/srv/irclogs.ubuntu.com/2008/10/09/#ubuntu-mozillateam.txt

ftaif i drop the sed, http://paste.ubuntu.com/55440/00:00
ftathe script just need APPNAME to run $LIBDIR/$APPNAME00:00
asacfta: /usr/bin/x-www-browser => APPNAME=firefox-3.100:06
asacdoes x-www-browser link to firefox-3.1?00:06
asacdoes it work if it links to firefox?00:06
asacfta: can you run the same check with the patch as of now?00:07
ftamine, yes00:07
asacwould like to see the difference ;)00:07
asacand file a bug about that :) ... i am sure i will not be able to remember which corner cases were fixed by this ;)00:07
ftado you need $0 later on inside firefox ?00:08
asacyes thats the whole point of all these changes :)00:08
asaci need a good system $000:09
asacwhich isnt versioned if user runs unversioned link (or link of link etc.)00:09
asacwell not system $0, but good basename $0 ;)00:09
asaci already gave up on the hope that $0 might point to the right ;)00:10
ftawhat should I get with 3.1 ? firefox or firefox-3.1 (ie the Name from application.ini)00:10
ftai mean, what do you expect inside ff00:11
ftawell, i'll have a closer look tomorrow00:15
ftai need to reboot, new kernel, new nvidia, new everything..00:16
ftai have a bad feeling about this00:16
ftaas usual, X broke00:25
asac:(00:25
asacfta: when last rebooted?00:25
ftaeach time there's an update of either the kernel or the nvidia driver, the next reboot / X restart is a nightmare00:26
asacyeah. that reminds me, does fglrx exist in intrepid now?00:27
ftagasp, the pc speaker is back00:27
ftaxorg-driver-fglrx | 2:8.532-0ubuntu5 | http://archive.ubuntu.com intrepid/multiverse Packages00:27
ftafglrx-installer | 2:8.532-0ubuntu5 | http://archive.ubuntu.com intrepid/multiverse Sources00:27
asacfta: i mean ... does it work?00:28
ftano idea, my ati box is still running hardy00:29
asaci didnt try to use that for a while because people said that we might not have a working solution for release00:29
asacbut that was 4 weeks back00:29
asacmaybe there was a miracle ;)00:29
ftaapps/metacity/general/audible_bell has been (re)set00:32
asacfta: anything else we need to fix for next upload?00:34
asachttps://bugs.edge.launchpad.net/ubuntu/intrepid/+source/firefox-3.000:35
asaclooks like the script is the only targetted fix left00:35
asachttps://bugs.edge.launchpad.net/ubuntu/intrepid/+source/xulrunner-1.900:35
asachmmm00:36
asacwhy is that thing still open?00:36
asacfta: there are still a bunch of abrowser transitions things open :(00:44
asachttps://bugs.edge.launchpad.net/ubuntu/intrepid/+bugs?field.searchtext=&orderby=-importance&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package=00:44
asaci guess thats me?00:44
ftai was done with all mine00:45
ftaBug 19778600:46
ubottuLaunchpad bug 197786 in prism "Prism apps think I'm offline and no way to change it" [Undecided,Invalid] https://launchpad.net/bugs/19778600:46
asacyeah you can inject the pref ... but that would make NM completely ignored00:46
asacmost likely we will have a real fix for that in time (i hope)00:47
asacthe master is bug 19188900:47
ubottuLaunchpad bug 191889 in pidgin "[MASTER] [WORKAROUND] "Offline Mode" feature fails to detect proper online state for networks that are managed outside of network manager." [Undecided,Confirmed] https://launchpad.net/bugs/19188900:47
ftai wish i could update prism too00:49
ftaand push ff3.100:49
[reed]I wish 802.1x worked on 8.1000:49
[reed]but nooo00:49
[reed]we all have to make sacrifices00:49
ftalol00:50
asac[reed]: have you tried the module paramter?00:50
asac[reed]: does your wired work again?00:50
[reed]wired works again00:50
[reed]thanks to the new kernel00:50
[reed]but still no 802.1x00:50
asacok. thats good news at least ;)00:50
asac[reed]: have you tried the module parameter? or didnt i ask you to test that?00:50
[reed]you asked me, and it didn't work00:50
[reed]but I can try again later00:50
[reed]that was two kernels ago00:50
[reed]:p00:50
asac[reed]: the disable_hw_scan=1 thing?00:51
asacok00:51
[reed]asac/fta: seen http://lists.gnu.org/archive/html/savannah-hackers-public/2008-10/msg00006.html ? :)00:53
asac[reed]: did you write that by yourself? :-P00:55
asac[reed]: but anyway. good info there00:55
asac[reed]: you should send that to debian too ;)00:56
fta"Free software != always secure" :)00:57
ftahttp://ubuntuforums.org/showthread.php?t=94202301:02
[reed]asac: yes, I wrote it myself!02:22
