/srv/irclogs.ubuntu.com/2008/03/23/#ubuntu-devel.txt

slangaseklamont: are your permissions on /var/lib/misc/shadow.db different than your perms on /etc/shadow?  Does /sbin/unix_chkpwd have appropriate s[ug]id bits for reading both?00:20
lamontls -l /etc/shadow /var/lib/misc/shadow.db00:21
lamont-rw-r----- 1 root root   1454 2008-03-17 12:28 /etc/shadow00:21
lamont-rw-r----- 1 root shadow 8192 2008-03-17 06:01 /var/lib/misc/shadow.db00:21
lamontls -l /sbin/unix_chkpwd00:21
lamont-rwxr-s--- 1 root shadow 21216 2007-10-01 12:49 /sbin/unix_chkpwd00:21
lamontthat's in the mode where I get to login on another vt and kill the screensaver.. :-)00:22
lamontslangasek: oh, and I've only confirmed this on gutsy, btw... :-(00:23
StevenKlamont: I note my /etc/shadow isn't root:root00:24
lamontStevenK: what is it?00:25
StevenKlamont: root:shadow, like /var/lib/misc/shadow.db00:25
lamontinterestingly, it's root:shadow on some other boxen... hrm00:25
StevenKIt isn't Gutsy, it's that install? :-)00:26
lamontwell, making /etc/shadow root:shadow didn't change the fact that it wouldn't unlock00:29
* lamont brb00:30
slangaseklamont: er, so why are other users not allowed to execute unix_chkpwd?00:30
slangasekthat's brokez0r00:30
slangasekand not the permissions shipped in the package in gutsy00:31
StevenKHey, yeah. That doesn't help either.00:31
slangasekrather, it's specifically what breaks it00:32
slangasekjdong: what version of hal do you have installed?00:33
lamontmeh.00:37
* lamont goes to close his bug as invalid00:37
slangasekhurray! :)00:38
lamontthere is a certain wrongness in closing one's own bugs as PEBCAK00:40
slangasekso how did the perms end up that way, exactly? :)00:41
lamontiz long storry00:42
lamontnothing like filing a bug to get educated. :-)00:52
FujitsuCan somebody please give back ikiwiki?00:57
brazilhi - mobile-player_0.4_all.deb and mobile-application-service_0.9-5_i386.deb in hardy universe have 10 extra bytes over what the Packages file says, and the md5s don't match. Who should this be reported to?00:58
FujitsuI thought they fixed that.00:58
brazilI'm using archive.ubuntu.com (was using uk at first)00:59
brazilhave run multiple times, checked by hand etc.00:59
FujitsuIt must have unfixed itself, I guess.00:59
Nafalloehrm... isn't uk the same one?01:00
brazilI get different IPs01:00
Fujitsua.u.c has several.01:00
brazilhmm - there's 3 IPs for archive - let me try each01:00
StevenKuk.archive != archive01:00
Nafallooh. uk is gb :-)01:00
Nafallodidn't know that one was mapped :-P01:00
StevenK*.archive.ubuntu.com is, actually01:01
NafalloStevenK: yea, which was why I was surprised uk was mapped to gb :-)01:01
brazil.31 .45 and .46 are all bad01:04
StevenKDo they actually download via apt-get?01:04
brazilI'm using reprepro to build a hardy mirror01:04
astheglorioushey peeps01:17
astheglorioushow's it hanging01:17
brazilso who needs to be told about this to get it fixed up?01:17
asthegloriousI have no idea01:18
adrian2002caso how do i get started developing...im a pretty good programmer, pretty new to ubuntu(a month)...ive looked at the wiki and I still don't know how to get an <experienced developer to help me along>01:27
elkbuntu_Adri2000, #ubuntu-motu is a better place to ask :)01:28
adrian2002caokey-dokey :P01:28
=== elkbuntu_ is now known as elkbuntu
=== YokoZa1 is now known as YokoZar
emgentheya people01:51
=== Pici` is now known as Pici
sladeniwl3945 b0rkage02:40
jdongslangasek: whatever's latest on Hardy as of 48 hrs ago02:53
xhakerHobbsee: could you please reupload your ubuntu-restricted-extras02:54
xhakerHobbsee: now with open-jdk-602:54
slangasekjdong: then you probably have the broken hal installed, please see my follow-up to the n-m bug02:54
jdongslangasek: ah, ok, so I need to catch-22 a new hal :)02:55
jdongI'll deal with that02:55
slangasekyeah, not a nice thing to have hit the archive the day after beta :/02:56
slangaseksladen: your laconic comment suggests the same issue to me...02:57
Hobbseexhaker: why?02:58
xhakerHobbsee: the packages are openjdk-6* sorry02:58
xhakerHobbsee: because every package that was built with icedtea7 is being transitioned to openjdk-602:58
xhakerso you should use openjdk-6 for sanity :D02:58
xhakerthe dependency should be on icedtead-gcjwebplugin02:59
xhakerthat package installs the browser plugin and pulls openjdk-6-jre03:00
