[00:07] when you go on "See full publishing history" you can also see the dates when it happened [02:01] zooko: Accepted. You may get a reject mail, but that was me rejecting the previous one. [02:01] jdong: ^^ [02:03] ScottK: *nods* cool :) [02:03] zooko will be happy [02:10] Does anyone else have issues with graphics in FF 3.5? [02:10] Changing a gfx config in about:config in FF fixes this. [02:10] All SIS comps seem to be affected by this color correction issue. [02:11] *SIS video chip [02:11] gfx.color_management.mode [02:11] when set to 0 fixes this. [02:12] Hello? [02:25] Coolg1026: #ubuntu+1 for Karmic help. [03:19] Oh boy oh boy oh boy... [03:19] * zooko sees that tahoe-lafs is gone from the queue... [03:19] * zooko searches on packages.ubuntu.com... [03:19] Okay, it is not there yet, but I'll bet it will show up soon after a regularly scheduled job! [03:19] zooko: I accepted the binary too, so once the current publisher run finishes, it should be installable. [03:20] zooko: Probably another 25 minutes plus however long it takes for it to get to your mirror. [03:21] zooko: https://launchpad.net/ubuntu/+source/tahoe-lafs [03:21] zooko: I imagine it seems like we're unbelievably anal about the licensing stuff, but we really do have to be careful. [03:22] Whoo-hoo! :-) [03:22] I really enjoyed seeing the intimate details of how picky you are about licensing. Very interesting. [03:22] I have a deep interest in open source licensing already. [03:23] Whoo-hoo! There is is! I'm now going to announce it excitedly on tahoe-dev mailing list, my blog, etc. [03:24] ScottK: well, we think we have to be :P [03:25] lifeless: Certainly. [03:26] Hm, well I think this means it is time to upgrade my main development workstation from jaunty to karmic. :-) [03:26] * StevenK looks at the NBS queue and goes 'argh' [03:28] * ScottK figures then that StevenK would faint if he reviewed the archive rebuild test results. [03:29] I doubt iut [03:29] s/iut/it/ [03:29] The good news is that slightly over 7/8 of the packages in the archive can be built in the archive. [03:30] which still leaves 1/8 that don't? [03:31] Here goes the big apt-get dist-upgrade. :-) [03:31] Yep [03:31] zooko: Use update-manaer [03:31] zooko: do-release-upgrage -d [03:32] Hm. Are you sure? I've heard of that before, but I've been using apt-get dist-upgrade for, you know, almost a decade now, most recently two days ago to update some servers from feisty->gutsy->hardy. [03:32] I'm a traditionalist stick-in-the-mud and I hate new things. [03:32] It will often work, but it's not the supported method. [03:32] But, if you say so... [03:32] And it will break sometimes. [03:32] Feisty->Gutsy was particularly bad for me. [03:33] I do it sometimes too for testing, but it has work arounds for known problems so your upgrade is more likely to succeed. [03:33] Whoops, I accidentally installed upgrade-manager from karmic instead of jaunty. Oh well. :-) [03:33] last I recall, dist-upgrade is still supported but doesn't have the specific handling for corner cases [03:34] Right, so if you hit those cases, you are suddenly unsupported.... [03:34] By the way, is update-manager shared with debian? [03:34] There was a Google SOC project to bring it to Debian. [03:34] * zooko reasd update-manager --help. Cool, aufs. [03:34] Not sure how it worked out. [03:35] by the way, it says "No new release found". [03:36] Ah, that's because it already thinks it as at Karmic. Whoops. [03:36] Hm. Might as well try out dist-upgrade just in case I can help out anyone else who tries this and hits a snag after Karmic is release. ;-) [03:37] zooko: -d to say "Look for development releases" [03:40] StevenK: I did. The problem was I accidentally apt-get installed update-manager from karmic, and it brought in base files, so now it thinks it is already at Karmic. [03:41] \o/ [03:41] ScottK: I would like to rename the project page in launchpad from allmydata-tahoe to tahoe-lafs. Ideally a vestigial "forwarding" page would be left at allmydata-tahoe pointing to tahoe-lafs for a while (a few years?). Do you