It can't be late, so therefore it will be on time. |
|
[05:05] <Hobbsee> right, there we go. |
|
[05:33] <LimCore> interpid have video distortions (shows random noise instead texture in time to time) |
|
[05:34] <LimCore> and if you run serveral opengl programs like amoeba --nosound --windowed & then it shows more quircks, then freezes (no limit on mem usage?). report? |
|
[05:35] <fabrice_sp> Hi. Some MOTU willing to sponsor bug #284786? |
|
[05:35] <ubottu> Launchpad bug 284786 in icedove-nostalgy "icedove-nostalgy: Depends: icedove (>= 1.5) but it is not installable" [Undecided,Confirmed] https://launchpad.net/bugs/284786 |
|
[05:35] <fabrice_sp> (for Intrepid) |
|
=== fabrice_sp_ is now known as fabrice_sp |
|
[05:39] <Hobbsee> how do users not manage to have a /boot/grub? |
|
[05:40] <Hobbsee> hm, unless they're running lilo, i guess |
|
[05:41] <StevenK> Right |
|
[05:41] <StevenK> Or aren't on x86 hardware |
|
[05:45] <LimCore> when reporting a bug, the it-affects-me-also flag should be ON by default. where to report this? |
|
[05:45] <jdong> malone. |
|
[05:45] <jdong> but it's kind of pointless |
|
[05:45] <jdong> we know the bug affects the original reporter |
|
[05:46] <wgrant> There are at least two bugs filed on that issue. |
|
[05:46] <jdong> fabrice_sp: looking at nostalgy... |
|
[05:46] <fabrice_sp> jdong: great. Thanks! |
|
[05:47] <Hobbsee> LimCore: #launchpad forall launchpad stuff. |
|
[05:47] <LimCore> I once said that installing sshd should really strongly warn that you are about to run an ssh SERVER. Its up for debate, but if we agree, then the fix is trivial |
|
[05:47] <wgrant> We don't agree. |
|
[05:47] <wgrant> And even if we do, it's not for this channel. |
|
[05:48] <LimCore> then which one |
|
[05:48] <Hobbsee> and not this close to release. |
|
[05:48] <Hobbsee> it's not a release critical bug, therefore is extremely unlikely to get fixed. |
|
[05:49] <jdong> when I install a package called openssh-server, I expect..... |
|
[05:49] <Hobbsee> jdong: to not have to read, think, or do anythingelse. duh. |
|
[05:50] <LimCore> your unwillingness to improve software to make it more user friendly / idiot proof a bit dissapoints me. is it just the close date of 8.10 or what |
|
[05:50] <Hobbsee> LimCore: oddly enough, yes. Have you even heard of regressions? |
|
[05:50] <LimCore> Hobbsee: yes. ubuntu, not gentoo or debian |
|
[05:50] <jdong> what? |
|
[05:50] <wgrant> I am now officially confused. |
|
[05:51] <jdong> well, fabrice_sp, your icedove^H^H^H^Hthunderbird plugin is uploaded. :) |
|
[05:51] <Hobbsee> LimCore: presumably also advocates patching out 'rm', and other dangerous commands. |
|
[05:51] * Hobbsee shakes head. |
|
[05:51] <jdong> they should prompt for each file |
|
[05:51] <LimCore> ubuntu is not gentoo or debian - it should make it hard for someone to shoot self in the foot. Especially if all it takes it so use password like "bob3" and install random packages to "try out the stuff" (both happen) |
|
[05:51] <wgrant> Technical Services at uni does that on their Solaris machines. |
|
[05:51] <wgrant> So everybody ends up shooting themselves in the foot at home. |
|
[05:51] <jdong> can we move the silly discussions off to devel-discuss? |
|
[05:52] <wgrant> jdong: Heh. |
|
[05:52] <jdong> I'd actually like to use these channels for development as intended. |
|
[05:52] <fabrice_sp> jdong: greaaat! Only 53 packages left that don't install on my amd64 ;-) |
|
[05:52] <Hobbsee> jdong: me too. |
|
[05:52] * Hobbsee waves the warning flag |
|
[05:52] <jdong> fabrice_sp: almost there... at 5-aday, that's... oh wait... |
|
[05:52] * wgrant feels threatened. |
|
[05:54] <fabrice_sp> jdong: and as some packages are not amd64 ready, I'm sure it's a lot less on i386 ;-) |
|
[05:54] <jdong> fabrice_sp: I hope so. Thanks for taking an interest in installability tests :) |
|
[05:54] <wgrant> fabrice_sp: See http://qa.ubuntuwire.com/debcheck/ |
|
[05:55] <wgrant> See the Depends line in particular. Those are uninstallable. |
|
[05:56] <fabrice_sp> Actually, I'm getting the list with a command I saw here: apt-cache unmet -i | grep "^El paquete" | wc |
|
[05:56] <fabrice_sp> ("^El paquete" because of the spanish language, of course) |
|
[05:58] <jdong> a bit OT, where has imbrandon been these past few months? I feel like I'm out of the loop |
|
[05:59] <wgrant> You're no more out of the loop than others, AFAIK. |
|
[05:59] <Hobbsee> fabrice_sp: it's even accepted. |
|
[06:02] <fabrice_sp> Hobbsee: do you mean that spanish language is accepted? |
|
[06:02] <Hobbsee> fabrice_sp: no, your package. icedove-* |
|
[06:03] <fabrice_sp> Hobbsee: Ahhh. It's not the first 'icedove*' package I patch |
|
[06:03] <fabrice_sp> ;-) |
|
[06:03] <fabrice_sp> to thunderbird* |
|
[06:03] <Hobbsee> fabrice_sp: right, but at the moment, we're in a freeze, so everything's being manually accepted. |
|
[06:03] <Hobbsee> (or manually declined) |
|
[06:04] <fabrice_sp> I thought we had a kind of 'exception' for Universe |
|
[06:04] <fabrice_sp> for a few days |
|
[06:04] <wgrant> Yes, but it still needs to be manually pushed through. |
|
[06:05] <wgrant> The whole archive is frozen, but universe/multiverse uploads don't need an exception request - archive admins like Hobbsee just wave them through. |
|
[06:05] <fabrice_sp> So thanks to all archive admins to make that easier ;-) |
|
[06:05] <Hobbsee> you're welcome |
|
[06:08] * NCommander hugs Hobbsee |
|
[06:09] <Hobbsee> NCommander: :) |
|
[06:09] <Hobbsee> nixternal: how does https://help.ubuntu.com/ubuntu/desktopguide/C/video.html get updated? |
|
[06:10] <nixternal> Hobbsee: it doesn't...the desktop guide has been for a long time |
|
[06:10] <Hobbsee> nixternal: missing word, and if that's the case, can you update !multimedia please? |
|
[06:11] <jdong> is "sudo /usr/share/doc/libdvdread3/examples/install-css.sh" really our preferred way of getting CSS packages? |
|
[06:12] <jdong> doesn't that script still fetch marillat's Debian debs? |
|
[06:12] <wgrant> Is there a better way? |
|
[06:12] <fabrice_sp> Hobbsee: and obviously to you for approving it :-) |
|
