/srv/irclogs.ubuntu.com/2008/03/15/#ubuntu-devel.txt

sistpotycould an archive admin please let the ghc6 upload through? thanks!00:34
nixternalin sispoty style: could an archive admin please rock out the keurocalc-kde4 upload? thanks :)01:09
=== emgent is now known as vodka
=== vodka is now known as emgent
=== dmb_ is now known as dmb
IOUdoes anyone find the 2.6.24-12 kernel slower than the 2.6.24-11 ?03:08
IOUhrm i joined the wrong room to ask that, My Apologies03:13
Hobbseecody-somerville: re; recommends; only for metapackages04:56
HobbseeFujitsu: which version of compiz actually worked/05:03
Hobbseewas it the last one, uploaded on the 7th, or 2 before/05:03
Amarantheh?05:04
HobbseeAmaranth: the special keys is back again05:07
HobbseeAmaranth: it actually went away for a while05:07
Hobbseenow i've hit it twice in less than 24 hours.05:07
Amaranthspecial keys?05:07
Hobbseealt, meta, shift, win, etc05:07
Hobbseepress-and-hold, focus05:07
Hobbseethere's a bug open.  no one's [been able to, bothered] to track it down.'05:08
Hobbseeno caps, either.]05:08
Amaranthoh, that's xorg stuff05:09
Hobbseeand i can't seem to remember how to downgrade05:09
Amaranthcompiz can't fix it, it happens with metacity too05:09
Hobbseei know, but i didn't think xorg had changed05:09
Hobbseeoh wait, it is05:09
Hobbseeer, did05:09
RAOFHobbsee: You mean the bug where when you hold a key down and make a mouse event that key is then permanently pressed?05:12
HobbseeRAOF: i'm holding no key.  but effectively, yes.05:12
HobbseeRAOF: yeah, looks like it.05:32
HobbseeRAOF: i think it's a phantom key though - it doesn't seem to produce any output on the terminal05:32
Hobbseewhereas pageup, etc, will just give pageup spam on the terminal05:32
FujitsuHobbsee: Check using xev next time it happens.05:55
macogwIs the Rhythmbox maintainer around?06:49
macogwguess not06:52
=== sourcode_ is now known as sourcode
jdub6.06 mailman security update bug seen?09:20
jdub            mlist.subject_prefix = Utils.canonstr(09:20
jdub        elif property == 'info':09:20
jdubin /var/lib/mailman/Mailman/Gui/General.py09:20
=== hunger_t is now known as hunger
jdubmakes mailman inoperable (daemon doesn't run)09:21
jdubemgent: ping?09:22
jdub(oh, that was gusty not dapper)09:22
mdkeAmaranth: interesting, I haven't heard of that; will investigate09:37
vik_3278Hi, is this the correct channel to ask about manual back-porting?09:54
vik_3278or is that #ubuntu ?09:54
vik_3278no one here?09:56
jpatrickvik_3278: #ubuntu-motu09:57
jpatrick!weekend | vik_327809:57
ubotuvik_3278: It's a weekend.  Often on weekends, the paid developers, and a lot of the community, may not be around to answer your question.  Please be patient, wait longer than you normally would, or try again during the working week.09:57
vik_3278kk cheers09:57
vik_3278yeah I guess it's the weekend huh.09:57
jdubbug mentioned above is #20233210:12
sorenjdub: What? There's a bug about people not working enough over the weekend? :p10:15
affluxapport is just removing the need-i386-retrace tag from bug 201071 :(10:18
ubotuBug 201071 on http://launchpad.net/bugs/201071 is private10:18
Hobbseesoren: now that should be filed.10:21
ogra_cmpcasac, around by chance ?11:18
pittigeser: oh, xulrunner-dev works? why did you previously patch it to build with firefox?11:38
geserpitti: I merged the last version because mono-tools was missing a build-dependency and did the test-build with universe enabled so I didn't catch that libxul-dev was in universe11:43
pittiah, ok11:43
mdkepitti: did you poke ubuntu-docs from gutsy-proposed to gutsy-updates already?11:58
jeromegScottK: hello12:02
jeromegjdong acked the pidgin backport, if I provide a debdiff, could you please upload it for me ?12:02
emgentjdub: pong12:37
emgenti'm working to it, sorry for delay12:37
jdubemgent: no worries, just wanted to make sure it became known quickly :-)12:42
jdubemgent: thanks!12:42
emgentnp :P12:42
emgentstupid error in ../Mailman/Gui/GUIBase.py12:44
jduband /var/lib/mailman/Mailman/Gui/General.py12:46
jdubsee #202332 (if you haven't already)12:47
emgentyes I saw12:47
emgentanyway, i will attach new patch, but i think that it will upload 17/18 th.. weekend time..12:48
