[00:06] bryceh: ok, the fglrx package appears to check out as DTRT, I think the submitter of that bug simply didn't use the Ubuntu fglrx package [00:07] slangasek, yep, that's what I thought. It's sort of a known donut hole. I put it in our spec and test plans to try to handle the corner case better but I think tseliot had too many other tasks to juggle [00:08] the problem is that the changes for the fglrx packaging don't trickle into AMD's releases for 2-3 releases === Lutin is now known as Guest14183 === jussi01 is now known as Guest4656 === sam_ is now known as samferry === jelmer_ is now known as jelmer === ]reed[ is now known as [reed] === mnepton is now known as mneptok === sconklin is now known as sconklin-gone [02:40] hrm... seems the disk UUID mapping gives preference to the snapshot over the original volume.... not good. [03:34] psusi: What, still?!? [03:35] * soren facepalms [03:38] kees: Would you say bug 460906 is important enough to be fixed this late in the cycle? [03:38] Launchpad bug 460906 in lvm2 "disk/by-uuid/foo symlink points to snapshot rather than the origin" [Undecided,Confirmed] https://launchpad.net/bugs/460906 [03:38] * soren thinks so [03:40] that suonds like a potentially problematic bug [03:41] Indeed. I spotted it quite late in the karmic cycle (a few days before release, apparantly) and then forgot all about until just now. [04:39] soren: I still don't have much of an opinion about it. I think by-uuid for a snapshot is kind of meaningless (i.e. the lv name is the unique part) [04:39] soren: there is logic for the link either direction. [05:22] doko_: I've confirmed that bug #417757 *is* still present in lucid [05:22] Launchpad bug 417757 in eglibc "[karmic regression] all network apps / browsers suffer from multi-second delays by default due to IPv6 DNS lookups" [High,Fix released] https://launchpad.net/bugs/417757 === Guest4656 is now known as jussi01 [07:03] Can I make pbuilder automatically run lintian on the .changes file it generates? If so, how? [07:09] Good morning [07:09] beuno: aptitude> hm, not sure; mvo? === tkamppeter_ is now known as tkamppeter [08:19] good morning [08:22] <[reed]> anybody know why Ubuntu's apache2 doesn't link against libssl like Debian's does? [08:36] [reed]: from the PTS page (http://packages.qa.debian.org/a/apache2.html) you can find a link to the patch between Debian and Ubuntu. That usually explains why. [08:37] I can't see anything obvious there, so maybe you misdiagnosed it? [09:00] pitti: Hello, please wait a moment before closing bug 536914. I found out there are other few keys that needs fixing (<255) I'll tell you when I finish. [09:00] Launchpad bug 536914 in udev "[Lucid Alpha3][khk-004 testcase failed] Click on Internet hotkey opens Home folder on Acer Laptops" [Medium,Fix committed] https://launchpad.net/bugs/536914 [09:02] primes2h: hm, I thought I reopened it already? [09:02] ah, I did [09:03] primes2h: ok, waitin gthen [09:03] pitti: Yes, I mean, there is still other 2 keys I have to triage [09:03] pitti: thanks [09:06] pitti: I also sent a message to our Italian Testing ML asking Simona (Aspire 1640) to tell me about her full keyboard layout. [09:06] pitti: I'll let you know ASAP [09:10] pitti: Would I break anything if I were to include the checking of hd[a-z] in the 60-cdrom_id.rules udev rule? This is needed for lucid only, as there is no libata version of powerpc/mac's IDE driver. A driver is supposedly in 2.6.33, so this rule change is only needed till lucid+1, and its safer than backporting a driver in the kernel. [09:10] lwindow 1 [09:11] oops sry guys ;) [09:11] TheMuso: this sounds very similar to what Debian has to do; I thought mbiebl already had a patch for IDE CD-ROMs, let me check [09:13] TheMuso: http://git.debian.org/?p=pkg-utopia/udisks.git;a=blob;f=debian/patches/10-ide-cd-support.patch;h=59a19e4eaf414b9b8bdf0588898d964ff95613f3;hb=HEAD [09:13] TheMuso: it basically adds the blkid call for ID_CDROM_MEDIA hd* [09:13] TheMuso: is that what you need? [09:14] pitti: No, I mean actually editing 60-cdrom_id.rules to check for add|change nodes on hd[a-z] for IDE CD devices that are on the pmac driver in 2.6.32 and below, which doesn't have a libata equivalent. [09:14] I'm just wondring whether checking those nodes shoudl they exist will break anything. [09:15] so the change is in udev itself. [09:15] ah, I see [09:15] TheMuso: hm, Debian should have that change as well then, otherwise mbiebl's udisks patch couldn't work at all [09:16] true [09:16] * TheMuso checks sid's udev. [09:16] TheMuso: so, I don't think it would hurt at all, but it'd be the first and only delta to upstream that we have in udev, and I don't think Keybuk will like it a lot [09:16] TheMuso: OTOH there's nothing wrong with shipping a 60-cdrom_id-ide.rules in a powerpc specific package :) [09:16] pitti: RIght, I can understand that. [09:17] pitti: There is that too. I'll have to see how that could be included in d-i, as d-i installation from alternates on powerpc are broken due to this issue. [09:18] TheMuso: oh, right, I could just stuff the new rule into debian/ in udev [09:18] TheMuso: and only install it on powerpc, to be on the safe side [09:18] pitti: that works [09:18] pitti: there is an LP bug for this, reported by someone else. I'll fetch the number. [09:19] TheMuso: confirmed, sid's udev has the rule like that [09:19] pitti: oh ok you beat me to it. [09:19] TheMuso: ok, please assing it to me; I'll have a quick discussion with Keybuk about how he wants this packaged, and get to it [09:19] bug 534912 [09:19] Launchpad bug 534912 in udev "[Lucid powerpc alternate] No common CD-ROM drive was detected." [Undecided,Confirmed] https://launchpad.net/bugs/534912 [09:20] pitti: ok thanks. === Guest14183 is now known as Lutin [10:33] cjwatson: hi, are you ok with the name "SUPPORTED_OVERWRITES.hints" for the fine-tuning file for lucid-support-timeframe? I will put it in lp:~ubuntu-core-dev/ubuntu-seeds/platform.lucid - I'm happy about better matching names [10:39] cjwatson: or maybe SUPPORT_TIME_OVERWRITES.hints ? === dpm is now known as dpm-bbl [11:00] kenvandine: hi, are you here? :) [11:01] mvo, hey, I need advice to fix bug 130289 [11:01] Launchpad bug 130289 in apt "Encode any ":", "@" or "/" within the user and password field in proxy settings." [Medium,In progress] https://launchpad.net/bugs/130289 [11:01] jibel: let me have a look [11:02] mvo, the problem is with the way apt reads it's configuration and parse encoded characters. [11:02] e.g %40 is stored internally as @ [11:02] It's then really difficult to parse an URI with password containing these characters [11:02] mvo, Is there any way to read the raw value of a configuration item ? [11:03] If there's no trivial solution I'll update the URI parser. [11:06] jibel: something like QuoteString etc from apt-pkg/contrib/strutils.cc does not help I assume? [11:08] mvo, no it will quote the whole string. if there something like Acquire::http::Proxy "http://user2:p%40s%2Fs%3Aword@localhost:3128"; [11:09] mvo, I need the value http://user2:p%40s%2Fs%3Aword@localhost:3128 not user2:p@s/:word or worse user2:p%40s%2Fs%3Aword%40localhost%3A3128 [11:10] jibel: could we read the string from the config, use the URI class to convert it, and then QuoteString/UnQuoteString only on URI.User, URI.Password? [11:11] jibel: I guess it would make a lot of sense to add a GetRaw to the configuration class in apt itself, let me have a look [11:13] I love seeing fixes for race conditions by adding ‘sleep 3’. :-) (foo2zjs changelog) [11:13] mvo, That's what I was thinking too. Add a FindR method and a