[06:12] <Hobbsee> jdong: no, it has'nt for quite some time. |
|
[06:12] <Hobbsee> jdong: it got modified tofetch medibuntu ones. |
|
[06:12] <wgrant> Ah. |
|
[06:12] <jdong> Hobbsee: sweet, awesomeness! |
|
[06:12] <Hobbsee> fabrice_sp: :) |
|
[07:21] <eddyMul> I'm trying to fix LP 257724 by re-packaging xinit and making /etc/X11/xinit/xinitrc executable |
|
[07:21] <ubottu> Launchpad bug 257724 in vnc4 "vnc4server startup script error" [Undecided,New] https://launchpad.net/bugs/257724 |
|
[07:22] <eddyMul> problem: initial /etc/X11/xinit/xinitrc is not executable. installing my newer xinit package doesn't override xinitrc's permissions |
|
[07:22] <eddyMul> any suggestions? |
|
[07:32] <fabrice_sp> eddyMul: How are your applying the change? |
|
[07:35] <eddyMul> fabrice_sp: ... good question. I was able to make /etc/X11/xinit/xinitrc executable in the .deb, but when I install the .deb into my system, the old xinitrc is not overwritten. |
|
[07:35] <eddyMul> fabrice_sp: ... good question. I was able to make "/etc/X11/xinit/xinitrc" executable in the .deb, but when I install the .deb into my system, the old xinitrc is not overwritten. |
|
[07:36] <eddyMul> fabrice_sp: I use "GDebi" to "apply" the change |
|
[07:36] <G__81> I would like to be part of the Ubuntu Developer team how do i start off. I am not good at packaging stuff but could be really helpful in Bug Fixing,Testing, Documentation and Marketing . I have been a contributor to Fedora |
|
[07:37] <fabrice_sp> eddyMul: I think you should use a postinst script |
|
[07:38] <eddyMul> fabrice_sp: I'll try that! thanks |
|
[07:39] <fabrice_sp> G__81: you can begin there: https://wiki.ubuntu.com/MOTU/Contributing |
|
[07:39] <fabrice_sp> eddyMul: you're welcome |
|
[07:39] <G__81> hi fabrice_sp just correct me if i am wrong, is MOTU only for packaging ? |
|
[07:42] <fabrice_sp> G__81: No. It's all about the Universe and Multiverse repository. You can have more info at https://wiki.ubuntu.com/MOTU |
|
[07:42] <fabrice_sp> that means: They are community members who spend their time adding, maintaining, and supporting as much as possible the software found in Universe |
|
[07:43] <fabrice_sp> (by the way, I'm not a MOTU :-) ) |
|
[07:43] <G__81> so MOTU is the only way to contribute ? |
|
[07:44] <G__81> This is what the wiki says |
|
[07:44] <G__81> MOTU contributors are the people who are interested in contributing to Ubuntu and are learning how to package and work in the Ubuntu development community. |
|
[07:44] <G__81> But if i am not good at packaging or rather have not done packaging before |
|
[07:44] <fabrice_sp> no: I'm not a MOTU, and I am contributing |
|
[07:44] <G__81> then i cannot become a MOTU at all :) |
|
[07:45] <fabrice_sp> and packaging in debian world is a lot easier than in others |
|
[07:46] <fabrice_sp> there are several 'steps' before becoming a MOTU. You can have a good picture there: https://wiki.ubuntu.com/UbuntuDevelopers |
|
[07:47] <G__81> fabrice_sp, Ok to introduce myself, I work on the kernel as a hobby, and contribute to fedora. I have used ubuntu but currently using Fedora 9 in my laptop. I thought i could start contributing to Ubuntu. My efforts in Fedora are not being recognized unfortunately :( |
|
[07:47] <G__81> but thats alone not the reason too |
|
[07:48] <G__81> or fedora does not have a proper system to do so. The LaunchPad is really good in that sense you get points and stuff like that but you dont have that in Fedora |
|
[07:49] <fabrice_sp> a simple way to begin is to help solving bugs |
|
[07:49] <G__81> i am also a fedora ambassador for my region |
|
[07:49] <G__81> fabrice_sp, yeah exactly |
|
[07:50] <G__81> fabrice_sp, i am very much interested in the kernel too so how easy is to hack a kernel running Ubuntu |
|
[07:50] <G__81> i hear people saying that Ubuntu is for end users and hacking a kernel or playing with the kernel is tough in Ubuntu. is it so ? |
|
[07:52] * G__81 waits |
|
[07:52] <wgrant> This isn't really a relevant channel. |
|
[07:52] <wgrant> But I've never found problems doing it. |
|
[07:53] <G__81> wgrant, ok |
|
[07:57] <NCommander> hey asac |
|
[07:59] <G__81> so i have decided to contribute to Ubuntu |
|
[08:00] <G__81> I am going to start my journey keeping my fingers crossed . I really look forward for this |
|
[08:00] <G__81> i really look forward to be part of this team |
|
[08:05] * G__81 wishes someone welcomes me |
|
[08:05] <G__81> :) |
|
[08:05] <slangasek> G__81: the #ubuntu-kernel channel is probably the best place to start getting involved with kernel maintenance in Ubuntu; the kernel team has git trees for the Ubuntu kernel, so that should be familiar to you, you can feed changes to them in the usual git-y ways - though of course with the kernel as large a project as it is, we prefer as much of the changes as possible to be pushed upstream first |
|
[08:06] <G__81> slangasek, yeah thanks |
|
[08:09] <G__81> thanks for your reply and i wish to start off from today |
|
[08:09] <G__81> i hope i have a good time with this team learning and exploring new things :) |
|
[08:09] <slangasek> G__81: I don't know how things are in Fedora, but around here I find that there's a pretty big gulf between kernel folks and userspace folks; so while we're happy to have community kernel developers get involved, you're not likely to find much practical assistance for kernel matters on this channel. :) |
|
[08:10] <slangasek> ("pretty big gulf" -- I think once someone touches the kernel it sucks up all their attention until there's no room left in their brain for userspace) |
|
[08:10] <G__81> slangasek, yeah sure i ll log in to those and apart from that i also like to contributing on the user space |
|
[08:10] <G__81> it would be more the userspace than on the kernel space |
|
[08:11] <NCommander> slangasek, so the fact that I started working to fix the lpia kernel has destroyed my userspace maintenance abilities? |
|
[08:11] <wgrant> NCommander: Yes. |
|
[08:11] <NCommander> slangasek & wgrant: you should have warned me! |
|
[08:11] <G__81> be right back |
|
[08:11] <slangasek> it's a gradual process |