ftahttps://wiki.ubuntu.com/UbuntuBugDay/20081009  it's moving09:30
asacyeah09:54
asacstrange why they look at 111 New bugs instead of 631 though ;)09:55
asac \o/ -> moving everything to inbox-seen :/10:16
asaclets hope that allows me to process mails again :(10:16
sebnerasac: flashupdate \o/10:18
gnomefreakyay i found it ;)10:19
gnomefreaksddoes moz-devscripts need mercurial as a depend?10:26
gnomefreakasac: s/sddoes/asac does10:26
gnomefreak# Error: missing dependencies. Please install mercurial at /usr/share/mozilla-devscripts/mozclient/mozclient.pl line 20610:27
asacsebner: yeah ... that reminds me that I should try it ;)10:27
asacgnomefreak: apt-get install hg10:27
asacgnomefreak: it should be a recommend10:27
gnomefreakok thanks10:27
asacas not all features depend on hg10:27
asacbut in general users should have it installed10:28
asacfta: ^^10:28
gnomefreakhg == not a package10:28
sebnerasac:  ^^ like me but no time for now. soon I have my 2nd "perfektionsfahrt" if you know what's that ;)10:28
asacdamn i changed my password and now i forgot it :(10:29
asaccannot install anything right now :(10:29
sebnerasac: wth O_o10:29
gnomefreakthats bad10:29
asacso lets hope that this gnome sessions doesnt crash :/10:29
gnomefreakwait there are a few password "cracker" packages in repos10:30
asachmm ... maybe i added my ssh key so i can log in as root10:30
sebnerasac: write a posteit :P10:30
gnomefreakasac: hg-buildpackage hg-load-dirs?10:30
asacgnomefreak: ?10:30
asacnot sure what you mean ;)10:31
asacwe dont use hg-buildpackage10:31
asacwe only use it to produce the orig from hg.mozilla.org10:31
gnomefreakasac: what hg package? there is no "hg" package in repos10:31
asacgnomefreak: yeah its simply called "mercurial"10:31
asacthought you already looked for that10:31
asac;)10:31
gnomefreakah ok already installed did that when i got there error10:32
asacsebner: when i made my fuehrerschein i didnt have a "perfektionsfahrt" ... only "ueberlandsfahrten"10:32
asac;)10:32
gnomefreak:) error is gone10:32
asaccool i can use my virtual box VMs again10:34
asacsomething fixed the vboxdrv for me10:34
sebnerasac: grrr10:37
asactoo bad ... but my debian VM doesnst start X anymore :/10:37
asacso most likely i have to trash that anyway10:37
sebnerasac: btw, do you know if there exist a openjdk6-plugin for firefox? Couldn't find one and had to install sun java plugin -.-10:37
gnomefreakjdk == devel right?10:38
gnomefreakif so i dont remember a plugin for FF to use *jdk10:39
sebnergnomefreak: well there is a sun-java6-plugin that's the one for firefox und I couldn't find a openjdk6-plugin (openjdk6 is the opensource java)10:41
gnomefreaksebner: im looking in archives there were ~6 that i saw none said "6" and none were a plugin but still running through archives. im assuming you are still using intrepid?10:42
sebnergnomefreak: sure ^^10:42
sebnergnomefreak: though it's for my parents pc10:44
gnomefreakhmmmmmm seamonkey-browser i guess that means seamonkey has been separated and now "seamonkey" is a meta10:44
gnomefreaksebner: it is Ubuntu? or Winblows?10:44
sebnergnomefreak: ubuntu of course ;)10:45
sebnermaybe it is10:45
sebner<gnomefreak> sebner: it is Ubuntu? or Winblows?10:45
sebnerargh10:45
sebnersry10:45
sebnericedtea-gcjwebplugin10:45
gnomefreakthat is not jdk ;)10:45
sebnerI searching for a firefox web plugin10:46
sebnernot developing java things ;)10:46
gnomefreaksun-java6-jdk is it10:46
sebnernope10:46
sebnerjava development kit is for developing java applications10:46
sebnerthe right thing is *-plugin10:47
sebnerbut I want the opensource java ;)10:47
gnomefreaksebner: than thats should be good but it should only be pretty much same as sun-java6-plugin10:47
sebneryes but as I said there is no openjdk6-plugin ;) but I think it's the icedtea-gcjwebplugin10:48
gnomefreakyeah i noticed10:48
sebnerkk ^^10:48
sebnerasac_: got your pw?10:49
gnomefreakdoesnt openjdk-6-jre bring in plugin for web?10:49
sebnernope10:49
gnomefreakit is the plugin as im seeing10:50
gnomefreakor atleast a bit different than the plugin10:51
sebnerjre is normally for normal java application and it wasn't working here so I installed the sun java plugin which worked10:51
asac_fta: you think this is "unsafe":10:51
asac_                for i in /usr/lib/firefox-3.*/.autoreg; do10:51
asac_                        touch $i10:51
asac_                done10:51
asac_?10:51