RawValue member to the Item struct [11:23] tkamppeter: Isn’t having a udev script call sleep EBW? [11:23] kenvandine: so far I couldn't get libproxy to recognise different server for https, and also it ignores auth. user and password in GNOME proxy settings [11:23] ion: elephantine beyond words? [11:23] oh, "evil bad wrong"? [11:23] bingo [11:24] yes, udev doesn't allow long-lived child processes === jelmer_ is now known as jelmer [11:27] mvo, the problem is elsewhere. apt reads and stores its configuration correctly but it converts and stores the string a second time. [11:30] kenvandine: seems like we will have to patch libproxy to support more than simple proxy handling to have proper support. [11:38] mvo: all-caps is reasonable to avoid clashes - how about SUPPORTED-HINTS? [11:39] EBW> I like "B&R" (Bad and Rong) [11:46] cjwatson: thanks, SUPPORTED_HINTS is fine, I will use that [11:58] asac/ogra: http://launchpad.net/builders/gourd [11:58] this will especially be a win if eglibc is FTBFS on imbe [11:59] mvo, the problem is in pkgAcqMethod::Configuration where all the config items are "dequoted" I guess [11:59] jibel: ohh, that makes sense [12:00] asac/ogra: since my testbuild of that version of eglibc on gourd resulted in bug 546808 [12:00] Launchpad bug 546808 in eglibc "armel v7 test failure: test-fpucw, tst-cpuclock2, tst-eintr1, tst-timer" [Undecided,New] https://launchpad.net/bugs/546808 [12:01] pecisk, can you check in #libproxy first? [12:01] pecisk, i would be surprised if that was the case [12:01] pecisk, nevermind... i see you did [12:01] rock on :) [12:01] kenvandine: i run trough code, it is very simple, they handle only http now [12:02] is there a long backlog of builds or other problems, since ppa packages seem to take many hours before getting built [12:02] problem is - if you want to use libproxy in Lucid, it would have rather large feature set to add, and it is kinda no no as far as I know [12:02] nevermind that code is kinda simple [12:02] jibel: apt-get update -o Debug::pkgAcquire::worker=true will show you the configuration messages that are send [12:02] run/ran/s [12:03] tjaalton: lots of uploads, and the extra machines that we generally steal to be ppas are being used for their primary role atm, it would appear [12:03] at least "many of" [12:03] lamont: ok, I'll wait then [12:06] though at least i386/amd64 buildd's don't have any "'currently building' build records" [12:07] https://edge.launchpad.net/builders [12:07] mvo, that's it. I removed the DeQuoteString and the password is ok. [12:07] hah, a lot accurate it seems :) [12:07] +more [12:08] * tjaalton bookmarks [12:08] damn daily/nightly builds ;) [12:11] cjwatson, hi, could you have a look at bug 545790, 545738, 454354. This is an odd error and there's an increasing number of reports since a few days. [12:11] Launchpad bug 545790 in dpkg "package PACKAGE failed to install/upgrade: error writing to '': Success" [High,Confirmed] https://launchpad.net/bugs/545790 [12:11] Launchpad bug 545738 in dpkg "package PACKAGE failed to install/upgrade: error writing to '': No such file or directory" [High,Confirmed] https://launchpad.net/bugs/545738 [12:11] Launchpad bug 454354 in dpkg "package PACKAGE failed to install/upgrade: error writing to '': Input/output error" [Medium,Confirmed] https://launchpad.net/bugs/454354 [12:12] cjwatson, there error is triggered when dpkg fflush an error message but the fflush fails. [12:12] cjwatson, The scenario seems to be: user run update-manager, system crash, reboot, apport triggers and report those errors. [12:13] lamalex, i dont think the test is a HW specific failure (fails on pegatron as well as on babbage afaik) [12:13] lamont, ^^^ [12:13] mvo, do you think adding an untouched raw config value is the right solution ? [12:16] jibel: any idea when it started appearing? [12:16] jibel: (BTW I really appreciate this sort of active triage/collation work and direct contact with developers on the hot bugs - it's invaluable) [12:17] $ wcgrep 'm_output.