|
[08:11] <wgrant> slangasek: But it is now inevitable. |
|
[08:12] <NCommander> ;.; |
|
[08:13] * NCommander wonders if his Hurd coding experience will save him |
|
[08:13] <slangasek> has Hurd coding experience ever saved anyone? |
|
[08:14] <NCommander> no, but I did port Linux's entropy framework to mach |
|
[08:16] <eddyMul> fabrice_sp: your advice to use postinst worked! thanx! will post debdiff soon.... |
|
[08:19] <YokoZar> slangasek: reminds me of a discussion I had at Wineconf. Alexandre Julliard (lead dev) described his main intrest in Wine as "anything low level. The lower level the better." I spoke later, introducing myself as "The opposite of Julliard" |
|
[08:20] <slangasek> heh |
|
[08:20] <slangasek> I wonder if any of my code in wine is still extant |
|
[08:22] <fabrice_sp> eddyMul: You're welcome :-) |
|
[08:23] <NCommander> YokoZar, lol |
|
[10:09] <didrocks> slangasek: hi :) |
|
=== eagles is now known as eagles0513875 |
|
[10:39] <Hobbsee> hmm. |
|
[10:40] <wgrant> I disagree. |
|
[10:40] <Hobbsee> wgrant: ? |
|
=== asac__ is now known as asac |
|
[10:57] <DktrKranz> persia, I did some tests on -rt flavour, everything is fine except shutdown. I rebooted/halted system three times and it hangs |
|
[10:59] <persia> DktrKranz, Hrm. OK. I'll bug abogani about that. I was planning to upload today (or maybe tomorrow). Shall I wait to see if we can patch that, or just go ahead? |
|
[11:00] <DktrKranz> persia, well... it could be just me, I have old hardware and I can't be sure it's related to that (such as my pulseaudio issues) |
|
[11:00] <DktrKranz> pushing it can be a good improvement |
|
[11:00] <DktrKranz> eventually, it can be fixed at a later stage (via SRU too) |
|
[11:01] <persia> DktrKranz, Could you update the bug with your encouragement? Although I was actioned with it in the last release meeting, I would like to have someone in MOTU Release stamp the bug for the records :) |
|
[11:01] <DktrKranz> sure |
|
[11:01] <persia> And yeah, the current linux-rt FTBFS, which isn't ideal for release. |
|
[11:02] <persia> Thanks. |
|
[11:05] <DktrKranz> persia, done. If you want, I can look for debugging messages while shutting down to better understand my failures |
|
[11:05] <persia> DktrKranz, If you could, and file a different bug, I'll make sure abogani sees it. |
|
[11:07] <DktrKranz> agreed. I'll do additional tests (since it will probably become my default kernel). If I catch alessio around, I'll ping him, he's italian too and we recently have a chat about -rt flavour |
|
[11:08] <persia> Cool. He was complaining about not having enough users or feedback, so that's probably a good thing. |
|
[11:09] <DktrKranz> I'm definitely a huge fan of -rt. I contributed a little in RT_PREEMPT development some years ago |
|
[11:09] <DktrKranz> just some minor fixes, though. |
|
[11:10] <DktrKranz> anyone familiar with KDE4 has an opinion about http://hattory.no-ip.info/intrepid/result/boson_0.13-4ubuntu1/boson_0.13-4ubuntu1.buildlog ? |
|
[11:10] <DktrKranz> s/opinion/fix/ ;) |
|
=== Adri2000_ is now known as Adri2000 |
|
[11:43] <Mathiasdm2> hi |
|
[11:44] <Mathiasdm2> I'm looking to include a deb package for I2P in Ubuntu. It's a Java program that currently installs all of its files in a single directory. |
|
[11:44] <Mathiasdm2> To meet the requirements for including I2P, I understand that executables must be in /usr/bin, config files in /etc/i2p, and so on. |
|
[11:44] <Mathiasdm2> I was wondering: would it be allowed to (during install) create an I2P directory, and add symlinks to these other locations? |
|
[11:44] <Mathiasdm2> And if it is allowed, where would one have to install I2P? In /opt, like openoffice.org? |
|
[11:45] <StevenK> Openoffice.org is not installed in /opt |
|
[11:46] <Mathiasdm2> ah, my mistake, then it's version 3 that I downloaded myself :p |
|
[11:49] <Mathiasdm2> is the procedure I described a good one? or would it not be allowed? |
|
[11:49] <RainCT> Mathiasdm2: no |
|
[11:49] <Mathiasdm2> any alternatives then? I can't find a lot of documentation on packaging java apps :( |
|
[11:50] <RainCT> Mathiasdm2: Debian packages can't place anything into /opt (afaik), and if you've a tarball meant to be placed into /opt then that's probably a pre-compiled one anyway, which isn't really ideal to be packaged |
|
[11:51] <RainCT> Mathiasdm2: I don't know Java, but perhaps looking at some existing package may help |
|
[11:51] <Mathiasdm2> well, I have sources available, along with an ant script that builds everything, but the problem is that i2p seems pretty much made to be put in a single directory (self-contained) |
|
[11:52] <Mathiasdm2> okay, I'll do that |
|
[11:52] <RainCT> (technically it's possible to place everything into /usr/share/<pkgname> and symlink it from there, but that's hardly the right way to do it) |
|
[11:52] <Mathiasdm2> yeah, I feared so |
|
[12:11] <asac> NCommander: pong |
|
[12:50] <didrocks> hum, I can't figure out what to do for localisation. I added an extra dialog in nautilus-share and wanted to update the according translation (po file), before rosetta handle it. Are there some good documentation for that? |
|
[14:19] <ion_> LP: #283343 – could someone please do the sync? |
|
[14:20] <ion_> No linky? https://bugs.launchpad.net/ubuntu/+source/gnokii/+bug/283343 |
|
[14:20] <ubottu> Launchpad bug 283343 in gnokii "Please sync gnokii 0.6.26.dfsg-3 (universe) from Debian unstable (main)." [Wishlist,Confirmed] |
|
[14:20] <ion_> The change in -3 fixes an encoding bug. |
|
[14:33] <persia> ion_, That bugs looks all set, and just needs an archive-admin. The archive-admins are scheduled to do stuff on Mondays, Tuesdays, Wednesdays, and Fridays usually. |
|
[14:34] <ion_> Alright, thanks. |
|
[14:35] <persia> ion_, also, for links, you want to use a syntax like "bug #123456" |
|
[14:35] <ubottu> Launchpad bug 123456 in xine-lib "podcast crashes amarok" [Undecided,Fix released] https://launchpad.net/bugs/123456 |
|
[14:36] <ion_> I’ll keep that in mind. :-) |
|
[14:43] <slytherin> superm1: hi |
|
[14:44] <LimCore> why are you people so negative? O_o |
|