asac_in case there is no firefox install that would break, woulnt it?10:51
gnomefreakProvides: java-runtime, java2-runtime, java5-runtime, java6-runtime10:51
gnomefreakthat should work *runtime10:52
asac_gnomefreak: its probably in a separate package (e.g. the plugin)10:53
gnomefreakasac_: its not unless he uses gcj10:53
gnomefreakopenjdk-6-jre provides the above10:53
gnomefreaknot really sure why jave2 is in that or java5 as there are not "free"10:54
asac_be assured, i am at least as confused about what is what for java in our repo as you are ;)10:55
gnomefreakthats why we have doko and/or the java team10:56
asac_to cause confusion ;)?10:56
gnomefreakOpenJDK Team <openjdk@lists.launchpad.net>10:56
asac_imo they should clean this up :-D10:56
gnomefreakto fix the confusion ;)10:56
asac_yeah. doko is working hard ... and what he gets from sun isnt always that great ;)10:56
gnomefreakthat is true10:57
asac_but thats just my guess :) ... nothing official10:57
gnomefreakjava has sucked for years IMHO open java doesnt help10:57
asac_well. open java would have helped10:57
asac_but it taking so many years, that it doesnt help in the end anymore ;)10:57
gnomefreakits still not "open" its restricted as i recall10:57
asac_yeah at least parts of it10:58
gnomefreakmakes no sence unless "parts" is just logo, name10:58
gnomefreaksense10:58
asac_well ... i think one part is the crypto part10:59
asac_that one can be replaced by the bouncycastle free stuff afaik10:59
gnomefreaki think you are right however i havent read the license since it was first included11:00
=== asac_ is now known as asac
asacme neither ... and i dont want to think about that ;)11:00
asachave enough bad dreams without that already ;)11:01
gnomefreakill be back updates are killing my proc for some reason11:01
gnomefreakill be back she wants breakfast and coffee before she leaves11:15
sebnerso foks11:24
sebnerfolks11:24
sebnerwish me luck11:24
sebnerbye bye11:24
asacsebner: toi toi toi11:25
persiaHello.  Someone pointed at the Fennec announcement to me today.  I seem to remember seeing that string in some mozilla packaging stuff when I was looking at the common scripts with fta.11:48
persiaIs the packaging preparation already underway, or is assistance required?  (Jaunty goal)11:48
gnomefreakany good guide on quilt?12:37
persiagnomefreak, What are you trying to do?12:38
persiagnomefreak, I usually just follow the recipe at https://wiki.ubuntu.com/PackagingGuide/PatchSystems for basic stuff.  If you need to do things like insert a patch in the middle of the stack or something, it's a bit more complicated.12:40
fta2persia, i have a branch for fennec since 0.3 (it's at 0.8+ now) but it needs my build-system patch in xulrunner-1.9, that's why i didn't push it to intrepid12:40
persiafta2, OK.  I won't try to recruit someone to chase it and come here for integration then.  Thanks for the confirmation.12:41
fta2persia, no assistance required required for the packaging, and i guess for jaunty, it's just a matter of pushing it12:42
fta2persia, is fennec needed for ubuntu ? or just ubuntu mobile ?12:42
persiafta2, Dunno yet.  Currently Ubuntu Mobile uses midbrowser, but some people would like to change.  A couple of the mobile developers were all excited about today's press release, so I figured I'd stop by and ask.12:44
fta2(my package needs /usr/lib/xulrunner-devel-1.9.0.*/sdk/build-system.tar.gz from xulrunner-1.9-dev)12:44
persiaI wouldn't think that Ubuntu Desktop would want it, just because their target devices can usually run full firefox.12:44
fta2persia, ok, i will start to publish previews in ppa then, so people interested could give it a try.12:46
persiaI suspect Ubuntu MID will probably want Minimo or MicroB.12:46
persiafta2, Don't rush on my account, although if you've previews available, I'm sure you'll have some users.12:46
persiaI heard a couple complaints about usability (like URL bar and scrolling), so it may make sense to get something to testers early, to help the feedback cycle and get something even more solid for Jaunty.12:47
gnomefreaki was just forgetting a command but i found it thanks12:50
asacpersia: we have a tutorial on how to package fennec12:58
asacwith xulrunner-1.9.112:58
asacpersia: https://wiki.ubuntu.com/MozillaTeam/XulApps/Packaging12:58
persiaasac, Thanks.  I'll pass that on.12:59
asacpersia: fta is the author of that page iirc13:00