*standard output' | wc -l [12:17] 35 [12:17] oh dear [12:18] cjwatson, in karmic, but the rate is higher since the latest beta [12:18] could just be increasing number of upgrades, then [12:18] jibel: you removed the UnQuoteString in ::Configuration ? or where? [12:20] jibel: there was a crash bug for a couple of hours in update-manager that may have triggered this [12:20] mvo, yes, but that's definitely not the right solution ;) [12:20] jibel: that is why I asked :) [12:21] mvo, but that confirms that's the problem since there's no more 407 from the proxy. [12:21] jibel: hm, hm, tircky to solve [12:21] ogra: hence the bonus points if it does fail on pegatron [12:21] jibel: but great that we know excactly what the problem is now [12:22] mvo, have some lunch. cu [12:22] jibel: see you (and thanks!) [12:26] lamont: can you try to build libplist on one of the newer builders? [12:28] asac: assuming it builds, how long does it run? [12:30] lamont: really quick [12:30] a couple of minutes [12:31] lamont: its the one that failed because of thumb2 instructions in the test binaries [12:31] /home/buildd/libplist-1.1/obj-arm-linux-gnueabi/swig/plistPYTHON_wrap.cxx:9369: warning: deprecated conversion from string constant to 'char*' [12:31] giggle [12:31] ah, cool [12:32] tests 100% passed. [12:32] * asac happy! [12:32] lamont: was that a real archive build? or just a test build? [12:32] e.g. are the bits now in ? [12:32] just a test [12:32] ah ok [12:33] forcing the build to happen there requires, um, more work [12:33] heh [12:33] * lamont goes to throw it against the wall a little [12:35] but that smells very good already :) === dendrobates is now known as dendro-afk [12:51] keybuk: You might find a udev script doing ‘sleep 3’ interesting. Oh, and the ‘sleep 3’ is a also ‘fix’ for a race condition. ;-) The package is foo2zjs. [12:51] *shrug* [12:51] probably works just fine in mandriva === rgreening_ is now known as rgreening === MacSlow is now known as MacSlow|lunch === oSoMoN_ is now known as oSoMoN [13:17] libparted segfaults in daily lucid server cd when trying to get partition tables from hard drives === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === sconklin-gone is now known as sconklin [13:23] pecisk: please submit a bug report with full logs === dpm-bbl is now known as dpm [13:39] c [13:40] will do [14:06] jibel: I updated bug #130289 with a possible solution, I would be interessted if it helps [14:06] Launchpad bug 130289 in apt "Encode any ":", "@" or "/" within the user and password field in proxy settings." [Medium,In progress] https://launchpad.net/bugs/130289 [14:07] ion, the sleep is called in a script which the udev rule starts in the background. The udev rule exits while the script is still running. [14:07] tkamppeter: why does it sleep though? [14:08] pitti: what's the easiest way for me to test that new ntfs-3g still works with udisks? [14:09] cjwatson: create an USB stick with an NTFS partition, plug it in, and see if it automounts? [14:09] cjwatson: I also have a test suite for udisks, which covers NTFS as well [14:10] http://cgit.freedesktop.org/udisks/tree/tests/run [14:10] you can just run that [14:11] cjwatson: but if an NTFS usb stick automounts, it should be okay; I don't suppose it changed the mount.ntfs-3g CLI, did it? [14:13] shouldn't think so [14:18] Please sponsor sync Bug #546482. It's bugfix & translation lucid-special LTS upstream release. [14:18] Launchpad bug 546482 in xiphos "Sync xiphos 3.1.3-1 (universe) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/546482 [14:24] Riddell: I attached a patch to bug #546024, you may want to check if the version is correct and commit to bzr (I don't think I have commit access to the relevant branch) [14:24] Launchpad bug 546024 in phonon "file overwrite error" [High,Confirmed] https://launchpad.net/bugs/546024 [14:29] Please see bug # 546929 as this affect new installations in today's netboot installer [14:29] pitti: thanks, ntfs test passes, uploading [14:29] \o/ thanks [14:29] didn't have a handy USB stick but I guess "fs: NTFS ... [cli] [ud] ok" is good enough [14:30] lemsx1: *please* don't file bugs against the 'lucid' project [14:30] cjwatson: right [14:30] cjwatson: out of interest, how much of the tests passed for you? [14:30] lemsx1: read the text in ALL CAPS on https://launchpad.net/lucid [14:30] s/much/many/ [14:30] pitti: oh, I just did './run FS.test_ntfs' [14:30] cjwatson: ah, fair enough [14:31] cjwatson: I was just curious, since I didn't get a lot of feedback from other people/distros about it yet (except from mbiebl) [14:31] but nevermind [14:31] lemsx1: I've reassigned your bug to the proper place [14:32] lemsx1: of course the developers of that text editor will keep on getting mailed about it, but ... [14:35] mvo: We get phonon from qt4-x11 now. I moved the bug to the right package. [14:35] ScottK: thanks [14:35] doko_: Does your gcc4.4 ia64 fix mean we can drop the ia64 specific work around in our next qt4-x11 upload? [14:36] hi people, if I configure GNOME proxy settings with auth. user and password, I can't get http_proxy as http://user:password@host:port, but simply as http://host:port. It is intentional? [14:36] ScottK: yes [14:36] doko_: Thanks. [14:38] mvo, it doesn't help. the problem is really with the global dequotestring in pkgAcqMethod::Configuration [14:39] mvo, we want some config vars to be unquoted and some other not to be. [14:40] mvo, more precisely, not dequoted at this moment. === jamie is now known as Guest14668 [14:44] mvo, couldn't we had an exclusion list of config vars ? And do not apply the DeQuoteString in the Cnf.Set call for those parameters ? [14:46] pecisk, hi, I can't reproduce. [14:48] jibel: I think that will work, could you attach a dump of the debug output please to the bugreport (from the pkgworker). I wonder if we can not quote some stuff specially so that on unquote its fully restored or have a different unquoe that ensures only what was quoted before gets unquoted. it sounds like this really what we want, that quote/unquote are fully reversible [14:49] jibel: proxy problem? [14:50] pecisk, sorry, I cannot reproduce the proxy problem [14:50] jibel: it gives you correct http_proxy value? [14:51] pecisk, set a proxy with user/pass. Open a new terminal and echo $http_proxy -> http://user2:p%40ssword@localhost:3128/ [14:51] strange [14:51] doesn't work for me [14:52] pecisk, file a report, I'll have a look. ubuntu-bug gnome-control-center. Thanks. [14:53] PlyBuk, err... Keybuk, how are things going with plymouth? [14:56] tseliot: stalled for the moment [14:57] Keybuk: can I have access to that patch/commit of yours which makes sure that deactivate and quit work correctly (just for testing locally), please? [14:57] tseliot: why not just pull from the branch and build that? [14:58] Keybuk: well, what kind of problems are you facing (so that I can be prepared)? [14:59] oh, just irritating messages on boot ;-) [14:59] I think I can deal with that [14:59] ;) [14:59] it may not show a splash screen also [15:00] now you're making it too easy for me to blame it on your work if my patch messes things up [15:00] :-D [15:00] jibel: could you please have a look at https://bugs.edge.launchpad.net/ubuntu/+source/apt/+bug/130289/comments/4 ? [15:00] Ubuntu bug 130289 in apt "Encode any ":", "@" or "/" within the user and password field in proxy settings." [Medium,In