[14:44] <LimCore> ok lets try this one: |
|
[14:44] <LimCore> would it be very hard to pack a text file, that would contain a fingerprint of medibuntu's key? |
|
[14:46] <LimCore> that would close a small (but existing) security hole: when installing medibuntu, there is no way to know if the medibuntu's keyring is actually the real signing key (or was it i.e. injected to our network connection etc etc). To fix it, the medibuntu keyring should be in officiall ubuntu's repo, signed by it; Or just put/sign in ubuntu.org information what should be the fingerprint of medibuntu;s |
|
[14:47] <LimCore> just packing medibuntu's keyring in ubuntu would fix it. |
|
[14:47] <LimCore> or text file (but then - translations) |
|
=== kop__ is now known as k0p |
|
[15:07] <AnAnt> DktrKranz: Hello, regarding bug 269855, I prepared a new upstream package, should I make a debdiff too ? |
|
[15:07] <ubottu> Launchpad bug 269855 in sl-modem "Please apply patch to fix issue with Si3054 modems" [Medium,New] https://launchpad.net/bugs/269855 |
|
[15:08] <AnAnt> superm1: can you help me with dkms problem ? |
|
[15:08] <DktrKranz> AnAnt, at this point, you may want to ask for a freeze exception, given changes provided |
|
[15:08] <AnAnt> DktrKranz: I thought that it will wait for jaunty, what do you think ? |
|
[15:09] <AnAnt> DktrKranz: I haven't got much feedback from ppl who issued bugs at both Debian & Ubuntu |
|
[15:09] <AnAnt> DktrKranz: I got feedback from only 1 person ! |
|
[15:10] <DktrKranz> AnAnt, if this change is important enough, it could be inserted in intrepid too, but jaunty option is good |
|
[15:10] <AnAnt> DktrKranz: well, it closes many bugs in Debian & Ubuntu |
|
[15:10] <AnAnt> DktrKranz: I am working at adding dkms support too |
|
[15:12] <ScottK> Hobbsee or StevenK: kio-sword can die now. |
|
[15:12] <DktrKranz> AnAnt, such a rewrite could be hardly approved for intrepid, though |
|
[15:13] <AnAnt> DktrKranz: I agree ! |
|
[15:14] <AnAnt> DktrKranz: the changelog is pretty BIG ! |
|
[15:14] <DktrKranz> I see :) |
|
[15:14] <AnAnt> DktrKranz: actually the updated one is on mentors.debian.net |
|
[15:14] <DktrKranz> are you maint on debian too? |
|
[15:14] <AnAnt> DktrKranz: well, I have a couple of packages there |
|
[15:15] <AnAnt> DktrKranz: and I found that sl-modem was orphaned, and mom's laptop needs it (it had that Si5043 chip), so I decided to take over ! |
|
[15:15] <DktrKranz> good! |
|
[15:19] <AnAnt> Si3054 that is |
|
[15:25] <AnAnt> DktrKranz: I got a question, bug 144090 has debian bug 489289 in its bug tracker, what does that mean ? |
|
[15:25] <ubottu> Launchpad bug 144090 in sl-modem "update the package to 2.9.10 or 2.9.11" [Undecided,In progress] https://launchpad.net/bugs/144090 |
|
[15:25] <ubottu> Debian bug 489289 in sl-modem-daemon "slmodem-2.9.11-20080629 available; all previous patches merged." [Normal,Open] http://bugs.debian.org/489289 |
|
[15:27] <DktrKranz> AnAnt, just a remote bug watch, to keep track of progress in debian |
|
[15:27] <AnAnt> DktrKranz: does that mean that fixing 489289 in debian will fix 144090 in LP ? |
|
[15:28] <DktrKranz> yes, if package is in sync and we're still before Debian Import Freeze |
|
[15:28] <DktrKranz> but those conditions are actually not satisfied |
|
[15:55] <tseliot> \sh, DktrKranz: can you have a look at my SRU, please (the debdiff is in comment 32)? https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-177/+bug/261816 |
|
[15:55] <ubottu> Launchpad bug 261816 in linux-restricted-modules-envy-2.6.24 "nvidia: Multiple versions in DKMS" [Medium,In progress] |
|
[16:01] <slytherin> StevenK: I am not sure if there is a bug for it already, but wasn't gnome-bluetooth supposed to be removed from archives? |
|
[16:02] <persia> slytherin, You can't do that until you port gnome-phone-manager to use a modern interface for bluetooth. |
|
[16:03] <slytherin> persia: I don't think gnome-phone-manager has anything to do with gnome-bluetooth. gnome-bluetooth is just old implementation of obex server and client |
|
[16:05] <persia> slytherin, gnome-phone-manager build-depends on libgnomebt0-dev |
|
[16:05] * slytherin chcks |
|
[16:05] <persia> slytherin, Note that gnome-phone-manager probably doesn't work over bluetooth (libgnokii doesn't support any of my phones, so I can't test). Porting would be good. |
|
[16:06] <slytherin> persia: At some point of time gnome-phone-manager used to work with bluetooth. Haven't tested for last 2 releases. |
|
[16:08] <slytherin> persia: One more question I have. The option for enabling file receiving is not available in bluez-gnome anymore. And gnome-user-share, which can enable this option, lies in universe. Do you think I should patch obex-data-server to have this option set by default, for bonded devices? |
|
[16:08] <persia> slytherin, RIght. It used to work. I expect that it broke in hardy. |
|
[16:09] <persia> slytherin, No. You'd first want to patch gnome-user-share to use a simpler webserver which might be able to be included on the CD, and it's far too late in the intrepid cycle to investigate these sorts of things. For Jaunty, I think that's a good idea. |
|
[16:11] <slytherin> persia: something like lighthttbd? |
|
[16:11] <slytherin> lighttpd I mean |
|
[16:13] <persia> slytherin, No. That's a whole megabyte installed. Think smaller, like dhttpd or micro-httpd. |
|
[16:13] <slytherin> hmm |
|
[16:14] <persia> space on the CD is precious. |
|
[16:14] <slytherin> but what is solution for now then? I have seen many bugs asking it is not possible to accept files. |
|
[16:15] <persia> slytherin, My memory is that installing gnome-user-share makes it just work, right? |
|
[16:15] <slytherin> persia: yes but is is in universe. |
|
[16:15] <slytherin> and also it will pull apache2 |
|
[16:16] <persia> slytherin, Yes, but that's just documentation. The archive is frozen. It's too late to try to change and test the changes to make it work. |
|
[16:16] <slytherin> oh, so frozen as in no changes at all? |
|
[16:16] <persia> The bugs should get duped, the workaround (install gnome-user-share) should be updated in the description of the master bug, and it should get fixed first thing in Jaunty. |