xhakericedtea-gcjwebplugin, it's late here.03:01
xhakerright Hobbsee ?03:01
* Hobbsee is looking03:01
jdongslangasek: ok, fair enough, thanks for your insight, I'll get out a USB disc03:02
slangasekit's possible that all of the other wpa2 problems people are reporting right now all come down to the same thing, but I'm not sure because so far I've only gotten confirmation from anyone that hal fixes the "eats my CPU" problem, which not everyone has mentioned seeing03:04
Hobbseeslangasek: how do you use sed to delete a line, and not leave blank space there instead?03:05
xhakerHobbsee: you'll see an upload from doko today, for the new openjdk code drop. there is a bug somewhere to trac the transition, but i don't know the number03:05
slangasekHobbsee: /thingymatched/d03:05
* slangasek worries that his reputation as a sed glutton may have preceded him03:06
Hobbseeslangasek: ah, thanks.03:08
xhakerHobbsee: found it, it's bug 20363603:08
ubotuLaunchpad bug 203636 in openoffice.org "replace icedtea-java7 references with openjdk-6 references" [Undecided,In progress] https://launchpad.net/bugs/20363603:09
xhakerfiled by doko himself :)03:09
Hobbseexhaker: done, thanks03:12
* xhaker hugs Hobbsee03:12
* Hobbsee hugs xhaker back03:12
Hobbseei'm beginning to hate all this java stuff, w.r.t. all the versions.03:12
xhakernow, why is licwidget0 and libcwidget3 installed side by side by default? licwidget0 is a dependency of aptitude and licwidget3 is orphaned :S03:14
Hobbseexhaker: of what?03:15
slangasekbecause aptitude is FTBFS with the new cwidget and needs to be updated in a freeze-appropriate way03:15
Hobbseeheh03:15
slangasekit's libcwidget0 which is supposed to go away03:15
xhakeric :)03:22
Fujitsuslangasek: Have you not pushed out the syncs you've performed yet?04:06
slangasekFujitsu: not yet, no04:09
Fujitsuslangasek: OK, thought it was a bit late for you to not have.04:10
slangaseknot in this time zone? :)04:13
FujitsuHobbsee: Can you please give back ikiwiki?04:16
HobbseeFujitsu: given back04:16
FujitsuHobbsee: Thanks.04:16
FujitsuGr, how'd that fail again.04:18
FujitsuIt succeeded twice in archive rebuilds two weeks apart.04:19
FujitsuHum, shouldn't Priority: required stuff be installed in the buildd chroots?04:46
Fujitsuliblocale-gettext-perl isn't.04:47
slangasekyou should still explicitly build-depend on anything that isn't essential: yes or part of build-essential04:49
slangasek(I think the buildds use a stripped-down package set for this reason, maybe?)04:49
FujitsuAh, thanks.04:50
Hobbseeit's interesting about how a lot of hte ubuntu iso's for the beta are being downloaded by ktorrent.06:16
FujitsuHobbsee: It's because KTorrent was the only graphical torrent client that didn't utterly suck, until Hardy.06:17
HobbseeFujitsu: ahhh06:17
FujitsuThat's why I and most other Ubuntuers I know used it.06:17
Hobbseefair enough06:18
=== \sh_away is now known as \sh
=== \sh is now known as \sh_away
=== doko_ is now known as doko
Methodi'm trying to add a patch to the kernel package but can't figure out how the kernel package works, how do i get a real source package rather than the linux_meta package?09:54
HobbseeMethod: probably #ubuntu-kernel, but it is a public holiday.09:55
=== evalles is now known as effie_jayx
lagacjwatson: where do you keep the preseed files for the alternate disks? is there a bzr repo somewhere?12:25
LoevborgScripts /etc/acpi/resume.d used to be run on resume. Is there an equivalent for this in hardy's hal-system-power-suspend-linux?13:04
mjg59You can add stuff to /etc/pm/sleep.d13:07
Loevborgmjg59: thanks. Any chance that /etc/acpi/suspend.d will get cleaned up or removed? It's really confusing to have several acpi subsystems in the system.13:09
j1mchello, all.  there don't seem to be new images created for xubuntu, but there are no problems being reported as part of the xubuntu daily cd health check.13:11
j1mcis there a reason for this?13:11
j1mchttp://cdimage.ubuntu.com/xubuntu/daily/current13:12
mjg59Loevborg: Comment should probably be added - we won't delete any user-added files, though13:12
Loevborgmjg59: sure, I was talking about a clean hardy install, where these files should not be necessary any more13:13
j1mceven the most recent daily-live builds are from the 18th: http://cdimage.ubuntu.com/xubuntu/daily-live/current/13:15