progress] [15:01] jibel: unless I'm missing something it should help [15:02] jibel: actually I guess QuoteString() should always quote % to ensure its reversible on DeQuoteString() === Mamarok_ is now known as Mamarok [15:05] mvo, log attached to the bug report. === dendro-afk is now known as dendrobates [15:05] lemsx1: thanks, this looks pretty widespread, I've cranked up the priority [15:09] cjwatson: thnx [15:10] jibel: thanks! with or without the patch [15:10] ? [15:10] cjwatson: sorry. i didn't see the text about the Lucid project... I'll follow that next time ;-) [15:11] cjwatson: wtf is Lucid editor... lol good one [15:12] cjwatson: I have another system that I ran into issues and I was able to fix the problem. it's an OQO (mobile device). I needed to blacklist the viafb driver. Is this something that I should file as a bug too? (hardware specific issue) [15:12] lemsx1: err, I'm not a kernel hacker so I don't know [15:12] lemsx1: I imagine you should file a bug if you're having problems [15:13] pitti: when you get a chance can you review 522509 please? thanks [15:14] cjwatson: against what package? or just Lucid in general? [15:14] lemsx1: the linux package in Ubuntu I assume, since it sounds like a kernel bug [15:40] bdrung: I think java is b0rken and that breaks eclipse: http://pastebin.com/9D3VEsXD (Just started a new project, wrote 1 line of code, pressed ENTER, crash) [15:42] sebner: result of "xulrunner --gre-version"? [15:42] bdrung: 1.9.1.8 , I read the bug report but it doesn't crash at startup, I can create a new project without problems, it crashes when writing code [15:43] sebner: it crash only on startup if the welcome page should be loaded, but it's the same issue [15:44] bdrung: aye, I'm seeing you are already working on it?! It's not urgent so I can wait some hours/days for the update. Else I would just have to use new xulrunner right? [15:46] sebner: launch eclipse with "eclipse -vmargs -Dorg.eclipse.swt.browser.XULRunnerPath=/usr/lib/xulrunner-1.9.2" [15:49] bdrung: works :) [15:49] sebner: k, i was right :) [15:51] bdrung: :) I guess you just need to change xulrunner-dev (>= 1.9.1.3-2) ? [15:52] sebner: nope. it was build against xulrunner-1.9.2. [15:52] bdrung: ohohoho [15:53] bdrung: so you have to hardcode xulrunner 1.9.2 in Depends? [15:54] seb128, kees: mind eyeballing http://bazaar.launchpad.net/%7Eubuntu-desktop/gnome-keyring/ubuntu/annotate/head%3A/debian/patches/04_clean_session_keyring.patch ? [15:54] sebner: nope - we B-D on xulrunner-dev, which pulls xulrunner-1.9.2 in lucid. [15:55] it's dynamically [15:55] bdrung: oh, Ic. Sorry for the noise, I'm not really up to date what uses and pulls in what :D [15:55] seb128, kees: I don't have a session.keyring, so I created a fake one, and tested in various situations (within session so far; building package and doing end-to-end testing with a full reboot now) [15:56] sebner: eclipse's xulrunner detection is totally fucked up. so we have to set it in our wrapper script. [15:57] bdrung: heh, wondering, how xulrunner can break eclipse when writing java code O_o [15:57] pitti, ok, will have a look in a few minutes when I'm done with what I'm doing now [15:58] seb128: cheers (not that urgent, I'll do more tests and then do other stuff) [15:59] sebner: type "System." and a dropdown list will open and a "tooltip" (rendered with xulrunner) will explain its details [15:59] bdrung: ah. now I understand, thx :) [16:00] bdrung: maybe they should switch to webkit :P [16:00] slangasek: so, I'm thinking, let's ship lucid without a splash screen? what do you think :p [16:00] sebner: i doubt that this will help [16:01] heh [16:01] Keybuk: +1 :-P [16:02] sebner: except webkit will be rewritten in java :P [16:03] bdrung: urgh, java ... xulrunner != java too [16:03] bdrung, http://twitter.com/migueldeicaza/status/10930739689 [16:03] Keybuk: well, at least I'll be in London for the release sprint already and will be able to attend your funeral :) [16:05] directhex: hää? webkit-sharp!?! [16:06] webkit-sharp and then jwebkit :D Eclipse 6 with jwebkit [16:06] slangasek: I think that, ultimately, as Release Manager, all failures are your fault [16:06] Captain is responsible for the actions of his Crew, and all that :p [16:07] Keybuk: may the grass be green above and the worms not that hungry beneath :) [16:08] Keybuk: needs to go down with the ship also iirc [16:09] The Captain goes down with the ship! [16:09] Vice-captain (Keybuk) too of course :P [16:10] Keybuk: I don't think I'm cut out for the navy :P [16:10] I'm just the cabin boy === dendrobates is now known as dendro-afk [16:10] is it a release goal to let the ship go down in ten seconds? ;) [16:10] Keybuk: roger that [16:18] cjwatson: I'm having a dpkg-source failure [16:18] and it keeps whining to me about v3 [16:19] dpkg-source: info: use the '3.0 (quilt)' format to have separate and documented changes to upstream files, see dpkg-source(1) [16:19] dpkg-source: unrepresentable changes to source [16:19] etc. [16:19] Keybuk: it means you've modified a binary file. [16:19] hyperair: how does v3 help here? [16:19] Keybuk: 1.0 only supports textual changes [16:19] Keybuk: v3 consists of multiple tarballs. [16:20] debian.tar.gz, orig.tar.gz [16:20] v1 consists of diff.gz, orig.tar.gz [16:20] the diff.gz can't contain binary changes [16:20] so you'll have to uuencode whatever it is you're doing [16:20] right [16:20] or use 3.0 [16:20] how can debian.tar.gz contain binary changes [16:20] is it just unpacked over the top? [16:20] i think so [16:20] not sure [16:20] it can't [16:20] oh it can't? [16:20] it only touches stuff in debian/ [16:20] well at least debian/ can contain a binary file [16:20] you can have binaries in there though [16:20] unlike in v1 [16:21] Note that the debian tar‐ [16:21] ball must contain a debian sub-directory but it can also contain binary [16:21] files outside of that directory (see --include-binaries option). [16:21] oh, cool [16:21] Laney: dpkg-source(1) disagrees with you [16:21] ok... [16:21] so how do I make one of these source packages? [16:21] Keybuk: debian/source/include-binaries is your friend [16:21] echo "3.0 (quilt)" > debian/source/format [16:21] and dpkg-source handles the rest [16:22] oh you'll have to disconnect your patch system from rules [16:22] and control [16:22] dpkg-source will take care of that [16:22] * cjwatson digs up an example [16:22] and only uses quilt [16:22] I don't have a patch system [16:22] is there an equivalent to lshal -m (for monitor) since the halsectomy? [16:22] (there is no debian/patches) [16:22] cr3: udevadm monitor [16:22] you don't need one [16:22] looks like you just put the filename of the binary in debian/source/include-binaries [16:22] Keybuk: thanks [16:22] so, for openssh [16:23] $ cat debian/source/format [16:23] 3.0 (quilt) [16:23] $ cat debian/source/include-binaries [16:23] debian/ssh-askpass-gnome.png [16:23] that's only in bzr, I haven't uploaded that yet [16:23] can I put wildcards in there? :p [16:23] browser-history has one too [16:24] Keybuk: try it and see =p [16:24] :D [16:24] I don't think so [16:24] you can build with --include-binaries once, and it'll dump them all in there [16:24] you can probably even put that in debian/source/options [16:25] any idea how to make pristine-tar not convert the bz2 to tar.gz ? [16:25] if you're completely confident that your clean rule works :) [16:27] I used to be able to troubleshoot brightness issues by echoing 50 (for example) into /proc/acpi/video/.../brightness, but one netbook on which the brightness keys don't work returns: write error: invalid argument [16:29] that's