know how to accomplish such a thing? [03:41] zooko: #launchpad [03:41] I guess I could g.. [03:41] What lifeless just said. [03:43] yeah update-manager -d tells me no new release found also [03:43] must be broken right now [03:46] What URL should I give to people accompanying encouragement that they should upgrade to Karmic in order to help beta test? [03:46] http://www.ubuntu.com/testing/karmic/alpha6 [03:46] Thanks. [03:48] ext4 default! You crazy kids. [03:49] crazy like a fox; and you know how that turns out [03:57] zooko: BTRFS! BTRFS! [03:58] *withholds output of mount on his present machine* [03:58] Yeah, btrfs is going to be great. [03:58] I messed around with reiser4 on my workstation at my day job last week. :-) [04:04] zooko: It's likely because you had the Karmic upgrade-manager that it didn't find a later release. [04:04] Right, now I see you figured that out alread. [04:04] y [04:06] Is ubuntu-bug a good way to report bugs? [04:07] I've been using launchpad web site, but I prefer not to use web ui if ubuntu-bug is good. [04:07] Hm, what is this thing that I'm talking about. Ithought someone mentioned it on this channel recently, but now I don't see it. [04:07] Oh well, I'll use the web site. [04:08] zooko: ubuntu-bug is the preferred method [04:08] ubuntu-bug is the cannonical way to file bugs, yes. [04:08] Ouch [04:08] Ouch what, the fact that "ubuntu-bug" doesn't appear in "apt-cache search ubuntu-bug"? [04:08] Eh, spelling is for the weak. [04:09] Oh, it matches "reportbug". [04:09] Don't use reportbug. [04:09] But at least in the bad old days, "reportbug" would send your bug reports to somewhere they weren't wanted. [04:09] RAOF: The ouch was for the pun, not the spelling. [04:10] * RAOF wonders idly why a pulse callback that could obviously be called exactly once actually gets called exactly twice, with NULL the second time. [04:12] So, how do I get ubuntu-bug? [04:13] It'll be in apport somewhere, right? [04:13] dpkg -S says... [04:13] apport :) [04:17] Thanks. [04:22] Yikes, apport won't work with my current system (not surprising), but the "yikes" is that launchpad takes me to a page telling me to use apport when I click "report a bug". [04:22] Hello launchpad. I'm talkin to *you*... [04:38] It's on purpose. I did file a bug suggesting the redirection to require using apport to report bugs not be applied to apport itself. [04:39] Ah, well I wasn't going to report a bug in apport. [04:39] I guess if you guys are sufficiently serious about this, then I should just take notes and then once apport works I can go ahead and report the bugs. [04:39] There's a good chance I'll have forgotten/lost interest but then, but oh well. [04:40] zooko: did you see the url? [04:40] I pointed you at it in #launchpad [04:43] lifeless: that URL -- https://bugs.edge.launchpad.net/ubuntu/+filebug -- redirects to https://help.ubuntu.com/community/ReportingBugs . [04:43] zooko: yes, and on that page [04:43] zooko: -> #launchpad [04:45] lifeless: okay I finally see it. Thanks! [04:46] lifeless: https://bugs.launchpad.net/ubuntu/+source/knode/+filebug?no-redirect gives me a "no such page" error. [04:47] the source package is kdepim I think [04:47] yes [04:47] change knode to kdepim [04:48] zooko: apt-cache show knode will tell you for sure what the source package is. [04:51] ScottK thanks. [04:51] You're welcome [04:52] I probably would have figured that out if it weren't past my bedtime. Maybe I should stop reporting bugs and go to sleep. [04:52] Sadly, I have to go to $dayjob in the morning, so I have the feeling that I can prolong the weekend and hack on open source for fun if I stay up. [04:55] zooko: Just remember that sleep is for the weak. [04:55] :-) [04:56] My wife was just reminding me that I've stayed up late for the last 5 days in a row and I'm getting really grumpy. [04:56] s/ak/ekend/ [04:56] I'm really happy about Tahoe-LAFS being in Karmic, though. [05:38] good night folks! [06:28] who takes care of emacs-snapshot? [06:38] how to use dh_link.. is it necessary to create debian/links file? coz i was going