asacpersia: if mobile team wants to package this, you should really line up with mozillateam13:01
asacdoing this on your own isnt really a great idea IMO :-D13:01
persiaAbsolutely.  That's why I'm here even before there's any decision or discussion on whether to package it :)13:01
asacwe would certainly be happy to have mobile team members contributing to our packages and would help to improve know-how diffusion i guess13:01
asaccool13:01
persiaI'd much rather coordinate to find out status, and if you guys don't have enough time, try to get you some more resources to help, rather than package it separately.13:02
asacyeah. that makes more sense13:02
asacmost work is probably initial bootstrapping of the packages. but if they are done using our best practices, maintaining them could probably be mostly done by us when bumping ffox and so on13:03
persiaRight, or if someone is *really* interested, I suppose they'd do best to help you guys, rather than trying to manage it themselves.13:03
asacack13:04
fta2asac, it's not with xul 1.9.1, it's with 1.913:12
gnomefreakgod i hope i fixed this13:12
gnomefreakill be back while this builds,13:12
asacfta2: oh opk13:13
gnomefreaksunbird 0 gnomefreak 1 ;)13:21
gnomefreakatleast fixed patch13:22
asac0.9 package ready?13:22
gnomefreakbuilding13:22
gnomefreakbut got past the failure13:23
asacgood13:23
gnomefreakif all goes well maybe an hour or so till its done and shortly after it will hit my PPA13:23
gnomefreakjust wish i could fix my sat. dish so i can watch tv again13:24
gnomefreakis there a problem with FF 3.1? its not showing up in preferred apps13:39
gnomefreakit just lists firefox13:39
gnomefreakasac: sunbird wont be ready for a while its a failure now13:41
gnomefreakok fixed the prefered apps just have to use custom command13:42
asacok13:46
gnomefreakasac: fta2 what the hell am i missing? http://pastebin.mozilla.org/55119913:46
gnomefreakam i missing a flag in rules?13:46
asacgnomefreak: yeah it lacks a ptch13:47
asacgnomefreak: look in the tbird package13:47
asacthe patch should be there13:47
fta2look at seamonkey-1.1.dev, there's a patch called debian/patches/11_fix_ftbfs_with_fontconfig.patch13:47
asacor there ;)13:48
gnomefreaklooking13:49
gnomefreakplease remind me why we use quilt14:36
gnomefreakoh yeah because i finally got used to dpatch than we switch to quilt14:37
* gnomefreak crosses fingers and goes outside14:38
gnomefreakok damnit i did everything right and its failing after applying the new patch14:42
fta2imho, quilt is much better, cleaner, faster14:48
gnomefreakasac: fta2 http://pastebin.mozilla.org/551239 would trailing whitespace in configure-autoconf2-13-reconfigure.patch cause this failure?14:48
gnomefreakfta2: dont forget a pain in the ass ;)14:48
fta2well, i won't call quilt new/add/push/pull/refresh a pain in the ass14:49
fta2i like it14:49
fta2#14:50
fta2cd build-tree/mozilla && QUILT_PATCHES=/home/gnomefreak/lightning-sunbird-builds/work/build-area/lightning-sunbird-0.9~cvs+nobinonly/debian/patches quilt --quiltrc /dev/null push -a || test $? = 214:50
fta2#14:50
fta2Patch bzXXX_ftbfs_fontconfig.patch does not exist14:50
asacyeah14:50
fta2you probably forgot the bzr add14:50
asacgnomefreak: bzr add debian/patches/bzXXX_ftbfs_fontconfig.patch14:50
gnomefreakforgot add14:50
gnomefreakok brb trying it14:50
gnomefreakok that fixed it thanks i knew i forgot something :)14:59
* gnomefreak looking over error and i dont see anything that points to forgetting bzr add15:06
gnomefreakmaybe not all patches being there15:07
fta2asac, nspr/nss.head reverted15:24
asacfta2: yeah ;)16:01
asacifupdown is really painful16:01
asaci rarely say that i hate something. but touching that is really pain imo16:01
asaci really cant believe that we have such an esoteric package as the foundation of ubuntu/debian16:02
asacnetworking16:02
asacnext try16:12
saivannasac : To answer your question for bug 259214, it is reproducible on a fresh intrepid installation. It is due to the fact that ethernet connexions use "system settings" so if you change the settings to static IP, the computer will use DHCP on the next reboot16:12
ubottuLaunchpad bug 259214 in network-manager "wired connection settings are lost after reboot" [Medium,Confirmed] https://launchpad.net/bugs/25921416:12
asacsaivann: its fixed in upload i did a few minutes ago16:13
saivannasac : Really, nice to know, thanks!16:13
