/srv/irclogs.ubuntu.com/2008/12/07/#ubuntu-motu.txt

* jdong would like to point motu-sru and those interested in SRU's in general to read pitti's comment: https://bugs.edge.launchpad.net/ubuntu/+source/transmission/+bug/295040/comments/1400:19
ubottuLaunchpad bug 295040 in transmission "multiple download works strangely" [High,Fix released]00:19
jdongI agree with the message wholeheartedly; it doesn't matter if something's called 1.34-0ubuntu1.1 or 1.35-0ubuntu0.1.... it's all about what it does and how it fits with the goals of SRUs00:20
=== Elbrus is now known as Guest52612
=== paul__ is now known as Elbrus
ScottK-laptopjdong: Sure thing.  Look at what clamav we released Intrepid with and what it has now ...00:36
* jdong nods00:36
jdongI'd like to do the same thing for VLC, but am wading through quite a bit of unnecessary or "unnecessary" packaging changes since 0..9.400:36
jdongthe latter definition of "unnecessary" is probably 'it fixes some obscure bug I don't understand yet'00:37
ScottK-laptopfixes/causes, but sure.00:37
ScottK-laptopIt is VLC after all.00:37
jdongmeh I don't expect VLC users to find little bugs here and there to be a surprise anyway :)00:38
jdongor a security problem every other week...00:39
jdongso far thank goodness -fstack-protector has simply kept the bugs down to DoS annoyances.00:39
__ironhi00:40
__ironjust a question to mailinglist00:40
__ironis it desired to introduce himself on mailinglist ?00:43
jmarsdenWhat is the rationale for there being an application that has source packages in Universe, but apparently no binary packages according to rmadison?  Is this "normal"?  (I'm looking at argus )01:16
* ScottK-laptop looks01:17
ScottK-laptopjmarsden: Try rmadison argus-server01:18
jmarsdenAh... OK.  Thanks.01:18
ScottK-laptopThe rationale is rmadison only looks up the package name you give it.  If the binary is named differently, it doesn't show up.01:18
jmarsdenOK.  There's no way to search on partial package names (other than apt-cache search) ?01:20
* jdong curses skype01:23
jdong[   81.368501] type=1503 audit(1228608776.425:11): operation="inode_permission" requested_mask="::r" denied_mask="::r" fsuid=1000 name="/sys/devices/system/cpu/" pid=6648 profile="/usr/bin/skype"01:23
jdongplease don't touch that....01:23
jdong[   81.780716] type=1503 audit(1228608776.834:12): operation="inode_permission" requested_mask="r::" denied_mask="r::" fsuid=1000 name="/home/jdong/Private/.mozilla/" pid=6653 profile="/usr/bin/skype"01:23
jdongor that either.01:23
jdongfortunately, it doesn't crash when I don't let it figure out what kind of CPU I have, my /proc/net/dev list, or my Mozilla profiles.01:23
emgentsomeone know in what package is Gtk.pm ?03:00
emgentah found.03:01
jdongemgent: as a long-time contributor it's probably in your interest to keep apt-file around and updated on your machine :)03:04
emgentjdong: yeah found it03:04
emgentbut my perl script was a little issue03:05
emgentheheh03:05
emgentnow i found it. and i fixed perl script too.03:05
keesScottK-laptop: I wonder if we could package clamav in ubuntu to take advantage of upstart so that clamd-killing-DoS vulns just cause it to restart?03:07
jdongkees: do we have a set of packaging conventions yet how to or if to use upstart event files?03:08
jdongI'd like to see more upstart work done :)03:08
jdong*pokes UDS people*03:08
keesjdong: all i have is examples from Keybuk's blog.  :P03:09
jdonghaha :)03:10
jdongthat's a bit out of date-ish too :)03:10
jdongthe job format AFAIK is due to change soon-ish03:11
jdonghence why I wanted to know what's the big-picture roadmap for adopting Upstart03:11
keesme too.  :P03:11
