[00:07] <Hamra> debian should have a better bug tracking system! it's just.... old!
[00:19] <chrisccoulson> Hamra - I just saw this synaptic wishlist you sent to debian
[00:20] <chrisccoulson> generally, it's a good idea to quote the ubuntu bug number in bugs that are forwarded upstream :)
[00:21] <Hamra> ok, it's the first time i do anything upstream, i'll remember next time
[00:22] <Hamra> is this bug 275590 really valid? it could be his hardware, or could be an unintentional double-click, it happens 1 in 50 times!
[00:25] <chrisccoulson> thats a difficult one. i'm sure i've seen people talk about mouse clicking problems in firefox before
[00:25] <chrisccoulson> it would be worth finding out if firefox is the only application that it happens in
[00:41] <Hamra> in bug 266946, on what basis was it actually confirmed and set as high? is there any proof that firefox actually doesn't need any of these packages?
[00:45] <wgrant> Hamra: There is proof. Common sense, and thinking the dependency chain through.
[00:45] <wgrant> I think it's a duplicate, however.
[00:46] <wgrant> (the problem stems from firefox's Recommends of ubufox)
[01:28] <Hamra> good night everybody
[03:40] <Hobbsee> !5-a-day
[03:41] <Hobbsee> yeesh.  only 1595 karma!  I must have been inactive for a while!
[03:43] <Hobbsee> and i haven't even hit the stats for yesterday.  shameful.
[04:07] <dholbach> good morning
[04:36] <mrooney> dholbach: hello!
[04:36] <dholbach> hi mrooney
[04:38] <mrooney> this rhythmbox patch for bug #42686 is making me understand more C than I anticipated
[04:38] <mrooney> if only devs would integrate patches instead of waiting until they can no longer be remotely cleanly applied to be interested in them
[05:21] <bucket529> Need bug#268677 changed to Wishlist - Debian is working on a package, but it's not finished.
[05:35] <bucket529> Need bug#194212 changed to Wishlist. Packaging Request
[05:36] <Hobbsee> bug 194212
[05:36] <Hobbsee> hm, that'd be a workflow bug.
[05:37] <Hobbsee> debian bug 442032
[05:39]  * Hobbsee wishes the guy would have changed the title to something a little more approriate, too.
[05:40]  * Hobbsee marks it invalid anyway
[05:55] <Hobbsee> dholbach: can i have a pony?
[06:06] <dholbach> Hobbsee: did you ask your parents first?
[06:07] <Hobbsee> dholbach: no.  but i did feature on 5-a-day :P
[06:08] <dholbach> I'll check with Santa... I think there was something about "asking ones parents about the pony" in the rulebook
[06:08]  * dholbach goes out for a run
[06:11] <Hobbsee> haha
[13:58] <salty-horse> hi. I have some problems with libgdl dependencies that prevent it from being upgraded. is it a known problem?
[14:00] <Hobbsee> salty-horse: which problem?
[14:01] <salty-horse> I have libgdl-1-0 in version 0.7.11-1. there's an upgrade available to 2.24.0-1, but python-gnome2-extras has not been updated, and it depends on the older version
[14:02] <salty-horse> also, python-gnome2-extras-dev HAS been updated, but it depends on a newer python-gnome2-extras which doesn't exist
[14:03] <persia> salty-horse: There's been a bit of a catch and loop with build timings.  It ought be sorted reasonably soon.  Check the build state on LP if you want to watch the progress.
[14:03] <salty-horse> persia, it's been like this for about a week
[14:03] <salty-horse> I'll check the builds on lp
[14:04] <persia> salty-horse: Yes, there's been a lot of confusion about it.  I think I saw traffic indicating it might be sorted earlier today, although either I could be mistaken, or those sorting it could be mistaken.
[14:09] <salty-horse> ouch:
[14:09] <salty-horse> Setting up libc6 (2.8~20080505-0ubuntu7) ...
[14:09] <salty-horse> init:io.c:724: Assertion failed in nih_io_message_send: message != NULL
[14:09] <salty-horse> Aborted (core dumped)
[14:26] <salty-horse> persia, here's the problem. what do I do now? https://launchpad.net/ubuntu/+source/gnome-python-extras/2.19.1-0ubuntu10/+build/724795
[14:27] <persia> salty-horse: Well, you can check who uploaded it 23 minutes ago, and if you think you have a solution to the problem, perhaps chat with them to share your results.
[14:27] <persia> If you don't have a solution yet, you can wait for them to try again.
[14:27] <salty-horse> persia, I think that build fails for a long time now. seb128 made the last change
[14:27] <persia> If you watch the build status, you'll be able to see progress on resolving your issue.
[14:28] <persia> salty-horse: Look at your link again.  Last build started 24 minutes ago.
[14:28] <salty-horse> what does it mean? maybe it rebuilds it again and again, failing each time?
[14:29] <persia> salty-horse: The problem may have been around for a while, but that doesn't mean it's being ignored: in this case you can see it being actively worked on.
[14:29] <persia> No, when it fails it waits for a developer to take some action.