through a makefile which has something like this... dh_link -p [06:39] wrapster, Creation of debian/links or debian/${package}.links is optional, but it's often easier to read for others than embedding it in debian/rules. [06:39] persia: so could you tell me how to use it.. the man page hardly has anything [06:39] So, if you're creating from scratch, or adding links, best to use debian/links, but if you encounter a package that embeds it, best to follow that when adding more links. [06:40] wrapster, What are you trying to accomplish? [06:40] assume it this way.. im trying to repack gcc by creating a symlink to /usr/sfw/bin/gcc from /usr/bin/gcc [06:40] how do i go about it. [06:41] as an eg [06:41] the similar was achieved through native commands but i want to use dh_link [06:41] to learn... [06:42] For that example, I'd not use dh_link, as I think you're seeking to use an alternate gcc, and would therefore recommend investigation of the alternatives system. [06:42] That said, use of a debian/links file (in the format of source destination) is easiest. [06:42] Otherwise, it's dh_link [debhelper options] [-A] [-Xitem] [source destination ...] [06:43] good morning === santiago-ve is now known as foursixnine [06:45] persia: thats what i did.. and in the rules file just add dh_link is it? [06:45] Yes. [06:45] under the proper rule ofcourse [06:49] and when i do create the links.. how should the path be specified.. generally if its hardcoded in the rules file.. an i want a link at usr/sfw/bin/ then this is what i will do.. ln -s debian/ ... is that how i write the links files as well. or just /usr/sfw/bin/ [06:50] From the manpage: "Be sure you do specify the full filename to both the source and destination files (unlike you would do if you were using something like ln(1))" [07:04] persia: ok cool.. thats what i've done lets see how it goes.. [07:12] Heya, tried asking this question in ubuntu-kernel but I guess no one knows or its not appropriate. i have just built a 2.6.31kernel with make-kpkg and I have used the --initrd switch however it still says ' please update the initramfs manually '. is there a way i can build an ubuntu package which will do this automagically on installation ? [07:13] I would like to just install the images and have everything configured if that is possible? [07:28] persia: http://pastie.org/633028 [07:28] i get this error.. [07:29] the links file has these.... [07:29] ### comments in the first line ;;;; [07:31] paste your link file [07:33] persia: (man) A file named debian/package.links can list pairs of files. [07:33] what i had created was debian/links [07:33] let me see if this change works... [07:33] then i'll do a pastie. [08:09] slangasek: Hello [08:52] persia: i still get this http://pastie.org/633028 error [08:52] And you still haven't pasted the call that gets the error :) [08:52] http://pastie.org/633064 [08:53] and thats the links [08:53] And the dh_links line in debian/rules ? [08:53] one moment [08:55] http://pastie.org/633066 [08:55] My, that's ugly. [08:56] y [08:56] i just started so could you tell me why? [08:57] For this package, I'd probably skip dh_link, and use a construction like that used for libintl.so [08:57] why [08:57] what difference does it make? [08:57] coz if you look at the rules thats exactly what i've done [08:57] rm debian/$@/usr/lib/charset.alias [08:57] rm debian/$@/usr/lib/libintl.so.8* [08:58] note that the second line is broken [08:58] its got a space, not a tab [08:58] thats not valid Make syntax [08:58] I think it's ugly because: 1) it uses cp rather than install, 2) it uses rm in build, rather than using it in clean or patching the build system, 3) It manually splits the package, rather than using dh_install, It forcibly strips the binaries, rather than using dh_strip (so there's no ddeb). [08:59] I could probably go on, but the short form is: it does it the hard way, and it's long. [08:59] however the breakage is the space-instead-of-tab [08:59] so i cannot use dh_link [08:59] in this case [09:00] Well, you can, but I wouldn't. That said, if you