hunmaathi. can you explain me, what the msgid 'Frobnicate!' of policykit-gnome means? And how is it different from 'Tweak!'?14:12
vykmorodI'm just wondering if anyone can tell me something about Python + OpenGL + Tk/tcl core dunos in FiestyFawn14:17
vykmorod*dumps14:17
persiahunmaat: When one frobnicates, one is adjusting something as one wishes.  When one tweaks, one is typically seeking to improve the results.14:19
hunmaatthanks, I thought something like this. but do not know distinct hungarian words for them :/14:20
hunmaat[I don't now if it makes sense to translate these examples...]14:20
hunmaatk14:20
persiahunmaat: Frobnicate is jargon (http://catb.org/jargon/),  If you translate it, "Tweak" is good.  The distinction is mostly arbitrary.14:22
hunmaatI see, but the next string is 'Tweak!' :(14:23
persiaMaybe translate "Tweak" from English to Hungarian, and simply conjugate "Frobnicate" as jargon?14:24
hunmaatthis might be an ability, but the English knowledge in Hungary is quite poor, so lots of people will have no idea about it's meaning14:27
persiahunmaat: True, but "Frobnicate" isn't really an English word either, and I'm not sure of any other solution.14:28
hunmaatI think I will simply skip the full examples section (~40 string). I can't find it on GUI.14:29
johanbrI'd say "Frobnicate" shouldn't be present in anything user-visible in English either.14:32
hunmaatI haven't seen the source, but based on the filenames I think, this is an unused example policykit implementation.14:33
hunmaatwith dummy actions14:34
sistpotydoko: please give back missinh on sparc, thanks!15:10
=== warp10_ is now known as warp10
sivangdoes anyone remember how to inspect contents of a package using python apt?16:39
lagasivang: there are lots of examples in the python apt package16:40
lagamaybe you can find something there?16:40
sivanglaga: I will try, thank you16:57
jwalHi.  I am looking for some documentation, preferably a risk assessment, for the decision to try and include Firefox 3 in Hardy and what will happen if Mozilla have not released it in time.  Any pointers?17:13
sistpotyjwal: maybe asac can help you there17:14
asacjwal: what for do you need that info?17:18
jwalasac: I'm just curious17:19
=== spacey_ is now known as spacey
asacjwal: there are several compromises involved in this.17:21
asacjwal: not being officially final is not the most important things for us (though its what we have hoped for). Its more important that we ship a product that rocks, while being able to guarantee our users security support throughout the whole hardy support cycle for instance.17:23
asacdoing a major version upgrade in a LTS because it turns out that we cannot provide security support is just not a good idea.17:24
jwalI hadn't even considered that.  Do Mozilla make any promises about how long they will provide security fixes?  (I am now starting to research this on mozilla.org)17:27
asacjwal: 6 month after the next version is released17:28
asacso consider that 3.0 is final in may, we will get a security support for 2.x till december or something17:29
asacjwal: we (distros) entangle to extend the security support, but unless thats proven to work we cannot rely on that17:29
asacwe do that for 1.5.0 (as in dapper) for now17:30
jwalSounds like you are no longer expecting to be able to ship a final Firefox in 8.04 LTS17:30
hunmaatafaik, nor 1.5 was final17:31
asacjwal: you never know ;)17:32
jwalasac: Presumably this has been discussed with the people at Mozilla.  Have they been able to make any promises?17:35
orbisvicismaking sure im doing this right: get source, extract .tar.gz, extract diff.gz, patch -p <diff.gz17:36
asacjwal: they are aware, yes. you never get any promises though17:36
sorenorbisvicis: It might help if you said what you were trying to do. If you're trying to make pancakes, you're not doing the right thing at all.17:36
sorenIf you're trying to extract the source for ubuntu packages in the hardest possible way, you're close :)17:37
orbisvicissoren this is an ubuntu dev channel, its about modifying/patching ubuntu source ; )17:37
mjg59orbisvicis: apt-get source packagename17:38
mjg59Or dpkg-source -x foo.dsc17:38
orbisvicisum unfortunately this is alsa-plugin_1.0.15 for 8.04 while i need it for 7.1017:38
orbisvicisso I dl'd it off the ubuntu website17:39
orbisvicisoh i was so close17:39
jwalsoren: It's about 41 days too late for pancakes.  More likely trying to boil an egg.17:39