jdubemgent: the problem is, the current security update breaks mailman12:49
emgenti know.12:49
Hobbseeemgent: please don't bug me in query, but is this for hardy or gutsy?12:50
jdubgusty12:51
Hobbsee<sigh>12:51
Hobbseeemgent: where is it then, and have you actually tested this one?12:51
Fujitsuemgent: Ping a security person as soon as you have a tested fix; you may be able to get it done earlier.12:51
cody-somervillesuperm1, ping12:52
emgentok i will do.12:53
erle-will monodevelop 1.0 and mono 2.0 be available in hardy?13:28
pittimdke: ah, no; the changelog failed to include a bug ref, so I didn't see it13:37
pittimdke: done13:38
Hobbseehi pitti13:40
* pitti hugs Hobbsee13:40
* Hobbsee hugs pitti back13:41
=== cjwatson_ is now known as cjwatson
cjwatsonemgent: has anybody been raised on the phone for this yet? security regressions => bad news13:54
emgentcjwatson: debdiff attached and now i'm putting up fix in my PPA too.13:56
cjwatsonI'll make some calls13:57
emgentkees sleep.13:57
emgentcjwatson: ok cool.13:57
emgentonly kees and jamie can upload in -security13:58
cjwatsonit's not an absurd hour for Jamie13:58
emgentok thanks cjwatson13:59
iwksehi all, anybody tried to install a live with a usb-cdrom device?14:00
iwksei'm experiencing a bug in the initramfs14:00
iwksewhile using an usb-cdrom device the new_root is mounted before the cdrom is recognized giving the (initramfs) prompt14:02
iwkseand after few seconds it recognize the cdrom and allow me to mount it14:02
iwksenow...i would really like to boot from initramfs but i don't know how to do14:02
iwkseso i was searching a way to delay the mounting of root14:03
iwkseroot/new_root14:03
iwkseanybody experienced it?14:03
Hobbsee!weekend | iwkse14:06
ubotuiwkse: It's a weekend.  Often on weekends, the paid developers, and a lot of the community, may not be around to answer your question.  Please be patient, wait longer than you normally would, or try again during the working week.14:06
mdkepitti: sorry about that - and thanks!14:09
Mithrandiriwkse: I installed hardy in that configuration at least.14:09
pittiemgent: hi14:09
yao_ziyuani want to suggest something, very important14:10
yao_ziyuani want a Language tray icon in a freshly installed ubuntu/kubuntu, just like windows does14:17
yao_ziyuanbecause many east asian users are used to find their input methods there14:18
yao_ziyuanwell, essentially, it's a SCIM tray icon14:18
yao_ziyuanso what i mean is, install and run scim by default14:18
yao_ziyuanno matter what the installation/system language is14:18
ogra_cmpcyao_ziyuan, did you try out hardy ?14:20
yao_ziyuandoes hardy do that?14:21
ogra_cmpcit does exactly that if an asian language is used :)14:21
ogra_cmpcand you can easily enable it on all others with one checkbox14:21
yao_ziyuanbut an east asian user is used to select his language from a tray icon, not from a system menu14:21
Hobbseeyao_ziyuan: ubuntu does that.14:23
ogra_cmpcwhy wouldnt an asian user not install his native language ?14:23
ogra_cmpci guess you rather mean keymap than language14:23
yao_ziyuan...14:24
yao_ziyuanbecause of windows experience14:24
yao_ziyuantrust me. a little default tray icon can lead you to wonder14:24
Hobbsee...with the endless popups and the spyware.  a really good idea to duplicate it.14:24
ogra_cmpcif you have installed any of the asian languages the scim applet is started by default14:25
yao_ziyuanthen, put a default tray icon that opens the install language dialog box14:25
ogra_cmpcbut you have to select your language before boot of the cd anyway ... why add another applet ? you cant boot hardy Cds without having selected the language14:26
ogra_cmpcif that language is asian the scim applet will be in the panel14:27
cjwatsonyao_ziyuan: we install it by default for all languages in hardy, but we don't run it because scim's default keybindings confuse non-Asian users14:28
cjwatsonyao_ziyuan: this was actually quite a big issue over the last week; it got accidentally turned on by default and caused a great deal of confusion14:28
cjwatsonyao_ziyuan: so I think we're fine as we are in hardy :-)14:28
yao_ziyuanmany keybindings in scim are not necessary even for east asians14:31