fix the issue that lifeless pointed out, it ought work a bit better. [09:00] ok thanks [09:01] The reason I'd use a construction like that for libintl.so is because I think it's best practice to try to match the style of a rules file when modifying it: this tends to cause less confusion than mixing styles for the next person. [09:02] oh ok. [09:03] secondly.. there is one more question... now that i'm making some changes to the actual pkg.. i would like to give it a different version no.. how do i do it? [09:03] changelog? [09:03] Yes. You'll want to update the changelog with `dch -i`. This tries to guess the right next version for you. [09:05] just dch -i? [09:05] yes. From within the package directory. [09:06] ok [09:07] supposing i want to write something that i've changed for references then i do dch -i it opens up the log there beside the * i write what i want and save? === dholbach_ is now known as dholbach [09:08] wrapster, Yes. Add more '*'s if youare doing more than one thing. Try to stay within 70 characters. [09:11] how do i save it .. no save option.. should i first ^x then save or writeout? [09:12] Which editor launched? [09:13] nano [09:13] and the file is changelog.dch [09:14] I think you need ^O and then ^X (but I'm not much of a nano user). [09:15] ^o is writeout.. === asac_ is now known as asac [09:16] ok so it says file name to write to and gives me changelog.dch . shall i remove the .dch or keep it ? [09:17] wrapster: neither [09:17] save and quite the editor [09:19] lifeless:there is no save option so i will have to ^x it will save in the buffer but then again asks the same question to save in .dch file? [09:22] instead of this cant i manually edit the chagnelog and add what i want ? [09:23] wrapster: save and exit [09:24] <\sh> moins [09:24] wrapster: I don't know what forsaken editor you're using, so you'll have to figure that out yourself. I recommend you set EDITOR to whatever you normally use [09:24] how do i set the editor? [09:24] export EDITOR=vim [09:24] or whatever [09:26] lifeless: editor is not set at all dont know from where its picking up nano.... $EDITOR is null [09:26] if ther is a env var like that :( [09:29] wrapster: perhaps you should follow my recommendation then? [09:29] I have to say it scares me that you are changing core toolchain packages while you're still learning standard Linux facilities. [09:30] well... im learning .. [09:30] hihu \sh , nice post yesterday :) === yofel_ is now known as yofel [09:56] <\sh> sebner: depends...for me it was not nice to see brown scum on my ballot paper [09:56] \sh: heh, how much % got they? [09:57] *many [11:02] <\sh> sebner, not so many...but read this: http://www.spiegel.de/politik/deutschland/0,1518,650151,00.html [11:07] \sh: yeah, it's pretty pretty stupid .. sick! [11:11] it's in germanese :/ [11:25] directhex: google translator :P [11:31] hmm emacs-snapshot's broken [11:32] =( [11:45] huhu sistpoty|work [11:46] hi folks [11:46] hi sebner [11:46] sistpoty|work: I was bored so I already did the paperwork for xmoto, you just need to test *g* and ACK [11:50] sebner: hm... let's see if it still has the "too difficult for sistpoty" bug :P [11:54] sistpoty|work: heh, yep. I suggest a deep investigation *cough* [11:54] haha [12:02] sistpoty|work: gnahh, thx for the hint. changed [12:03] sebner: no problem [12:08] sistpoty|work: bah LP is soo buggy, I can't edit the bugreport [12:09] sebner: heh, happened to me this weekend as well [12:23] this bug ? https://bugs.launchpad.net/malone/+bug/423924 [12:23] Launchpad bug 423924 in malone "Entity-body was not a well-formed JSON document when updating bug description" [High,Triaged] [12:30] if i change the version number while installing i get an error saying it depends on some other software which is not of the same version.. [12:30] whats that suppose to mean? [12:34] wrapster: I can't parse your question (are you editing the .deb to change a version number in it while apt is reading it? that doesn't sound like a good idea :) [12:35] hmmm.. im changing the version no in the changelog file :) [12:35] not a good idea aye? [12:36] I still can't parse your question [12:36] the version number of the built source package is determined from the changelog, so adding a new version there is perfectly reasonable [12:38] mzz: thats exactly what I'm doing [12:38] then perhaps it'd help if you gave specifics (what package, what version numbers, what error message) [12:38] ok.. [12:38] one moment. === Tonio__ is now known as Tonio_ [12:57] mzz: the dpkg-buildpackage comes through but dpkg -i gives me this http://pastie.org/633066 [12:58] wrapster: looks like gcc and cpp want to be kept in lockstep [12:59] yeah. the versions need to match, so both packages need to be installed at once [12:59] wrapster: so you need to build a matching version of cpp (if that's not built from the same source package) and install both at the same time [13:00] wrapster: also, I strongly agree with earlier comments about the toolchain not being the best place to start experimenting [13:01] wrapster: (not just because they're nontrivial packages, but mainly because you risk running into weird failure at a distance much later on because of changes you make now) [13:10] looks like dependency versioning is wrong, 4:4.2.3-1-1nexenta4 has "-1" twice in it === doko__ is now known as doko [13:24] randomaction: yeah [13:25] shucks!! [13:52] hey all... is keyserver down? [13:53] flithm1: Known issue. [13:53] thanks === ember_ is now known as ember === tgm4883` is now known as tgm4883 [15:25] Heya gang [15:27] heya bddebian [15:27] Hi sebner [15:28] Hey bddebian [15:33] Hello fabrice_sp__ === fabrice_sp__ is now known as fabrice_sp [15:36] hi bddebian [15:37] Heya sistpoty|work === nixternal_ is now known as nixternal === roaksoax_ is now known as RoAkSoAx [16:04] Not sure what else I need to do for LP bug 424576 - I posted a debdiff and diff.gz, do I need to upload to REVU? Also depends on libmixlib-config-ruby uupdate to 1.0.12, which can be synced from Debian Testing (LP bug 420674) [16:04] Launchpad bug 424576 in chef "Chef new upstream version 0.7.10" [Medium,Confirmed] https://launchpad.net/bugs/424576 [16:04] Launchpad bug 420674 in libmixlib-config-ruby "Update libmixlib-config-ruby to new upstream version 1.0.12" [Undecided,New] https://launchpad.net/bugs/420674 [16:14] how does the freeze interact with new packages for universe? [16:14] lamont: you need to get a feature freeze exception [16:15] Laney: ok [16:15] from motu, or RM? [16:15] motu-release [16:15] !ffe [16:15] uvf is Upstream Version Freeze. For an exception, see https://wiki.ubuntu.com/FreezeExceptionProcess [16:19] ta [16:19] and I'll worry about it for post-beta, if not karmic+1 [16:20] MOTU team, quick packaging question: I created a package under Jaunty, it is compat level 7. I cannot get it to build in Launchpad on Hardy, because the compat level is too high. Can I "downgrade" my compat level? [16:21] irvingpop: yes, but it may require changes beyond just changing the number [16:21] irvingpop: If the package doesn't use any debhelper 7 features, yes. [16:21] irvingpop: possibly extensive changes [16:21] Hardy is compat level 6, correct? [16:21] hardy-backports has dh7 [16:22] Could that help me in a Launchpad PPA? [16:22] I'm not sure if PPAs build against -backports [16:22] When I required debhelper >= 7, it wouldn't build on Hardy due to missing dependency [16:23] I changed that to 6, and changed the compat file to 6, but it still won't build on Hardy. [16:23] on a PPA? [16:23] (clarification) on the Hardy PPA build box [16:23] you can either fix it to work on 6 [16:23] or upload the dh7 backport to the ppa [16:23] Laney: you can select -backports as a dependency to ppa [16:24] yeah if that works :) [16:24] oh cool, how can I do that? [16:24] I don't really use ppas that much [16:24] irvingpop: there should be a link 'Edit dependencies' [16:24] oh, how cool is that? [16:24] trying now [16:25] selecting backports will work [16:25] unless you use 7.0.50 features ;) [16:36] guys im back with one last problem please help.. like i was saying i created a few symlinks under