keesI'll ask him at UDS03:11
jdongawesomeness03:11
jdongblame it on me if necessary, I just want to see more Upstart in jaunty!03:11
keeshehe03:12
ScottKkees: Interesting idea.03:54
=== _Nicke_ is now known as Nicke
zerwasI don't know if this is the right place for my question - i will try it: Would it be O.K. to have a package of a p2p system (namely http://freenetproject.org ), that has an integrated update system and let it use this update system?11:30
lagait's usually not OK11:33
lagaAFAIK :)11:33
zerwaslaga> the problem is that freenet always needs the latest builds to stay usable. and it's an anonymous network so every time a new build would come out, one would have to update through the package management, which is non-anonymous11:34
RainCTzerwas: how often are there updates?11:35
zerwashm, depends. in the last time 3 times a week, sometimes a few weeks without updates11:36
tombeeHow would I go about actually getting involved with development?  I've read the wiki about MOTU :)11:37
zerwasit's still Freenet 0.7 though11:37
lagazerwas: and how is getting the .deb initially "anonymous"?11:38
zerwasit isn't, that's the problem11:39
RainCTtombee: look for a bug and fix it :)11:39
tombeeRainCT: I'm really a beginner to development :P looking to learn11:40
tombeeStruggling to find an IDE I even like, used vim mostly before, but now my projects are becoming more complex it becomes more difficult to handle the files this way11:41
RainCTtombee: do you want to contribute with packaging or coding?11:41
tombeeIdeally coding, but I'm not sure my skills would be immediatly up to scratch.11:42
RainCTtombee: for coding the best would be to choose a project to which you want to contribute and get involved there. MOTU is about packaging11:43
tombeeHehe I see ok :) Was just pointed in this direction with the 'beginning to get involved', thanks anyway :)11:44
zerwaslaga> so, what do you suggest in which way i should build the package?11:45
lagazerwas: well, how is the problem "i can't download the package anonymously" different from " i can't update the package anonymously"?11:46
zerwaslaga> what i meant was: if one can see publicly when someone installs new build it might reduce the anonymity. But this problem is only second-tier. a package that would not use the update mechanism would be useless one week after jaunty got released11:49
lagazerwas: yeah. maybe you can get a standing exception11:51
zerwasthat would be nice. i haven't found a better solution up to now11:53
lagai'd suggest asking on the MOTU mailing list, explaining the problem etc.11:56
zerwasokay. am i allowed to write on this ml without being a MOTU myself?11:57
lagayes, but you need to subscribe AFAIK11:57
lagaif you don't subscribe, your mails will be moderated11:57
lagai think ;)11:58
zerwask. thank you very much for your help so far :)11:58
anakronHi ALL12:02
anakronHi RainCT12:02
RainCThi :)12:02
anakron:) i found another one12:02
RainCTyeah, I've seen12:02
anakroni have one question12:05
anakrona tool called nss-updatedb is designed to work with libpam-ldap and libnss-ldap12:05
anakronbut they dont appear in suggests or depends12:06
anakronand it cant work fine without these libs12:06
anakronso, i want to make a patch for it :) changing dependencies+12:06
anakroncan you help me?12:06
RainCTanakron: It's pretty much like your first patch. But I'd recommend asking in #ubuntu-server before you touch this (to be sure that the change is appropriate)12:07
anakronok :)12:07
anakronRainCT, I can12:28
DRebellionAre there plans to backport lintian 2.0.1 from jaunty-> intrepid? Currently intrepid is using 2.0.0.13:21
DRebellionSorry, I meant to say that the jaunty version of lintian is 2.1.0, not 2.0.113:57