asacsaivann: please verify and let me know if its not fixed ffor you :)16:13
asacbut i hope it is16:13
asacwell i verified that it is here16:13
saivannasac : Which upload version is that?16:13
asacthere might still be rough edges16:13
asacsaivann: svn20081008...16:13
saivannasac : Thank you, I'll give feedback16:13
asacsaivann: https://edge.launchpad.net/ubuntu/intrepid/+source/network-manager/0.7~~svn20081008t224042-0ubuntu116:13
saivannasac : I'll have to wait for AMD64 build, thanks16:15
nxvlasac: i'm quite confused on how to patch mozilla16:16
jdstrandnxvl: just out of curiosity, is this for a security update?16:16
asacnxvl: we use quilt and branches16:16
nxvlasac: and i'm not sure if it is because i don't know bzr or that everything is tarballed :P16:16
nxvljdstrand: yup16:16
asacnxvl: please state first what you are working on :)16:16
nxvljdstrand: well, stuff like16:16
jdstrandnxvl: those kind of patches need to go through asac first16:16
jdstrandasac: due to our relationship with mozilla16:17
jdstrandnxvl: ^16:17
nxvlok16:17
asacnxvl: whats this about?16:18
* asac reboots and hopes for the best16:34
gnomefreak;)17:05
gnomefreakuploading sunbird17:05
bdmurrayasac: does jazzva maintain foxyproxy?17:08
gnomefreakbdmurray: i think he built it last but let me check17:09
asacbdmurray: he is the "extensions" lead ... but its team maintained in the end17:09
gnomefreakupload+browser makes browser not work17:10
bdmurrayHmm, I took the liberty of subscribing him to bug 27946617:11
ubottuLaunchpad bug 279466 in foxyproxy "Error writing settings file: file:///usr/lib/firefox-3.0.3/foxyproxy.xml" [High,Confirmed] https://launchpad.net/bugs/27946617:11
bdmurrayIs he on irc much?17:12
asacbdmurray: he usually is here every night. but then he is on a holiday trip till next week atm :/17:13
bdmurrayAh, good for him! ;)17:14
gnomefreakwhen did bzr-builddeb start making a dir backup.bzr in the top level dir?17:14
saivannasac : I tried network-manager svn20081008 in a i386 VirtualBox installation (not completely up-to-date) and the problem persist, if I try to change the system setting from DHCP to static, it fall back to DHCP after a reboot17:17
asac saivann have you restarted everything after the update?17:19
asacgnomefreak: thats not builddeb, but bzr upgrade17:19
gnomefreakah17:19
saivannasac : Yes17:21
saivannasac : I'll still try the AMD64 build on a completely up-to-date computer when it will get in the repositories17:22
gnomefreakit works but i get javascript errors "window not defined17:24
gnomefreak"17:24
asacsaivann: soudns wierd that it doesnt work for you. not sure what this is up to. it will certainly fail if you have policykit issues17:24
asacsaivann: do you get at least asked for a passphrase?17:24
* asac reboot again17:24
saivannasac : Yes I guess so, that's why I'll try with a up-to-date intrepid17:25
saivannasac : No, not asked for the passphrase17:25
gnomefreakasac: pushing 0.9 to branch if you need anything with it17:28
gnomefreakow iceowl17:28
gnomefreakasac: do you have an up-to-date iceowl branch? mines a bit old17:30
gnomefreaki still get bzr upgrade warnings but i have latest17:32
gnomefreakbzr: ERROR: The branch format Bazaar-NG meta directory, format 1 is already at the most recent format17:32
gnomefreakok its building in my PPA atm i will be back later today i still have to fix my sat.17:33
gnomefreakoops forgot to test smart gui for sticking at ldconf17:34
asacgnomefreak: isnt the iceowl branch in mozillateam up to date?17:37
asacgnomefreak: you have to upgrade the remote branch too17:38
asacnot only your local one17:38
asacyou can do that with bzr upgrade URL17:38
asacbut take care to not abort that17:38
asacit takes quite a while17:38
asacfta_: there?17:48
asaccan you paste your /var/run/network/ifstate please :)17:48
bdmurrayasac: I've tested bug 277063 and it doesn't seem fixed to me17:51
ubottuLaunchpad bug 277063 in network-manager "Wired and wireless connection to the same router cause confusion" [High,Fix released] https://launchpad.net/bugs/27706317:51
asacbdmurray: does the order of the interfaces still change?17:52
bdmurrayasac: what do you mean?17:53
asacbdmurray: look at your routing table17:54
asacbefore you have a certain order (e.g. eth0 eth1) ... after its (eth1 eth0) in the routing table17:54
asaci am quite sure this has something to do with it17:54