yao_ziyuanit just needs a Ctrl+Space14:31
yao_ziyuanand a Shift+Ctrl14:31
cjwatsonthis was exactly what confused people14:31
yao_ziyuan Shift+Ctrl is optional too14:31
ogra_cmpcthhats exactly been the biggest issue :)14:31
cjwatsonpeople hit that a lot by accident. We are absolutely not turning that on by default, sorry14:31
yao_ziyuanwhat about Ctrl+Space?14:31
cjwatsonyes14:31
yao_ziyuansigh14:31
cjwatsonbut it's more prominent in hardy than it was in gutsy; I think it's still an improvement14:32
yao_ziyuanthen don't activate any keybindings unless the user chooses Chinese/Japanese/Korean in the tray icon menu14:32
yao_ziyuani mean, choose by mouse14:32
ogra_cmpcyao_ziyuan, thats what we do14:32
cjwatsonwe'd welcome a patch for that14:32
cjwatsonogra_cmpc: no, it isn't14:32
ogra_cmpcbut you select it earlier14:32
Hobbseecjwatson: yao_ziyuan is a kubuntu user.  he probably didn't see it14:32
ogra_cmpccjwatson, well, we force lang selection upon the user on boot14:33
yao_ziyuanmy concern is that almost all computer users in china are using backdoored windows14:33
cjwatsonogra_cmpc: if your locale isn't CJKV or Thai, scim isn't run14:33
cjwatson14:32 <yao_ziyuan> then don't activate any keybindings unless the user chooses Chinese/Japanese/Korean in the tray icon menu14:33
cjwatson14:32 <yao_ziyuan> i mean, choose by mouse14:33
cjwatson14:32 <ogra_cmpc> yao_ziyuan, thats what we do14:33
yao_ziyuanthat hurts their democracy enterprise greatly14:33
ogra_cmpcright14:33
cjwatsonogra_cmpc: yes, we make the user select a language at boot, but your description above is absolutely not accurate14:33
yao_ziyuani want them to switch to linux14:33
yao_ziyuanbut there has to be a way to activate a chinese input method similar to that in windows14:34
yao_ziyuanchoosing language upon installation is indeed good for novices14:35
yao_ziyuani didn't know vietnamese and thais also need input methods :)14:36
ogra_cmpcyao_ziyuan, is it a general habit in asia to not choose the sysem to be in your native lang ?14:36
yao_ziyuanno14:37
ogra_cmpcthen the existing sysem inded doesnt gain much ...14:37
yao_ziyuanbut for bilinguists like me14:37
yao_ziyuani prefer to remain english as default14:37
yao_ziyuanby no i mean yes14:38
ogra_cmpcheh14:38
yao_ziyuanthey choose native languages14:38
ogra_cmpcthats how i understood it :)14:38
ogra_cmpcso for the majority it will likely be an improvement as it is in hardy then14:39
yao_ziyuanin kubuntu 7.10, if you choose chinese upon installation, you will get a default font that displays english letters in an ugly fixed-width way14:39
yao_ziyuanyeah i also want to mention, if the user adds chinese, you should use the latest wenquanyi fonts14:40
yao_ziyuanbut ubuntu's chinese fonts are fine14:41
yao_ziyuani mean kubuntu14:41
ogra_cmpci think especially wrt fonts hardy has improved a lot, the person who is responsible for font and input handling since hardy lives in asia, you should probably take a look, beta wil come out soon14:41
yao_ziyuanit's a lot lagging behind14:41
yao_ziyuani'm downloading kubuntu hardy alpha 614:41
yao_ziyuanubuntu and fedora have perfect chinese display and input14:41
yao_ziyuanbut they still require you to add chinese from a system menu, rather than from a tray icon, which would be more prominent14:42
yao_ziyuanfor kubuntu 7.10, it sucks at its default chinese font and its way to enable chinese input: you have to "add language: chinese" AND THEN "set system language: chinese"14:43
yao_ziyuani didn't know i should "set system language: chinese" in my early kubuntu days so i had a lot of frustration getting scim working14:43
* ogra_cmpc cant comment on kde14:43
yao_ziyuani found the best way so far to enable scim in kubuntu is to install ubuntu first, and then enable chinese input in ubuntu, and then install kubuntu-desktop. and then i log into a kde session, and then Ctrl+Space will give me the input method.14:44
yao_ziyuanbut this way i still have to manually download the latest chinese fonts from wenquanyi's website14:45
yao_ziyuanor else chinese characters in my firefox will be junk14:45
jdstrandlamont: I think it should go in sooner than later, but I am working through upgrade scenarios and will be more conservative on upgrades from pre-hardy14:47