pmjdebruijnlo13:58
pmjdebruijnI'm trying to package something which uses the CC-BY-SA-3 license, but it's not included in /usr/share/common-licenses13:58
pmjdebruijnhow should I handle this?13:58
RainCTpmjdebruijn: include it in debian/copyright13:59
azeemyou'll have to quote the whole license in debian/copyright then13:59
pmjdebruijnverbatim?13:59
RainCTyes13:59
pmjdebruijnhmmm13:59
pmjdebruijnisn't it about time CC licenses got included in common-licenses?13:59
ScottK-laptoppmjdebruijn: No.  There isn't a huge amount of software licensed under CC.  The only real reason to add stuff is to save significant space in the archive.14:01
pmjdebruijnok14:01
=== nhandler_ is now known as nhandler
DRebellionsebner, Thanks for taking a look at cifer; I've fixed the issues you raised: http://revu.ubuntuwire.com/details.py?package=cifer .14:36
sebnerDRebellion: np, changes look good so far, in the evening I'll do a second review and then maybe advocate it14:38
pmjdebruijnhmmm, pbuilder included with intrepid, doesn't do jaunty...14:45
=== nhandler1 is now known as nhandler
DRebellionpmjdebruijn, enable intrepid-backports15:00
pmjdebruijnDRebellion: I just did15:01
DRebellionpmjdebruijn, and it works now?15:02
pmjdebruijnoh wait, I need to update debootstrap as well15:03
DRebellionpmjdebruijn, yep15:03
pmjdebruijnit works now15:04
pmjdebruijnone of the copyright holders of a package I'm trying to make, does not publish his email adres... should I include a link to his "contact page/form" online?15:14
rjune_I would say ask him15:23
lidaobinghelp review iptux 0.4.2-0ubuntu1(new package): http://revu.ubuntuwire.com/details.py?upid=418615:58
=== fta2 is now known as fta_uds
fta_udshi17:16
RainCThey17:27
RainCThas UDS already started?17:27
fta_udsnope17:27
iulianTomorrow.17:27
=== fta_uds is now known as fta
ftasorry :)17:28
sebnerfta: haha, one day to early :P17:28
RainCTiulian: I know.. I'm asking for the case there's some timezone weirdness :P17:28
RainCTfta: heh17:28
ftathat's the one nick for all channels effect ;)17:28
sebnerRainCT: well California is -9h so it's also sunday there :P17:29
StevenKIt's Sunday, 9:30am currently.17:31
StevenKAnd California is -080017:31
RainCTsebner: ^^17:31
rjune_fta: is ogra about?17:32
sebnerStevenK: well I'm not from the UK so -9h :P17:32
StevenKsebner: It's -19h from where I live17:32
* RainCT slaps sebner :P17:33
ftalol17:33
RainCTlet's agree that it is "Pacific Standard Time (PST) -0800 UTC"17:33
RainCT:P17:33
* sebner hides17:33
sebnerRainCT: math is calling :P17:34
RainCTsebner: you can't hide from them, Google is watching you :P17:34
RainCTmeh17:34
=== nhandler_ is now known as nhandler
=== nhandler_ is now known as nhandler
sebnerNCommander: aloha, is it allowed to add license headers with a patch(-system)?18:45
NCommandersebner, w.r.t. to what?18:45
sebnerNCommander: we don't accept source files with no license headers which means new upstrem release is necessary, I'm wondering if you can work around it with a patch (debian/pathes) which adds the license headers to the source files18:47
JontheEchidnaI wouldn't think so unless they have been added in an upstream svn/git/whatever18:47
NCommanderI would REJECT if I was an archive admin18:47
NCommanderAs a rule of thumb, unless there is a technical reason or a file is uncopyrightable, the headers should go in it18:47
sebnerkay18:47
NCommanderIf upstream is allowing you to add the headers and will add it in SVN/GIT18:48
NCommanderRepack the source tarball using DFSG, then note the reason iN README.source, and (optionally) in debian/changelog18:48
NCommander*DFSG versioning18:48