bdmurrayasac: I'll have to check again when first booted.17:55
directhexis there a version-specific xulrunner-addons i should use for 1.9-only or 1.8-only plugins18:20
sebnerflash is soooo f*cking b0rken. more than normally xD18:30
directhexsebner, i can never tell18:31
sebnerdirecthex:  ^^18:31
directhexsebner, right now i'm trying to work out the best way to let some foreigners have proxy access inside the uk, to help debug a plugin18:31
sebnerdirecthex: O_o18:31
directhexsebner, you're aware of ITV catchup?18:33
sebnerdirecthex: when you'd tell me what ITV is xD18:33
directhexsebner, one of our main TV networks18:33
sebnerahhh18:33
directhexITV catchup is their streaming browser-based service, which allows you to watch any of the shows they air, as long as you come from a british ip18:34
directhexexcept... the implementation is sick. twisted and sick18:34
* sebner needs a proxy xD18:35
sebnersilverlight?18:35
directhexsebner, if it was *just* silverlight, it wouldn't be sick18:35
directhexsebner, it's silverflashlight.18:35
directhexsebner, watch http://retro.apebox.org/moon/moon-itvcatchup.ogg18:36
* sebner is scared now18:36
* sebner hide18:36
sebners18:36
sebnerxd18:36
sebnerdirecthex: omg xD18:37
directhexsebner, sick, isn't it :)18:37
directhexsebner, it's a very high profile site, i want it working in ubuntu. so i need to find a way to let the devs be british, for debugging18:38
sebnerdirecthex: you're a nice guy  ^^18:41
asacdirecthex: there should be no 1.8 only plugin19:00
directhexasac, so there's no mechanism for it?19:01
KennethVenke1hello,19:01
KennethVenke1i'm trying to get started with triaging. I have some time at the moment, so i thought, why not start with the firefox hug day. (since firefox is part of my most frequently used software ;) )19:01
KennethVenke1So i was looking at bug #280101 to which i commented. What steps should i do to complete this triage.19:01
KennethVenke1(https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/280101)19:01
ubottuLaunchpad bug 280101 in firefox-3.0 "Firefox search bar points to wrong language wikipedia" [Undecided,New] https://launchpad.net/bugs/28010119:01
ubottuLaunchpad bug 280101 in firefox-3.0 "Firefox search bar points to wrong language wikipedia" [Undecided,New]19:01
bdmurrayasac: yes, the route ordering does change19:11
asacbdmurray: ok ... and you have two net routes right?19:20
asac(e.g. not one host route + one net route)19:20
asacdirecthex: well. if your plugin only uses npapi it should work everywhere19:20
asacdirecthex: if it uses XPCOM you need to link against the dependent glue of 1.8 ... then you can use the binary on 1.8 and 1.919:21
bdmurrayasac: I have two routes to the same subnet if that is what you mean.19:22
asacyeah thats dumb19:22
asacbdmurray: i will test that now :)19:23
directhexasac, there's a core plugin, which is NPAPI, and some per-engine glue, needed for advanced bits & pieces (e.g. allowing the plugin access to the DOM). currently there are two bits of glue, ff2bridge (1.8) and ff3bridge (1.9)19:24
asacbdmurray: i have almost the same setup. doesnt happen here19:26
asacbdmurray: what i do:19:26
asac1. up two devices that have the same subnet19:26
asac2. modprobe -v -r forcedeth (e.g. my wired driver)19:26
asac3. check that the routing was removed for the wired driver19:26
asac4. modprobe -v forcedeth19:26
directhexasac, it's not a huge problem for me as long as i'm only targetting debian (since only 1.9 exists in debian), but...19:27
asac5. check that the routing was happily added again19:27
asacbdmurray: ^^ for me ping never looses a package when doing that19:27
asacbdmurray: can you reproduce that by removing your module too instead of unplugging?19:28
asacbdmurray: unplugged and plugged again19:30
asacworked well too19:30
asacbdmurray: maybe your dns server changes and that causes issues?19:30
asacbdmurray: do you get the same dns through wired and wireless?19:30
asacbdmurray: hmm do you have different metrics for both devices?19:32
asacbdmurray: http://paste.ubuntu.com/55743/19:32
asacbdmurray: so wlan and eth have metric 2 and 1 respectively19:32
bdmurrayasac: no all the metrics are 0 except for 169.254.0.019:33
asacbdmurray: yeah. cant tell then thats the latest code :/ it shouldnt be 019:34
asacbdmurray: did you setup this stuff in connection editor?19:34
asacbdmurray: please paste the output gconftool-2 -R /system/networking19:35