jdstrandlamont: this isn't bind9-specific, we are working out the details for migrating profiles/etc for all apparmor enforcing packages14:48
lamontjdstrand: fwiw, bind9 runs just fine with the file on dapper... :-)14:48
jdstrandlamont: bottom-line, I hope to have some postinst and doc changes14:48
lamontah, for bind9?14:48
jdstrandlamont: heh, yes it would be fine on dapper14:48
jdstrandlamont: yes-- we will put the profile in complain mode on certain upgrades and enforce on new install14:49
jdstrandlamont: so as not to break a dapper-hardy upgrade14:49
lamontah, ok.  we'll want to chat about those, I expect.14:49
lamontbecause there are far more configs out there than the default config14:50
jdstrandlamont: all the details aren't worked out yet, but once I dig into it more, we can chat14:50
yao_ziyuancjwatson: how is it "more prominent" in hardy than in gutsy?14:50
jdstrandlamont: oh sure-- and this isn't bind9 specific14:50
=== jono is now known as highvoltage
jdstrandlamont: but it will affect the apparmor snippet in postinst somewhat14:51
lamontjdstrand: those fun little other-configs are the reason that bind doesn't force the chroot or user name on upgrades.  (well, doesn't default to chroot in any case)14:51
lamontright14:51
* lamont grumbles about jdstrand including a (&*^(&*)_)($&%)*_)(^*&%^(_ changelog entry in his patch14:53
lamontand manually applies the patch14:53
jdstrandlamont: oh-- you don't want the changelog entry too? I was trying to make it easier for you14:53
lamont1) changelog never merges cleanly14:54
lamont2) I autogenerate it at the very end (note the diff on any revision whose commit log says 'changelog: release' :0)14:54
jdstrandok-- I was thinking of the patch as a debdiff.  I'll know for next time14:55
lamontif I had just git-am'ed your patch, the changelog record would have been:  Fix for LP: #201954 (apparmor profile does not allow access to /var/lib/bind)14:55
lamontyou still have the git tree?14:55
jdstrandlamont: yes14:55
yao_ziyuankubuntu makes such a distinction between "installed languages" and "system language"14:56
lamont    Addresses-Ubuntu-Bug: 20073914:56
lamontthe changelog basically gets line one of the commit log (the short log print), and will add the LP# syntax if Addresses-{Debian,Ubuntu}-Bug: NNN is there14:56
yao_ziyuani want kubuntu to have a simple Languages listbox like ubuntu does14:56
yao_ziyuanand like fedora does14:57
yao_ziyuanthe Language Selector14:57
lamontjdstrand: and the sad part about me munging your change around is that it'll make for merge fun when you refresh after I push it.14:58
lamont(for you)14:58
jdstrandlamont: it'll help me remember14:59
jdstrand;)14:59
lamontlol14:59
lamonthrm....15:01
lamontjdstrand: so do we make /etc/bind/* read-only and force people to either edit the profile or "fix" their config to be conformant?15:01
lamontthe default fresh install should make it read only15:02
lamontupgrades may want it to be rw15:02
jdstrandlamont: I think it would be wise to be conservative in terms of functionality, and leave it rw. The unix permissions will then handle it properly.15:03
lamontah, true15:03
jdstrandlamont: from a security perspective, I would want it 'r', but it is really very little gain for potentially a lot of pain15:03
* lamont steals the bug, marks it fix-committed15:03
lamontyeah - the admin can always tweak the file - we shouldn't need to change it much once it's done "enough"15:04
* jdstrand nods15:04
* lamont really needs to push bug 191530 upstream15:05
ubotuLaunchpad bug 191530 in bind9 ""host" cannot see sites in .org" [Medium,Triaged] https://launchpad.net/bugs/19153015:05
lamontit's a simple case of host not going the distance15:05
jdstrandheh15:05
lamont.com nameserver return glue as A RRs (violation of spec), because it works better15:06
lamont.org nameservers don't.15:06
lamontso if your .org domain has nameservers outside of .org, then host doesn't print things.15:06
lamontof course, the real answer is "use dig. kthx"15:06
jdstrandinteresting-- haven't hit that myself15:07
lamontthere's a domain name in the bug15:08