sebnerNCommander: I see, there is a bad boy with no license headers and after my complains (revu) he added them himself -.-18:49
sebnerNCommander: where bad boy = upstream too18:49
Elbrusdoes anybody now documentation about how to do bootstrapping for packaging? I think bug 67544 needs that, and I can look somewhat into that if I find documentation.18:49
ubottuLaunchpad bug 67544 in fpc "PPC build of fpc fails" [Undecided,Confirmed] https://launchpad.net/bugs/6754418:49
Elbrusit works in debian18:50
Elbrus(apparently18:50
sebnerElbrus: NCommander is the FTBFS guy :P18:50
sebneranyway thx for your help NCommander and JontheEchidna18:51
NCommanderfpc has issues on PowerPC last I saw18:51
ElbrusNCommander: but it looks like it works for Debian: http://buildd.debian.org/pkg.cgi?pkg=fpc18:52
jdongAmperage: -85418:53
jdongnow what would it take to mae Linux do that too?18:53
Elbrusjdong: you can make money that way,18:54
jdongI am guessing the equivalent Linux solution is get a bigger battery.18:55
NCommanderlol18:55
jdongAFAIK OS X isn't even a tickless kernel... I'm curious as to how it pulls off more than a 2W improvement compared to Linux18:56
jdongand that's after I've optimized my Linux setup the best i could; ASPM/ALPM18:56
NCommanderjdong, OS X has a lot of the default services disabled to prevent things from paging out and thus it can spin down the HDD18:57
NCommanderthus the better power usage curve18:57
jdongdisk idling only results in about 0.4W savings here.18:57
jdongi.e. full spindown18:57
jdongI boot Ubuntu into casper to RAM18:57
jdongthe disk is spun down 45s into initial boot18:57
jdong11.5W wifi on, 11.2W wifi off18:58
jdongon OS X when I turn wifi off the battery meter jumps into the 6:30 arena18:58
jdong22 minute uptime, currently at 92%18:59
jdongtat's roughly 300 minutes total runtime including the bootup flurry18:59
jdongso the battery meter isn't lying right now18:59
DRebellionsebner, hi! About the RPATH issue with cifer: neither i or my co-developer are running amd64 machines, so we can't use chrpath to test amd64 binaries from my ppa. Am I right in thinking that a simple call to chrpath like this, << chrpath --delete $(CURDIR)/debian/cifer/usr/bin/cifer >> will get rid of the RPATHs and solve the issue?19:06
DRebellionI guess I need to build-depends on chrpath as well?19:07
sebnerDRebellion: right =)19:08
DRebellionsebner, great : )19:08
sebnerDRebellion: you don't need an amd64 machine. you can download the amd64 binary from your ppa, unpack it and when you run chrpath on the binary it'll complain19:08
DRebellionsebner, exactly19:08
DRebellionso i can't --list them19:09
sebnerDreamThief: list?19:09
DRebellionsebner, I downloaded from my ppa, extracted, and ran << chrpath --list ./usr/bin/cifer >>. It said: simrun@simrun-desktop:~/cifer-amd64/usr/bin$ chrpath --list cifer19:09
DRebellion`cifer' probably isn't a 32-bit LSB-first ELF file.19:09
DRebellionelf_open: Exec format error19:09
sebnerDRebellion: that means it has a rpatch issue, if you run chrpath on an i386 binary it won't complain19:10
DRebellionsebner, oh right. Ok then.19:10
sebnerDRebellion: and you don't need the --list option :)19:10
DRebellionsebner, btw, I don't need to check this build with DH_VERBOSE=1 because you already have?19:12
sebnerDRebellion: yes, it was just a side note for the future19:12
DRebellionsebner, ok, fixed those issues, now building. I'll let you know when its in revu again.19:13
sebnerDRebellion: ok, you maybe again want to upload it to your PPA and recheck the amd64 binary19:13
DRebellionsebner, yep, just a slight pain as i'll have to fork into two packages to bump the version number for the ppa :(19:14