bdmurrayasac: http://paste.ubuntu.com/55746/19:39
asacbdmurray: so those are not system connections?19:40
bdmurrayasac: what do you mean by system?19:42
asacbdmurray: in connection editor "system"19:43
asace.g. not bound to your user account19:43
asacbdmurray: "system setting" ... thats a check box19:43
asacwhich means that those connections can be started even before you logged in19:43
bdmurrayasac: the one's in pastebin are not however eth0 is a system connection19:44
asacbdmurray: ok so you have one system and the wireless is user19:56
asaccan you try to make both system or both user?19:56
bdmurrayasac: okay, fwiw I've never used the connection editor before so I'm uncertain how it got this way19:57
asacbdmurray: are you using managed=true in /etc/NetworkManager/nm-system-settings.conf?19:58
asacif so then your eth0 is most likely configured in ENI19:58
asacwhich would make that a read-only system config19:58
asacerr connection ;)19:58
bdmurrayasac: okay managed is false and eth0 is in ENI.  However, wasn't ENI the right way to go a release ago?19:59
asacbdmurray: well. it means that eth0 isnt managed by NM20:00
asacwhich might explain why NM doesnt know about the metric20:00
asacbdmurray: please set it to =true20:00
asacreboot20:00
asacand see if you still reproduce it20:00
asacor if your metric is fine20:00
asacmy guess is that it works with "managed=true"20:01
bdmurraygeez, reboot really?20:01
asacbdmurray: well ... stop NM20:01
asacthen sudo killall nm-system-settings20:01
asacthen start NM20:01
asacis the other way20:01
bdmurrayokay, but nm-system-settings.conf is new for Intrepid right?20:01
asacbdmurray: yes20:01
asacwe use the same by default that we used before20:01
asace.g. managed=false -> meaning that devices configured there are not managed by NM20:01
asac-> which causes issues as you can see now ;)20:02
bdmurrayOkay so this should be release noted then right?20:02
asachopefully until jaunty we can flip the default to managed=true and say good bye to ifupdown :)20:02
asacbdmurray: no i dont think so20:02
asacbdmurray: the behaviour didnt change20:02
asacif you dont do anything your /etc/network/interfaces devices will not be managed by NM20:02
asacthats the behaviour we had before20:03
asaci will blogg about it ;)20:03
asaci also sent an announcement to ubuntu-devel20:03
asac-> RFC:20:03
asacbdmurray: you can also read some details in README.Debian20:04
bdmurrayasac: I'm looking at that mail now20:04
asaccool20:04
asacbdmurray: the README.Debian might be more comprehensible20:04
asac:)20:04
bdmurrayasac: Why did it used to work in Hardy though?20:04
asacbdmurray: by pure accident then20:04
asacbdmurray:  you would need to observe the behaviour in hardy to tell for sure20:05
bdmurrayHaving eth0 unmanged and wireless managed seemed to be fine20:05
asacbdmurray: first try to verify that this is the bug please :)20:05
asacif it is, we can probably fix it20:06
bdmurrayasac: I've set managed=true and rebooted and see no change in behavior20:10
asacbdmurray: no metric set on wifi connection?20:14
asacbdmurray: does grep NULL /var/log/syslog give you anything?20:14
bdmurrayjust CPU stuff20:15
asacbdmurray: really. the metric in route -n should be 1 for ETHERNET and 2 for WIFI20:15
asacif that isnt the case something is seriously broken on your system ;)20:15
bdmurraymy system! hmmm20:16
asaci looked at the code. and there is really no other way ;)20:16
bdmurrayokay, I'll try and find what I broke then20:17
asacbdmurray: rebooting helps ;)20:18
asacbdmurray: do you see the eth0 connection now in NM?20:18
bdmurrayI'm rebooting again - give me a second20:18
bdmurrayasac: I see "Auth eth0" in nm now - it still seems to be a system setting20:26
asacbdmurray: thats ok20:28
asacAuth?20:28
bdmurraystill no metrics for the interfaces20:28
asacbdmurray: http://paste.ubuntu.com/55758/20:28
bdmurrayAuto - my bad20:28
asacbdmurray: is there a ifupdown (eth0) connection now?20:29
asacbdmurray: you need to apply that patch. maybe your find_route doesnt work. thats the only reason i see that could hinder NM from setting a proper priority on your device20:29
asacbdmurray: its really confusing that you always switch systems when testing this20:38
asacsometimes you have wlan0 and eth0 ... then eth1 and eth0 ;)20:38
asacsometimes you have nothing in /etc/network/interfaces (e.g. like in the bug log) ... sometimes you have configs there20:39