orbisvicisthen move the debian folder into the source folder17:39
jwalasac: If, for example, Hardy ships with Beta 5 will the final version be included in the security update stream, the updates stream or the backports stream?17:41
jwalasac: Sorry if this is on the wiki or launchpad somewhere (but I have been unable to find it)17:42
sorenorbisvicis: Just add a deb-src line to your sources.list and do an "apt-get source alsa-plugin/gutsy"17:43
asacjwal: we will track upstream through security17:43
orbisvicissoren, that is a viable option17:43
sorenjwal: Aw, I'm too late for pancakes?17:43
asacjwal: as always17:43
orbisvicisbut theoretically this was works as well ?17:44
sorenorbisvicis: Or dget http://archive.ubuntu.com/ubuntu/pool/main/a/alsa-plugins/alsa-plugins_blahblah.dsc, followed by dpkg-source -x.17:44
sorenorbisvicis: Not necessarily.17:44
sorenorbisvicis: Depends on how the package maintainer applies patches.17:45
orbisvicisoh17:45
orbisvicispatch -pX, or even a different patch program ?17:45
sorenorbisvicis: The .diff.gz might patch stuff outside debian/.17:45
=== Kopfgeldjaeger is now known as PrinzessinZelda
orbisvicisi hadnt thought of that, though in that case i could move the diff file before patching it17:47
sorenorbisvicis: There are quite a few patch management system. If the package maintainer always uses one of those, it's usually not a problem to just move debian/ around. If they don't use a patch management system or just doesn't always do it (it's sometimes troublesome), the .diff.gz will contain diffs against stuff outside debian/17:47
orbisvicisthough i like the dpkg-source -x method >elegant17:47
soren"apt-get source" uses "dpkg-source -x", too.17:48
sorenit just fetches the source first.17:48
sorenAnd yes, dpkg-source -x is the one true way.17:48
orbisvicisheh17:48
orbisvicishow did you get that link ... b/c it shows all the newest packages, what if say i needed the 7.10 version ?17:49
sorenorbisvicis: That one is there as well.17:50
sorenorbisvicis: I just typed it.17:50
=== PrinzessinZelda is now known as TentakelFrau
sorenorbisvicis: Every version from dapper to hardy should be tehre.17:50
sorenthere, even17:50
lagaTentakelFrau: scary.17:50
* asac out for a while17:50
orbisvicisyes17:50
TentakelFrau;d17:50
* orbisvicis ^^ blind17:51
=== \sh_away is now known as \sh
hunmaathas the policykit-gnome developer ever heard about gnome hig?18:20
jwalasac: Thanks for your help18:30
hunmaatThe UIF means strict stringfreeze?18:32
=== TentakelFrau is now known as eMo
lagaeMo: _scary_18:35
eMolaga: only if you don't know the context :)18:36
eMobut, well, ok, it's enough ;)18:36
=== eMo is now known as Kopfgeldjaeger
=== slangase` is now known as slangasek
slangasekasac: does ubufox not use the "Npp" metadata in Packages to figure out what plugin packages to install? bug #177514, icedtea-java7-plugin has been removed from the archive but firefox still won't automatically pick up icedtea-gcjwebplugin in its place19:26
ubotuLaunchpad bug 177514 in icedtea-java7 "firefox 64-bit IcedTea java not working." [High,Confirmed] https://launchpad.net/bugs/17751419:26
=== azeem_ is now known as azeem
=== \sh is now known as \sh_away
=== fta_ is now known as NextHop
=== NextHop is now known as fta
=== fta is now known as NextHop
slangasekasac: do you need any help triaging the information in bug #50214?  the set of permutations is huge, and a fair number of wireless drivers appear to still have the problem; are you expecting all of these to get fixed in the kernel for final?20:46
ubotuLaunchpad bug 50214 in network-manager "can't connect to hidden network" [High,Confirmed] https://launchpad.net/bugs/5021420:46
slangasek(... and if so, does the kernel team know this?)20:46
asacslangasek: help appreciated, i think i forked out individual bugs for specific chipsets i know of21:01
asacfor instance the iwl drivers appear to lack a patch in its own mac80211 module21:01
asacslangasek: we might consider to unmilestone the generic bug.21:03
slangasekasac: I guess I'd like to be sure first that it's empty of any remaining per-chipset problems that haven't been broken out, before unmilestoning it21:13
asacIll look21:14
asacbut thats definitly on my radar21:14
ScottK2asac: Is Bug #204868 on your list too?21:37