|
[16:16] <persia> frozen as in only changes that are critical for release. |
|
[16:17] <slytherin> damn, I should have thought about this before. |
|
[16:17] <persia> General rule is crashers, regressions, translations, install failures, that sort of thing. |
|
[16:17] <slytherin> regression as compared to hardy? |
|
[16:17] <persia> crevette and I discussed it a couple weeks ago, and crevette was planning to talk to upstream about switching to a different web server. |
|
[16:18] <persia> Yes. regression as compared to hardy, but the level of change can't be high : some regressions we have to accept (unfortunately). |
|
[16:19] <slytherin> hmm |
|
[16:21] <persia> slytherin, Basically, it's that end-users who want to do obex have to install an extra package in universe, right? |
|
[16:21] <slytherin> the reason this is regression is that the option moved form bluez-gnome to gnome-user-share. And can be solved by patching o-d-s to have option set by default. But I guess no point in putting the effort if it is not going to make in. |
|
[16:22] <persia> Wait : does one need to both install gnome-user-share, and change an option in obex-data-server? |
|
[16:22] <slytherin> Don't think so. I haven't really looked into the code. But it should be possible to just patch obex-data-server. |
|
[16:24] <persia> slytherin, Basically, I don't think we want to ask for gnome-user-share on the CDs for intrepid. If o-d-s needs a patch to make it so that users can just install gnome-user-share and it works for them, then it makes sense to apply that patch. If it works by installing the package without a patch, there's no need for the patch. |
|
[16:26] <slytherin> persia: I guess o-d-s will just need a patch. I will check. I am not proposing gnome-user-share for CD anyway, not this late. |
|
[16:27] <persia> slytherin, OK, I'm just worred about changing o-d-s in such a way that it would need to recommend or depend on g-u-s. As long as it works properly without g-u-s, and works better with g-u-s, then it's a good patch to push for intrepid. |
|
[16:27] <slytherin> Ok. |
|
[16:46] <slytherin> persia: looks like the preference is actually part of bluez-gnome gconf option set. It is just not available on UI. |
|
[16:47] <persia> slytherin, So basically it needs two things done : to set the gconf key, and to install gnome-user-share? |
|
[16:47] <slytherin> persia: mostly only first thing. |
|
[16:48] <persia> So setting that, even with gnome-user-share not installed fixes it? |
|
[16:48] <slytherin> It should. I am checking. My dongle is gicing me trouble again. |
|
[16:51] * ScottK notes some irony about the name of this spacecraft: http://www.cnn.com/2008/TECH/space/10/17/solar.mission.ap/index.html?eref=rss_topstories |
|
[16:52] <jdong> "Ibex, the size of a bus tire," *grin* |
|
[17:00] <slytherin> persia: can you simply try if changing value of gconf key apps/bluetooth-manager/receive_enabled enables/disables file sharing? I will provide the patched package in PPA so that someone else can test it. |
|
[17:06] <lfaraone> Any release team members about? |
|
[17:06] <ScottK> Maybe |
|
[17:07] <lfaraone> ScottK: would bug 274820 cover the addition of a sugar-jhbuild package that would allow developers to build devel versions of the sugar project from SVN? |
|
[17:07] <ubottu> Launchpad bug 274820 in sugar "Sugar feature freeze exception" [Wishlist,Fix released] https://launchpad.net/bugs/274820 |
|
=== fta_ is now known as fta |
|
[17:08] <ScottK> lfaraone: Why wasn't it uploaded before? |
|
[17:08] <lfaraone> ScottK: It didn't exist. |
|
[17:08] <lfaraone> ScottK: It's a new package. |
|
[17:09] <ScottK> lfaraone: I'd say no. Get it uploaded as soon as Ibex opens and then we can backport it. |
|
[17:10] <lfaraone> ScottK: Ah, OK. |
|
[17:11] <slytherin> ScottK: you mean 'jaunty' :-) |
|
[17:18] <DktrKranz> tseliot, I think I'm late for your SRU ;) |
|
[17:22] <tseliot> DktrKranz: thanks anyway ;) |
|
[18:17] <LaserJock> does anybody know what the difference between flashplugin-nonfree and adobe-flashplugin is? |
|
[18:21] <persia> LaserJock, Unless I'm mistaken, flashplugin-nonfree is a multiverse package that downloads flash from adobe and installs it, and adobe-flashplugin is a partner package that installs adobe flash. |
|
[18:22] <AnAnt> who can help me with a dkms issue ? |
|
[18:23] <AnAnt> I added dkms support to slmodem, my problem is that it only compiles one module but fails to compile the other |
|
[18:24] <AnAnt> hey ! sl-modem just got uploaded to Debian ! |
|
[18:28] <LaserJock> persia: ok .... but still that seems sort of like a duplication on a rather annoying package :-) |
|
[18:28] <ScottK> slytherin: yes |
|
[18:28] <LaserJock> persia: do you imagine we'd ever drop our Multiverse package and let Canonical maintain Flash? :-) |
|
[18:28] <ScottK> I'd vote yes. |
|
[18:29] <AnAnt> gnash is alright |
|
[18:29] <AnAnt> well, it didn't work well on youtube when I was behind a proxy though |
|
[18:30] <persia> LaserJock, We could, but derivatives might complain. |
|
[18:30] <LaserJock> hmm, that is a point |
|
[18:30] <ScottK> That'd be a reason not to drop any package ever. |
|
[18:31] <AnAnt> can't derivatives start using gnash ? |
|
[18:31] <persia> No it wouldn't. There's a difference between e.g. dropping the useless gnutls13 and dropping a potentially useful package. |
|
[18:31] <AnAnt> I'm that the last one to say this, I am working on a package that still doesn't work well with gnash ! |
|
[18:32] <persia> Mind you, I'd not be opposed to altering debian/control so flashplugin-nonfree couldn't build on Ubuntu |
|
[18:39] <ScottK> persia: Is there any reason derivatives can't use Partner? |
|
[18:39] <persia> ScottK, licensing. |
|
[18:40] <persia> Stuff in partner is typically only OK for Canonical to distribute, and may not work with a given set of patches applied by a derivative. |
|
[18:40] <ScottK> They can't distribute stuff that's in there, but no reason that they can't have it in their user's sources.list |
|