bdmurrayasac: I've always used the same system - the description, which has wlan0, is from the original reporter - not me.20:46
asacbdmurray: anyway. i could reproduce the bug i think with the hybrid approach (e.g. one unmanaged and a managed connection)20:48
asacbdmurray: however, the bug is particularly caused by a metric on the NM managed device20:48
asacif you have 0 everywhere on the metric then i cannot say20:48
asacits definitly another bug20:48
asacbdmurray: the unmanaged default would always have a default route with a metric of "100"20:49
asacthats the way ifupdown handles that20:49
asacalso if i have one unmanaged and one mangaed device i will always have two default routes20:49
asacbut you all the time have only one20:49
asacbdmurray: http://paste.ubuntu.com/55770/20:51
asacthats the routing table that i get ... and which causes packet loss20:51
asacthe default route of wlan has a higher priority than the default route  of eht0 (which comes from eni)20:52
asacbut the net route of eth0 has ahigher prio than the wlan0 route20:52
asacwhich explains why it doesnt work20:52
asacpackets want to go the wlan default route ... but try to do that through the eth net20:52
asacand i get packet loss20:53
asacanyway. whatever i do, the metric will be 2 for wireless devices and 1 for wired (given they are not unamanged)20:53
asacand i know that this was a bug fixed in this last upload :)20:53
asacwhich is the 8th oct build ;)20:54
asaci will think about the hybrid bug ... most likely i have to patch ifupdown again to use a default metric of 10 or so on nets that dont have a metric :(20:55
=== fta_ is now known as fta
ftathe hug day seems staled :P21:26
* sebner hugs fta :P21:26
ftalol21:26
directhexno hugs for me? :(21:36
* sebner also hugs the poor directhex 21:39
sebnerwell you are my mono hero so this is a true hug ^^21:39
directhexsebner, meebey's the hero. i'm more like robin to his batman, if anything21:40
sebnerdirecthex: sure but he's not here ;) btw, good progress with mono2?21:40
directhexsebner, i'm not getting very involved with that right now, mono is meebey's baby. mainly been working on moonlight and xsp21:41
sebnerdirecthex: well, I refered to him ::P21:41
directhexsebner, but i understand he's making progress when he finds time. there were licensing problems to solve first and foremost21:41
sebnergrrrrrr21:41
sebnerlicensing21:41
sebneryuk21:41
directhexlicensing is important. it's patents that can take a flying leap!21:42
asacfta: https://lists.ubuntu.com/archives/ubuntu-devel/2008-October/026676.html21:42
sebnerhowever folks, I wish everybody a good night (if they are in nearly the same timezone as I am ^^)21:44
fta"A debdiff is just a kind of patch and any Ubuntu developer who can't figure out how to apply some slightly differently-formatted patch in a matter of seconds shouldn't be an Ubuntu developer."21:46
ftaeheh21:46
directhexworse than debdiffs are debdiffs which ignore a package's patchsys21:47
directhexi can spend ages sometimes merging ubuntu changes that were poorly applied outside the scope of the existing patchsys21:47
asacfta: yeah ... i was mostly refering to the statements about MOTUs and debdiffs and sending people back and forth for nothing ;)21:47
directhexsebner, anyway, you know i only do backports for LTS, right? some other mug will need to offer mono2 packages for intrepid21:50
ftaasac, i agree with most of it. yet, i don't always like to send patches upstream.. for some projects, i don't want to have to create an account there and deal with a complex/weird/unknown/alien workflow, i expect someone else to take care of that, but i understand that probably noone will :(22:00
bdmurrayasac: what more needs to happen with bug 279083?22:12
ubottuLaunchpad bug 279083 in firefox-3.0 "firefox 3.0.3 on intrepid reports 3.0.1 as user agent" [Undecided,Confirmed] https://launchpad.net/bugs/27908322:12
ftabdmurray, i think i already fixed that22:16
bdmurrayfta: is there somewhere I could check?  I noticed some of firefox is in bzr22:17
ftabdmurray, hm, i fixed it for sure in abrowser, let me check22:21
ftait's in a branch, not released yet22:22
bdmurrayfta: okay, should I do something to the bug or can it get closed by a changelog?22:24
ftabdmurray, please assign it to me, i'll take care of it22:25
bdmurrayokay, thanks!22:26

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