ubotuLaunchpad bug 204868 in network-manager "NM can't use wireless networks" [Undecided,Confirmed] https://launchpad.net/bugs/20486821:37
slangasekScottK2: isn't 204868 really 204768 in disguise?21:38
* ScottK2 loooks.21:38
slangaseknice that 145652 is marked as a duplicate of 204868 :P21:38
slangasek204768 was the hal issue21:39
slangasekis this issue reproducible even with current hal?21:39
ScottK2slangasek: No.  Apparently it's not the same thing.21:39
ScottK2slangasek: 868 has a debdiff attached that multiple people claim has fixed nm for them.21:40
ScottK2It does appear to persist with the new hal.21:40
slangasekScottK2: yes, but AFAICS the only reason kernel_driver is NULL in the first place is because of the hal bug...21:40
slangasek"guard against a NULL kernel_driver value" does not address the root issue21:40
slangasekand I believe the root issue is the hal bug, unless someone can confirm they're seeing this with current hal21:41
ScottK2I see what you mean.21:41
ScottK2slangasek: It does seem like the bulletproofing would be generally a good thing.21:42
slangasekI agree21:42
ScottK2I'd say that 204768 exposed 204868.21:43
warrendaehi21:45
warrendaecan i ask a question that is not related to ubuntu here about C ?21:46
stgraberwarrendae: please read the topic first, it'll probably answer your question21:46
warrendaeok :)21:47
stgraberbtw, I wondered where we are supposed to redirect people looking for help with development on Linux, we usually redirect people to #ubuntu but it's probably not the right channel for you :)21:48
bryycestgraber: google maybe?  ;-)22:04
=== \sh_away is now known as \sh
ion_Yay, no alsaconf, and the module's autoprobe fails for the non-PNP ISA card integrated to this laptop.22:12
stgraberbryyce: would probably be a good start indeed :)22:18
dmbhey, we need someone to upload https://bugs.launchpad.net/ubuntu/+source/inspircd/+bug/201941 , is this the place to ask?22:26
ubotuLaunchpad bug 201941 in inspircd "FreezeException request-- Sync with  Debian unstable" [Wishlist,New]22:26
ScottK2dmb: The archive administrators have been subscribed.22:27
dmbso i just wait and it should be good?22:28
ScottK2dmb: They'll get to it in the next few days.  No need to do anything else.22:28
ScottK2Yes22:28
dmbjust making sure :D22:28
dmbScottK2: you were one of the ackers wern't you?22:28
ScottK2For motu-release, yes.22:28
slangasekasac: so what needs to happen to make ubufox happy with icedtea-gcjwebplugin?  It looks like the source package only looks at gutsy Packages files currently?22:36
=== \sh is now known as \sh_away
asacslangasek: yes, the plugin db needs an update run.22:59
slangasekasac: is that something only you're able to do because it's hosted somewhere, or is it something that should be doable within the source package and I'm just not understanding?23:01
asacslangasek: no ... thats on people.ubuntu.co23:02
slangasekok23:02
asacso i until now i run it regularaly23:02
slangasekasac: I guess I'll leave that bug to you then, and you'll close it when the db is updated?23:05
hspaanssomeone here responsible for gnome in Hardy?23:06
asacslangasek: yes.23:16
asacill treat it with high priority23:17
slangasekasac: ok, cheers23:17
asacslangasek: please assign to me23:17
slangasekdone23:18
slangasekOMG these NM/hal bug reports are driving me batty >_<23:18
slangasekit's probably going to take us to the middle of the week to stop getting false-positive confirmations from people with the bad hal23:18
LaserJockslangasek: well, the forums have it, that's usually a contributor23:19
slangasekthe forums encourage users to send "me toos" to bug reports?23:19
sistpotycan't you give and lp hing or s.th.? (thought lp would have such a feature nowadays, but I might be wrong)23:19
sistpotyhint even23:19
slangaseksistpoty: the issue is that people are sending follow-ups with incomplete information to the *wrong* bug reports23:20
sistpotyoh23:20
LaserJockslangasek: not specifically no, but if they give a LP bug number they'll all jump on it23:20
slangasek"yes, I have this driver and it's still broken" - ok, but what version of *hal* do you have installed?23:20
ScottK2slangasek: Be sure to share your pain with however did the hal upload the broke everything.23:20
ScottK2however/whoever23:20
slangasekpitti: <share>23:25
=== evalles is now known as effie_jayx

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