[18:40] <persia> (thinking of derivatives like MEPIS, Mint, Arch : not flavours) |
|
[18:40] <ScottK> Yes |
|
[18:40] <ScottK> Gotta run |
|
[18:40] <persia> No guarantee it even works though. Same reason people don't use Debian Multimedia as a deb source for Ubuntu. |
|
[18:41] <G__81> Hi people let me introduce myself. I have been a fedora user and contributor and have installed Ubuntu 8.04 for the first time and i really like it and i am really interested to getting involved in contributing to Ubuntu |
|
[18:41] <G__81> It would be really great if some one could suggest me with the road ahead. |
|
[18:42] <G__81> I am really looking forward for this. I can help in Bug Fixing, Testing Packages, Documentation and of course as an ambassador too |
|
[18:42] <G__81> I work on the Linux Kernel as a hobby so i could chip in some help there |
|
[18:43] <G__81> i was really amazed on seeing Launch Pad and how it recognizes all the contributions made by the developers. It really motivates the developers to attain more points |
|
[18:44] <G__81> is anyone here ? |
|
[18:44] * G__81 waits |
|
[18:44] <csilk> yeah |
|
[18:44] <csilk> I am |
|
[18:44] <G__81> hi csilk nice to meet you |
|
[18:44] <csilk> likewise |
|
[18:45] <G__81> i am from india and want to be part of this team :) |
|
[18:45] <csilk> Fantastic |
|
[18:45] <AnAnt> I am |
|
[18:45] <G__81> Hi AnAnt |
|
[18:46] <G__81> I have been contributing to fedora in terms of bug fixing, testing packages, documentation team and also as an ambassador for fedora |
|
[18:47] <AnAnt> well, my contributions aren't much, I just do some simple packaging and bug reporting ! |
|
[18:47] <csilk> G__81, feel free to read __ http://www.ubuntu.com/community/processes/newdev |
|
[18:48] <G__81> csilk, thanks and i did read through it as i was so enthusiastic :) |
|
[18:52] <AnAnt> is there still a need for sun-java6 now that openjdk is there ? |
|
[18:53] <jdong> I don't believe OpenJDK6 is 100% compatible with sun-java6 |
|
[18:53] <jdong> though it's extremely close |
|
[18:53] <csilk> Was just about to say that |
|
[18:53] <persia> It's not a goal to make them 100% compatible, as I understand things. |
|
[18:54] <csilk> until there is 100% compatibilty there will be a need for the sun one |
|
[18:54] <jdong> persia: correct |
|
[18:54] <persia> We still also have JDKs for 1.4 and 1.5. |
|
[18:54] <jdong> csilk: even if they are 100% interchangeable that doesn't mean we should drop the Sun ones... |
|
[18:54] <persia> If everything in Ubuntu can be migrated to OpenJDK, the others can be dropped. |
|
[18:54] <G__81> csilk, is the wiki the right place to start with ? |
|
[18:54] <G__81> i have created a LP account |
|
[18:54] <G__81> or rather i already have one : |
|
[18:54] <G__81> :) |
|
[18:55] <jdong> persia: our code can be migrated but what about stuff the user intends to run? As a platform IMO we should continue to offer Sun's supported VMs |
|
[18:55] <G__81> is there any mailing list that i ve to subscribe to ? |
|
[18:55] <G__81> so that i introduce myself and start off. |
|
[18:55] <csilk> jdong, i only use the sun one because I can't risk incompatibility, whatever code I wright at home needs to work in my University lab |
|
[18:55] <AnAnt> G__81: ubuntu-dev I think |
|
[18:55] <persia> jdong, Actually, at this point, our code can't be migrated. Anyway, OpenJDK meets Sun Certification tests, and we should be able to phase out the old ones. |
|
[18:55] <csilk> *write |
|
[18:56] <persia> Sun probably wants us to drop them more that we want to drop them anyway. |
|
[18:56] <handschuh> hi, can anyoen help me with debhelper commands? |
|
[18:57] <jdong> persia: cool |
|
[18:59] <handschuh> I am trying to change che permissions on a dir that was created with dh_installdirs |
|
[19:00] <G__81> is there a mailing list for Ubuntu-motu ? |
|
[19:00] <AnAnt> G__81: yeah, |
|
[19:00] <iulian> AnAnt, G__81: Actually no, ubuntu-devel is moderated, only developers can post to that list. You might want to subscribe to ubuntu-motu@l.u.c. OTOH, there are hundreds of bugs that are waiting to be fixed, just pick one, or maybe more and fix them. |
|
[19:01] <G__81> iulian, ok so its ubuntu-motu and there is an other thing called ubuntu-motu-mentors right ? |
|
[19:01] <G__81> whats that for ? |
|
[19:01] <persia> I'd recommend subscription to all of ubuntu-devel-announce@, ubuntu-devel@, ubuntu-devel-discuss@, and ubuntu-motu@ : each has a slightly different role, but all are useful. |
|
[19:01] <AnAnt> yeah, I really hope someone look at a couple of bugs I reported against the kernel |
|
[19:02] <iulian> G__81: I think there is a wiki about the mentoring programme. One sec. |
|
[19:02] <iulian> G__81: https://wiki.ubuntu.com/MOTU/Mentoring/Junior_Contributor |
|
[19:04] <G__81> iulian, thanks a ton |
|
[19:04] <G__81> iulian, i would like to start off and would read the links that you gave me and will subscribe to the mailing lists first and introduce myself and would start off from there |
|
[19:05] <G__81> i am really looking forward for working with this team and hopefully do something really good for the Ubuntu Project |
|
[19:05] <iulian> That would be excellent, indeed. |
|
[19:06] <G__81> yeah thanks iulian have a nice day will log in back here once i go through the docs |
|
[19:07] <G__81> cya all |
|
[19:07] <iulian> You too. |
|
[19:07] <G__81> iulian, have one question |
|
[19:08] <G__81> iulian, i gave this sudo apt-get install ubuntu-restricted-extras |
|
[19:08] <G__81> and now its asking me for some fonts and java installation.This wouldnt break anything right ? |
|
[19:09] <RainCT> G__81: no, or at least it shouldn't :) |
|
[19:09] <iulian> G__81: No, it's safe. I didn't encounter any problems with it. |
|
[19:10] <G__81> its installing some fonts and some exe files |
|
[19:10] <AnAnt> yup, it is safe |
|
[19:10] <G__81> Extracting cabinet: verdan32.exe |
|
[19:10] <AnAnt> it will extract some fonts from the exe files |
|
[19:10] <G__81> :) |
|
[19:10] <AnAnt> that's msttcorefonts package |
|
[19:10] <RainCT> G__81: yep, it's installing windows' fonts |
|
[19:10] <G__81> oh |
|
[19:10] <G__81> ok thanks |