bluefoxicydid somebody pull Quake from the repos?19:35
DRebellionsebner, ok, the ppa build is published. Downloaded the amd64 deb, extracted, ran chrpath:19:43
DRebellionsimrun@simrun-desktop:~/cifer-amd64/usr/bin$ chrpath cifer19:43
DRebellion`cifer' probably isn't a 32-bit LSB-first ELF file.19:43
DRebellionelf_open: Exec format error19:43
DRebellionare you sure chrpath built for i386 can read amd64 binaries?19:43
AggroI got a package in Debian called Cppcheck and I would like to package it for Ubuntu. Is there some easy way to re-using existing work, or do I need to read Ubuntu's manuals about packaging and re-do everything (possibly reusing existing files?)19:57
AggroI'm not afraid of work, just want to make sure before I start working that I'm not doing something stupid ;)19:57
nhandlerAggro: If it is in Debian, it will get synced into Ubuntu19:57
Aggronhandler: I don't need to do anything? It is automated process?19:58
Aggronhandler: I am the package maintainer for it in Debian19:58
nhandlerUp until DIF it is automated.19:58
Elbrusnhandler: DIF= Debian import freeze???19:58
nhandlerCorrect Elbrus19:59
nhandlerAfter that, you have to manually request a sync19:59
AggroI made a bug report: https://bugs.launchpad.net/ubuntu/+bug/30604219:59
ubottuLaunchpad bug 306042 in ubuntu "please sync package cppcheck from debian Unstable" [Undecided,New]19:59
Aggro"please sync package cppcheck from debian Unstable"19:59
nhandlerAggro: When did you upload it in Debian?20:00
Aggronhandler: It was accepted to Unstable yesterday I think20:00
Aggronhandler: But it has been in the queue for 3 weeks or so20:00
nhandlerAggro: You don't need a bug report. The package should make its way into Ubuntu within the week20:00
AggroI already made a bug report, should I undo it?20:01
AggroThe wiki page was a bit misleading20:01
Aggrohttps://wiki.ubuntu.com/UbuntuDevelopment/NewPackages20:01
nhandlerAggro: I would just close the bug report. There is no need for it20:01
Aggro"For packages in Debian, but not in Ubuntu file a bug..."20:01
nhandlerWhere is that?20:01
Aggronhandler: The url was given in my previous message20:02
nhandlerAggro: Yeah, the wiki page should probably be modified. I'll change it now20:02
crimsunAggro: likely will be synced within the next 72 hours20:03
AggroBut the progress is automated? All are Debian packages in Ubuntu? I had a feeling that there were less in Ubuntu?20:03
crimsunAggro: just about every source from Debian is in Ubuntu.  There are a few that are blacklisted.20:04
crimsunAggro: yes, up until DI, the process is automated (and thus the progress, too, I suppose).20:04
Aggronice20:04
crimsunDIF*20:04
Aggronhandler: If I close the bug, what status should I set?20:06
nhandlerAggro: I just updated the wiki page20:06
Aggroinvalid?20:06
DRebellionsebner, i'm going to go ahead and upload the updated package to revu, as i can't test if using chrpath was a success. if there is still a problem, let me know.20:06
nhandlerYeah, invalid is fine20:06
Elbruswhen do packages propogate from jaunty-released to jaunty?20:06
Elbruss/released/release20:07
geserjaunty-release?20:07
nhandlerAggro: Could you take another look at the wiki page and let me know if it is more clear now?20:07
Elbrusthat's where winff is https://launchpad.net/ubuntu/+source/winff20:07
Aggronhandler: You might want to clarify when is Debian Import Freeze (DIF), how does the person who is reading that know.20:08
AggroE.g. I have no idea when it is20:08
nhandlerAggro: Good idea. I guess I can link to the Release Schedule for Jaunty20:08
Aggronhandler: What about if the person is ready to wait until the freeze is over?20:09