debian//path and installed the result of dpkg-buildpackage.. but its not reflected on my machine.. [16:37] however if i open up debian//path and see the files exist and a ls -l will display the source for that symlink as well [16:37] what could be worng? [16:46] Laney and cemc, thanks for the tip. Resubmitting now [16:47] I'll submit for sponsorship after the Karmic crazyness is over :) === nxvl_ is now known as nxvl [17:41] kees, ping === Tonio__ is now known as Tonio_ [18:09] hi [18:09] can i reply by e-mail to bug comments i receive from lp via e-mail? if so, is there anything i have to look out for? [18:18] e.g. what will happen with attachments/pgp signatures? [18:32] moldy: they are no problem [18:33] If I am right, they get stripped [18:34] kees, i already have the security updates debdiff for nginx: https://bugs.launchpad.net/ubuntu/+source/nginx/+bug/430064, however I do not know which kind of tests are required for the patched code or to test the vulnerability [18:34] Launchpad bug 430064 in nginx "Security fix in recent release 0.6.39/DSA-1884-1" [High,In progress] [18:44] RoAkSoAx: sounds like jdstand covered that pretty well on #ubuntu-hardened. If you can get the service up and serving files, that's a reasonable test that it still works. :) [18:46] kees, cool :) I already brought the service up and it serves files, so I guess I'll just wait till someone of you reviews it :) [18:46] RoAkSoAx: ok, cool === fabrice_sp__ is now known as fabrice_sp === porthose is now known as porthose|afk === Guest22498 is now known as YDdraigGoch [21:49] "This bug was fixed in... mplayer (2:1.0~rc3+svn20090426-1ubuntu7) karmic; urgency=low" That was just posted. When is it likely to show up in the archives? [21:53] Darxus: normally within an hour or two, but it's possible that updates are being let through manually since it's beta freeze time [21:54] ajmitch: Thanks. [21:54] dtchen: OK, I've grabbed the alsa-driver snapshot, I just need to find the proper way to build it :) [21:55] Quick devhelper question: I want to convert my package from using init to Upstart. What's the best way to do this AND remain compatible with Jaunty? [21:56] I see that Upstart is in /etc/init on Karmic, but in /etc/event.d on Jaunty. Plus, Jaunty's dh_installinit has no idea about Upstart === pochu_ is now known as pochu === ShadowChild is now known as lukjad007 === porthose|afk is now known as porthose [23:41] Folks: I have a user asking why there aren't PowerPC .deb's of a few packages: http://allmydata.org/pipermail/tahoe-dev/2009-September/002930.html [23:41] What's the answer? [23:42] zooko: I thought the powerpc support was ended [23:43] Aha. I'll google for that. [23:43] I'm pretty sure there are still some people who try to keep PPC up and running; TheMuso is one, from memory. [23:43] https://wiki.ubuntu.com/PowerPC [23:44] But it's a port, no longer official, IIRC. [23:45] zooko: are you sre there are missing dependencies there? [23:45] packages.ubuntu.com may only be showing the official architectures [23:46] launchpad shows them all, eg https://edge.launchpad.net/ubuntu/+source/pycryptopp [23:47] b#da [23:47] oops [23:47] Well, it sounds like I need to ask Danny O'Brien to hop into #ubuntu-motu. I don't know what's going on. [23:47] And I don't have an PPC machine to experiment with either. [23:48] zooko: yeah, it seems they should have no problems [23:48] zooko: if they are just looking at filenames they may be confused [23:49] james_w: he wrote "The powerpc builds don't seem to be available at that repository " [23:49] > (yet?). [23:49] ports.ubuntu.com is what they want to actually install it on powerpc [23:49] (so they can get the architecture-specific dependencies) [23:50] So I can post to the tahoe-dev mailing list saying to make sure they have ports.ubuntu.com in addition to http://us.archive.ubuntu.com/ubuntu/ ? [23:50] instead of I think [23:50] "are running karmic" should cover it :-) [23:51] let update-manager figure it out === jerbear is now known as jmcantrell [23:58] james_w: I didn't understand your last two sentences.