|
[19:11] <G__81> RainCT, Sorry to ask these questions as i have been using fedora :) |
|
[19:11] <G__81> using as well as contributing so am completely new to ubuntu process as such :) |
|
[19:13] <G__81> ok got to go now and be back after a few hours with questions/doubts :) |
|
[19:13] <G__81> nice talking to you all |
|
[19:13] <G__81> have a nice day |
|
[19:13] <AnAnt> G__81: questions about usage are better asked in #ubuntu |
|
[19:14] <G__81> AnAnt, yeah just found that |
|
[19:14] <RainCT> G__81: no problem, it's great that you want to get involved :) |
|
[19:14] <AnAnt> indeed |
|
[19:14] <G__81> RainCT, infact have already initiated the download of Ibex to give some testing reports |
|
[19:15] <G__81> Does Ubuntu work fine with Qemu ? |
|
[19:15] <G__81> I used to test Fedora using that so i guess Ubuntu should also work right ? |
|
[19:15] <G__81> i believe the installer shouldnt have problems |
|
[19:15] <G__81> anyways let me explore on it |
|
[19:15] <G__81> if it does not, then my work starts right there :) |
|
[19:15] <DktrKranz> AnAnt, I just noticed sl-modem has been accepted in sid, good news ;) |
|
[19:16] <AnAnt> G__81: ok, if you got issues regarding the upcoming release of ubuntu (currently Ibex), those can be discussed in #ubuntu+1 |
|
[19:16] <G__81> Oh thanks for that info AnAnt |
|
[19:16] <AnAnt> DktrKranz: yeah, a guy from debian sent me an email asking if I need sponsoring, so I said yes, at the same I was chatting with you earlier today |
|
[19:17] <AnAnt> DktrKranz: now I need help with adding dkms support ! |
|
[19:17] <AnAnt> dkms ? anyone ?! |
|
[19:17] <jdong> G__81: I have installed Ubuntu in both qemu and kvm (which is more or less modified faster qemu) |
|
[19:17] <DktrKranz> I'm not expert of such a system, sorry |
|
[19:18] <G__81> jdong, thanks will do it |
|
[19:19] <AnAnt> DktrKranz: shall I submit a sync request for jaunty now ? or shall I wait ? |
|
[19:20] <DktrKranz> AnAnt, you have to wait a little bit, it won't get processed until archives are reopened |
|
[19:21] <jdong> calc: is there any hope of reviving the side-by-side installation packaging for openoffice? |
|
[19:21] <calc> jdong: i kept running into more problems as i fixed them and rene (debian) said he doesn't think it would work at all |
|
[19:22] <calc> jdong: i'm sure with enough work it could be made to work eventually, but it hasn't been used in a long time apparently |
|
[19:22] <jdong> calc: yeah it hasn't been used since the 1.x->2.0 transition |
|
[19:22] <jdong> calc: I was just thinking from a backports perspective if it'd be more advantageous |
|
[19:23] <jdong> I'm already getting a lot of clammering about OOo3 for Hardy and that scares me without side by side installation |
|
[19:23] <calc> oh yea we were originally going to put it into universe as openoffice.org3 but it looked like too much trouble |
|
[19:23] <jdong> yeah; that's too bad :(. It would be the best compromise for people who want version 3 |
|
[19:23] <calc> jdong: i'm considering trying to get 2.4.2 into both hardy/intrepid when it comes out in early dec for bug fix reasons |
|
[19:24] <calc> jdong: but that would go into updates, not backports |
|
[19:24] <jdong> right, 2.4.2 sounds good for -updates |
|
[19:24] <jdong> for backports we'll really have to figure out what we want to do.... |
|
[19:25] <calc> also OOo 3.0 uses a lot of stuff from universe, not sure if that affects how it is backported |
|
[19:25] <jdong> newer stuff than what hardy's universe has? |
|
[19:25] <jdong> backports can build against all pockets |
|
[19:26] <calc> jdong: oh ok, maybe newer than hardy's but not newer than intrepid |
|
[19:26] <calc> jdong: i didn't know if it could pull from all pockets |
|
[19:26] <calc> i haven't tried building 3.0 against hardy |
|
[19:27] <AnAnt> is OO3 stable ? |
|
[19:27] <calc> AnAnt: upstream calls it that |
|
[19:27] <jdong> AnAnt: FWIW I've been using it on OS X without any issues |
|
[19:27] <calc> AnAnt: i wouldn't call it that until at least 3.0.1 is released |
|
[19:27] <jdong> but I wouldn't use that as a metric for stable |
|
[19:28] <calc> they were finding new serious bugs until a week before release |
|
[19:28] <jdong> lovely |
|
[19:28] <calc> it appears when they didn't find any new serious bugs for one week they released it (i'm pretty sure that is really how it works) |
|
[19:28] <jdong> lol |
|
[19:28] <calc> with 2.4.0 they found major bugs shortly after they released it |
|
[19:29] <calc> of course that is also why they release a .1 release a month later, heh |
|
[19:29] <jdong> calc: does our 2.4.x support the office 2007 formats or is that a 3.0-only feature? |
|
[19:29] <jdong> I think that's what most people are clamoring about. |
|
[19:29] <calc> jdong: even our 2.3 does |
|
[19:29] <jdong> calc: wow, cool |
|
[19:29] <calc> not sure about 2.2 though |
|
[19:30] <calc> it might not be as good as the 3.0 support, well hopefully they are fixing bugs relating to it, but it definitely can open the files |
|
[19:30] <calc> official openoffice didn't add support to open the files until 3.0 but it has been in ooo-build for a long time now |
|
[19:31] <calc> i also added support for some more Office 2007 mimetypes with the last 2.4.1 upload to intrepid |
|
[19:31] <jdong> cool |
|
[19:31] <calc> a few of the types weren't being identified by shared-mime-info |
|
[19:32] <calc> so now those can be double click opened as well, you could always open them from inside of openoffice though |
|
[20:07] <LaserJock> calc: if OO.o can open Office 2007 .docx files with equations they'd be ahead of Microsoft |
|
[20:08] <LaserJock> I just got my mac upgraded to Office 2008 for Mac so that I could read my students files and it won't open equations from 2007 |
|
[20:10] <LaserJock> I guess it shouldn't surprise me anymore but I was really shocked that MS would do that |
|
[20:10] <calc> heh, don't know about equations |
|
[20:11] <LaserJock> I should get OO.o 3 on it and compare |
|
[20:15] <laga> jdong: hey. just for the record: your upstart event.d files don't work in intrepid :) |
|