probably why the brightness keys don't work, then [16:29] Keybuk: I'll try to rebout with noacpi [16:30] mvo, with the patch [16:42] <\sh> UDS-M approved by company , means we are coming :) [16:44] w00t [16:46] <\sh> well...one week sponsored by our employer [16:52] pitti: reading it now... [16:54] pitti: looks good to me. [17:00] kees: thanks === deryck is now known as deryck[lunch] [17:08] pitti: is there a bug report about fglrx and jockey already? [17:08] tseliot: for what? [17:08] pitti: to be able to install it again [17:08] tseliot: ah; I'm not aware of one [17:08] do we care about fglrx anymore? [17:09] Keybuk: yes, in some cases we do [17:10] pitti: ok, I'll look for it and file a new one if necessary (and assign it to me) just to keep track of things [17:10] tseliot: thanks [17:10] np === sconklin is now known as sconklin-lunch [17:11] pitti: re bug 546436, it's esata (I commented on the bug too) [17:11] Launchpad bug 546436 in udisks "Does not detect LUKS storage device as hotpluggable" [Medium,New] https://launchpad.net/bugs/546436 === MacSlow|lunch is now known as MacSlow|capoeira [17:11] kees: ah; itz kernel bug then; AFAIR I saw a similar one, I'll dupe it [17:12] pitti: how is it a kernel bug? I didn't think it could distinguish between "internal" and "external" for esata, so it was simply the sudden appearance that should trigger the mount? [17:12] tseliot: you mean this? bu 494699 [17:12] bug 494699 [17:12] Launchpad bug 494699 in fglrx-installer "Does not support current Lucid kernel (2.6.32) or xserver (1.7)" [Critical,Fix released] https://launchpad.net/bugs/494699 [17:13] tseliot: ups, sorry [17:13] tseliot: misunderstand [17:13] ah, that was bug 153768 [17:13] Launchpad bug 153768 in hal "External SATA (eSATA) removable disk detected as system-internal" [Unknown,Confirmed] https://launchpad.net/bugs/153768 [17:13] primes2h: I guess this one would be more like it: bug #496225 [17:13] Launchpad bug 496225 in jockey "Jockey cannot install fglrx driver" [Undecided,Confirmed] https://launchpad.net/bugs/496225 [17:13] kees: well, what is "sudden"? [17:13] tseliot: sure ;-) [17:14] kees: did that ever work? the original bug (153768) is ancient [17:15] kees: so, I don't think we'll ever be able to automount esata drives which are already plugged in during boot, but I think we can do something in udisks for stuff that got plugged in at runtime [17:17] does it matter when it is detected? either way it should show on the places menu shouldn't it? [17:17] Keybuk: any ideas on this? bug #532436 [17:17] Launchpad bug 532436 in nvidia-graphics-drivers "nvidia driver sometime does not load at boot" [Undecided,New] https://launchpad.net/bugs/532436 [17:17] or can I assign it to plymouth? [17:18] pitti: yes, worked in jaunty and karmic fine [17:18] pitti: right, I'm plugging it in at run time [17:18] kees: I'd really be interested in a devkit-disks --monitor-detail and udevadm monitor -e --udev logs from karmic when you plug in an esata drive [17:19] I don't see how this should ever have worked, TBH, but maybe there was something funky [17:21] tseliot: sounds like a kernel issue [17:21] "driver does not load" => linux [17:22] Keybuk: driver loads if plymouth is uninstalled => plymouth, no? === beuno is now known as beuno-lunch [17:22] no [17:22] still linux [17:22] heh [17:23] Keybuk: can you comment on that in the bug report, please? [17:23] if it's the glx driver, it's in the right place [17:23] pitti: hm, will that work off a karmic liveCD? [17:23] kees: yes, it should [17:23] Keybuk: ok, thanks [17:23] pitti: okay, I'll give it a shot. [17:23] kees: cheers [17:23] all he's done by uninstalling plymouth is change a race [17:24] what he's describing is "driver doesn't load sometimes, but does if I restart sometimes, add sleep, etc." [17:24]