MithrandirI'm wondering if we should think about shipping some compat versions of older tools, so people can still go host $blah, and get similar-formatted output, but from dig.15:08
lamontand I might have those details wrong, too.15:08
Mithrandirditto for traceroute/mtr.15:08
=== pbn_ is now known as pbn
jdstrandyeah, I get 'sourceforge.org has no NS record' too15:10
lamontand if you look at the dig output, you can see why15:12
lamont;; flags: qr rd; QUERY: 1, ANSWER: 0, AUTHORITY: 3, ADDITIONAL: 015:12
lamontvs dig ns sourceforge.com @a.gtld-servers.net15:13
lamont;; flags: qr rd; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 315:13
* jdstrand nods15:14
* lamont would welcome someone going through http://bugs.debian.org/src:bind9 and triaging them15:14
lamontmost of which are "does this still happen" questions for the submitter...15:15
* lamont wonders about debian bug 408432 and whether it was fully addressed (well, in debian, so I could close it...)15:16
ubotuDebian bug 408432 in bind9 "BIND remote exploit" [Important,Open] http://bugs.debian.org/40843215:17
jdstrandlamont: I can dig into it on monday (pun intended)15:18
lamontheh15:18
lamont#431663: dig can be made to crash using -f, named pipes, and signals15:26
lamont"and your point?" :-)15:26
WartornJust wanted to let you know, the "hardware testing" crashed on me, when it came to the mouse-test and i pressed next, now my cpu keeps running at 100%, even after a reboot15:26
Wartornand it keeps alternating between the cores, where one keeps running at around 5% and the other at 100%15:27
iwkseany developers around here? :)15:47
jpatrick!weekend > iwkse (:))15:47
iwkse:)15:47
iwkselazy hackers15:48
=== spacey_ is now known as spacey
=== fta is now known as fta_
=== fta_ is now known as fta
pitti!ask | iwkse15:54
ubotuiwkse: Please don't ask to ask a question, ask the question (all on ONE line, so others can read and follow it easily). If anyone knows the answer they will most likely answer. :-)15:54
VadiIs there a list of requirements somewhere for getting an @ubuntu.com forwarding email address?15:57
jpatrickVadi: you have to be a member15:57
iwksepitti: i did the question before, but i can repeat it..wait15:57
jpatrickubotu: tell Vadi about member15:58
iwkse<iwkse> i'm experiencing a bug in the initramfs15:59
iwkse<iwkse> while using an usb-cdrom device the new_root is mounted before the cdrom is recognized giving the (initramfs) prompt15:59
iwkse<iwkse> and after few seconds it recognize the cdrom and allow me to mount it15:59
iwkse<iwkse> now...i would really like to boot from initramfs but i don't know how to do15:59
iwkse<iwkse> so i was searching a way to delay the mounting of new_root15:59
Vadijpatrick: Ok, thanks15:59
=== Traxer is now known as Traxer|on
=== Traxer|on is now known as Traxer
superm1cody-somerville, hey what's up?18:21
=== Seveaz is now known as Seveas
=== dijital1_ is now known as dijital1
alex-weejplease can a developer take a quick look at https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/201127 ? i've posted the changes that I think are suitable for resolution19:28
ubotuLaunchpad bug 201127 in network-manager "(Hardy) please remove Network Manager Editor from Internet and Preferences" [Undecided,New]19:28
enarxe0hi19:51
RotundIs there any chance of getting a new package in for Hardy yet?  Something that's in community?20:10
LaserJockRotund: like an updated package or a completely new one?20:11
=== sudobash is now known as bashsudo
=== bashsudo is now known as baneva
=== baneva is now known as banevade
ScottK2In January serpentine dropped from Main to Universe.  If it's going to stay there, there's some stuff that can be re-enabled.  I'm working on an upload to support python-xml removal.  I can deal with that at the same time unless someone knows of plans to promote it again?20:55
=== Amaranth_ is now known as Amaranth
slangasekScottK2: nope, I'm pretty sure that dropping it in favor of brasero was deliberate21:03
ScottK2OK.  Just checking.  Didn"t want to do a bunch of work someone would just have to undo later.21:04
ScottK2Thnks21:04
jpatrickKopfgeldjaeger: ##fix_your_connection22:09
slangasekmjg59: ping22:54
=== calc_ is now known as calc
emgentheya23:57

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