[20:27] <james_w> is writing to files from multiple threads concurrently going to cause segfaults? |
|
[20:27] <rockstar> james_w, it could. |
|
[20:27] <james_w> hey rockstar |
|
[20:27] <rockstar> james_w, hi! |
|
[20:32] <jdong> laga: yeah, I'm not surprised |
|
[20:33] <jdong> laga: never updated them for intrepid's new upstart format |
|
[20:33] <laga> jdong: i'm going to tinker with $stuff |
|
[20:33] <laga> oh, there is a new format. |
|
[20:33] <jdong> laga: sweet, lemme know if you make progress :) |
|
[20:33] <laga> yay for stable formats :) |
|
[20:34] <jdong> laga: well it was well announced that it was going to change :) |
|
[20:34] <laga> jdong: i probably won't convert everything, i just want a working mythbuntu ;) |
|
[20:34] <laga> jdong: just trolling :) |
|
[20:34] <jdong> lol |
|
[20:36] <laga> jdong: i assume the documentation on upstart.ubuntu,.com is current |
|
[20:37] <persia> laga, It's not : that's for a newer version than we have in intrepid. |
|
[20:37] <jdong> :) |
|
[20:38] <LaserJock> yay for stable documentation |
|
[20:38] <persia> Well, the reason we didn't update upstart for intrepid is that it's undergoing massive change upstream, and isn't currently as stable as we want. Should be able to get in sync again for Jaunty. |
|
[20:42] <laga> jdong: where was the format change announced? |
|
[20:42] <laga> hardy and intrepid seem to have 0.3.9 |
|
[20:43] <rockstar> james_w, are you going to be in London this week? |
|
[20:43] <jdong> laga: oh, maybe I am wrong about that then |
|
[20:43] <james_w> rockstar: I may be swinging by this week or next to meet with you guys. Nothing confirmed yet though. |
|
[20:44] <james_w> rockstar: blame jml :-) |
|
[20:44] <jdong> laga: you might want to start from http://jdong.mit.edu/~jdong/macbook/etc/event.d/ that event.d |
|
[20:44] <rockstar> james_w, great. |
|
[20:44] <jdong> laga: I think one of the LP branches has a poorly machine-generated upstart :) |
|
[20:46] <laga> yay! the upstart package doesn't come with a description of the event files. |
|
[20:46] <jdong> laga: might want to talk to #upstart |
|
[20:46] <jdong> it's been a while since I've used event.d syntax |
|
[20:58] <LaserJock> anybody here try irssi-plugin-xmpp by any chance? |
|
[20:58] <LaserJock> I wonder how it is relative to bitlbee |
|
[21:54] <james_w> could someone confirm I'm not being stupid please? |
|
[21:54] <james_w> 0x0 in a stacktrace is NULL, yes? |
|
[21:56] <james_w> and calling var->something when var=0x0 in the stacktrace will cause a SIGSEGV? |
|
[21:57] <coppro> yes |
|
[21:57] <persia> james_w, Can you show the stacktrace? |
|
[21:57] <persia> 0x0 is NULL. |
|
[21:57] <james_w> thought so |
|
[21:57] <persia> calling 0x0->something may or may not segfault, depending on how it's used. |
|
[21:58] <persia> (and how it's trapped). |
|
[21:58] <james_w> ah |
|
[21:58] <james_w> http://launchpadlibrarian.net/17571430/Stacktrace.txt |
|
[21:58] <james_w> file_monitor_remove_watch (monitor=0x9240d60, watch=0x0) |
|
[21:58] <james_w> but one line before that is called it does watch->something |
|
[21:58] <persia> What's the code for that line? |
|
[21:59] <persia> Just file_monitor_remove_watch(monitor, watch); ? |
|
[22:01] <james_w> if (g_slist_length (notify->watch->notifies) == 0) { |
|
[22:01] <james_w> file_monitor_remove_watch (monitor, notify->watch); |
|
[22:01] <james_w> I think I may have it though |
|
[22:09] <persia> Yeah, that's an example of how you can use 0x0->something safely :) |
|
[22:09] <persia> Of course, it returns zero (0x0), which breaks that particular test. |
|
[22:09] <persia> (note that g_slist_length takes a pointer as an argument) |
|
[22:10] <james_w> yeah, notifies is a pointer |
|
[22:10] <james_w> how is that a safe use? |
|
[22:10] <james_w> or rather, what makes that safe? |
|
[22:10] <persia> 0x0 is guaranteed to contain 0x0 |
|
[22:10] <persia> So you get notify->0x0->0x0 which contains 0x0 |
|
[22:11] <james_w> ah, ok |
|
[22:11] <persia> Essentially, you can always read a non-complex value (integer, byte, char, size_t, etc.) from 0x0, and you'll get 0x0. |
|
[22:12] <persia> And actually, it might get messy : if you have a structure, and the head of the structure is at 0x0, you might end up pulling a value from somewhere like 0x8 |
|
[22:13] <persia> (which may not be zero) |
|
[22:17] <james_w> what would cause a read from /dev/tty9? |
|
[22:17] <james_w> assuming /dev/tty9 is my X |
|
[22:18] <persia> Well, either a rogue, or the server reading from /dev/tty |
|
[22:18] <persia> clients shouldn't be able to access it. |
|
[22:18] <persia> Note that "the server" doesn't necessarily mean xserver-xorg : it's more complicated now, and lots of things live on the server-side. |
|
[22:18] <james_w> but it would just be normal server activity? |
|
[22:19] <persia> It's unlikely server activity. xinput should do most of that processing. Ask on -devel or -x (I'd try -x first, but don't know how active it is at this time of week) |
|
[22:20] <james_w> pretty active right now |
|
[22:21] <persia> They can probably suggest what might be reading from there. A call to /dev/tty client side would get a virtual tty reference. |
|
[22:21] <persia> (e.g. /dev/pts/0) |
|
[22:23] <james_w> I think it's trying to track activity so that it can work out if you are idle |
|
[22:23] <james_w> so input device activity would fit with that |
|
[22:26] <persia> Right, except it shouldn't be getting that from /dev/tty : it should be using XInput2 |
|
[23:55] <james_w> and what do you know? Debugging statements make it work. |
|
[23:55] <wgrant> james_w: All good bugs are like that. |
|
[23:55] <james_w> this one is a peach |
|
[23:57] <persia> james_w, This is your null-pointer bug? Which debugging statement made it go away? |
|
[23:58] <james_w> a trick I got of Federico |
|
[23:58] * persia suspects that it ought do something other than have a debug statement |
|
[23:58] <james_w> access("MARK: Debugging string", F_OK) |
|
[23:58] <james_w> then you can just strace and grep the result for "MARK" |
|
[23:59] <james_w> I expect it just closes the race |
|
|