Aggronhandler: Then there is no need to submit a bug either?20:09
crimsunElbrus: that is jaunty proper.20:09
nhandlerAggro: DIF is in effect until after the official release.20:09
ElbrusAggro: right, but then it will take an extra release before it is available20:10
nhandlerIf they are willing to wait until the next Ubuntu release, no bug is needed20:10
geserElbrus: I don't know if "release" has any meaning there20:10
Elbrushttp://packages.ubuntu.com/search?keywords=winff doesn't show anything...20:10
geserElbrus: winff is in DEPWAIT: https://launchpad.net/ubuntu/jaunty/+source/winff/+builds20:10
Elbrusaha.20:10
sebnerDRebellion: ok, and yes I'm sure that chrpath built for i386 can read amd64 binaries though it seems it wasn't sucessful in your case20:11
sebnerhi geser =20:11
sebner=)20:11
geserHi sebner20:11
geserElbrus: and the successfull builds are in the NEW queue20:11
Elbrusgeser: you mean everything except  i386 and amd64?20:11
geseryes20:12
DRebellionsebner, hmm, that's strange... anyway i've fixed all the issues it seems then.20:12
Elbruslazarus-ide doesn't (or shouldnot) exist on ppc20:12
geserif winff exist only i386 and amd64 you just need to wait till it gets NEWed20:12
nhandlerAggro: The page now mentions when DIF is for jaunty20:12
Elbrusgeser: well, it will build on all systems (I hope) where lazarus-ide builds... (Arch is any)20:13
Aggronhandler: Should the timeline (week)? be added there also?20:14
Aggronhandler: How long it should take before it is not normal20:14
nhandlerAggro: I don't think the week will mean anything to most people20:14
nhandlerI think a date is fine20:15
nhandlerIf they really care about it, they can click on the link to the release schedule20:15
Aggronhandler: I mean the "it should be there within 1 week"20:15
nhandlerAggro: The one week thing isn't exact. It could be less/more than a week.20:16
Aggronhandler: So it is not a cron script? ;)20:17
AggroBut it should be obvious enough not to fool me as it is now20:17
nhandlerAggro: It might be a cron script, I honestly cannot remember how they handle the automated syncs.20:17
Aggronhandler: Thanks for your help, this was much simpler than I thought. I'm glad I didn't start making a package of it again ;)20:18
nhandlergeser: Any idea when you will be able to review my application?20:18
nhandlerAggro: You are welcome. Glad to have helped20:18
=== `Chris is now known as Chris60292
=== Chris60292 is now known as `Chris
=== `Chris is now known as dotCYM
jdongsiretart: what are your opinions on simply copying VLC 0.9.8a into intrepid-security?23:35
jdongI've only briefly looked at the diff and it doesn't look too bad23:36
RAOFHm.  Thinking of SRUs.  Any motu-sru's around?  It'd be nice to have some guidance on bug #28733223:53
ubottuLaunchpad bug 287332 in evolution-sharp "beagle-backend-evolution cant find libedataserver-1.2.so.9" [Medium,New] https://launchpad.net/bugs/28733223:53
* jdong looks23:55
jdongjeez23:57
jdonggonna need a moment to ponder this one :)23:57
RAOFBasically, there are two options: new upstream version, breaking API and ABI, or trying to patch in support for new e-d-s without breaking either.23:57
jdongprimary rdepends seem to be Beagle and Tasque?23:58
RAOFAnd gnome-do-plugins, and one other.23:58
RAOFgfax, that's it.23:59
jdongmmmkay and have we tested whether or not those things migrate properly, like build against the new evo-sharp?23:59
RAOFNot yet, actually.  I'll do that now.23:59
jdongI see beagle and tasque tested23:59
jdongPedro seems to have done those23:59
RAOFNifty.  I'll do gnome-do-plugins.23:59
jdongif the other two work, I'll be positive on the idea :)23:59

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!