[14:29] <salty-horse> ok
[14:29] <salty-horse> I'll wait then
[14:29] <salty-horse> thanks
[14:32] <salty-horse> hi seb128. aware of https://launchpad.net/ubuntu/+source/gnome-python-extras/2.19.1-0ubuntu10/+build/724795 ?
[14:32] <salty-horse> :/
[14:34]  * Hobbsee mutters something to seb128 about telling people off for sending mails about test building, and then not test building himself.  :)
[14:34] <Hobbsee> [23:32] <salty-horse> hi seb128. aware of https://launchpad.net/ubuntu/+source/gnome-python-extras/2.19.1-0ubuntu10/+build/724795 ?
[14:35] <seb128> bah, people start getting annoying, if you can't deal with installability issues don't run an unstable distro
[14:36] <seb128> Hobbsee: I never said I was testing my builds and didn't tell people off for that, I told you off for ranting about things you don't do yourself
[14:36] <Hobbsee> seb128: i usually do.  I thought that one was safe.
[14:36] <Hobbsee> seb128: besides, i wasn't rantin.
[14:36] <Hobbsee> er, ranting
[14:37] <seb128> well, if you give public advice you should be carreful to apply those to yourself for the days after giving the lesson ;-)
[14:37]  * persia thinks it's good to ask people to test build, and good to poke people when they don't.  Helps us all get the right practices.
[14:37] <Hobbsee> seb128: if my pbuilder hadn't exploded, like my akregator (again) and my email, i would have.
[14:38] <seb128> persia: I do test builds locally but that one is somewhat sbuild specific apparently
[14:38] <persia> seb128: I know.  I don't fault you it not building.  I've local sbuild if I can test it for you.
[14:39] <seb128> persia: that's ok lool debugged a similar issue in pygtk this morning so we just have to apply the same fix but thanks
[14:39] <persia> seb128: It's more that I think Hobbsee was right to remind people to test build & test install, as we'd been seeing a lot of growth in unmetdeps, and I think you were right to poke her when she missed because we all ought help identify anything that needs help.
[14:39] <persia> seb128: You think the pygtk fix will also work for -extras?  That's great news.
[14:41] <seb128> yes, that's the same xvfb issue
[15:07] <salty-horse> can anyone help me debug this warning?
[15:07] <salty-horse> Setting up capplets-data (1:2.24.0.1-0ubuntu2) ...
[15:07] <salty-horse> WARNING: Failed to parse default value `??????????? ?????? ;gtk-theme-selector.desktop,???????????? ??????????? ???;default-applications.desktop,??????????? ????;gnome-cups-manager.desktop]' for schema (/schemas/apps/control-center/cc_actions_list)
[15:08] <salty-horse> gconf-editor doesn't give me details on /schemas/apps/control-center/cc_actions_list
[15:10] <seb128> salty-horse: control-center.schemas has a broken translation
[15:11] <salty-horse> where can I look for it? I don't even have /usr/share/applications/gtk-theme-selector.desktop
[15:12] <salty-horse> kees, here? I get an assert when upgrading to the latest glibc: init:io.c:724: Assertion failed in nih_io_message_send: message != NULL
[15:13] <salty-horse> seb128, where should I report it?
[15:35] <Hobbsee> bug 273856
[15:41] <salty-horse> Hobbsee, any idea what to do with my broken translation issue?
[15:42] <Hobbsee> salty-horse: no.
[15:42] <salty-horse> any idea where to look or who to ask, or even what package should I file a bug under?
[16:09] <seb128> salty-horse: report it on bugzilla.gnome.org against the broken locale if that's not fixed yet
[16:26] <salty-horse> seb128, how can I figure out which locale is it? I have no idea how to debug it.. that's why I'm asking here
[16:36] <lfaraone> Is it bad practice to set priority on your own bugs?
[16:36] <lfaraone> Like setting 235941 to "low"?
[16:36] <persia> bug #235941
[16:37] <persia> lfaraone: It's discouraged, except in the case of setting to "Wishlist" when you're sure it's not important, or if you're someone who has taken responsibility for a given class of bugs anyway.
[16:38] <lfaraone> persia: Ok.
[16:39] <persia> For example, I try to triage and fix *every* bug in wildmidi or freqtweak.  As a result, I often triage my own bugs there, because I'm likely to coordinate the fix.  Similarly, when I report an issue complaining about some dialog text that's not quite right, I'll set it to wishlist.  For other bugs, I tend to try to find someone else to triage them (perhaps as trade for triaging their bugs).
[17:07] <kees> salty-horse: is the init bug related to locale?  I'm confused after reading the backlog.  :P
[17:07] <salty-horse> no, it's a different problem :)
[17:08] <salty-horse> and I have no idea where that coredump was written to :)
[17:08] <salty-horse> (I used "sudo aptitude")
[17:14] <kees> salty-horse: when does the init problem happen?
[17:14] <salty-horse> when I configure the package after installation
[17:15] <salty-horse> right after the "setting up libc6" message
[17:16] <Pici> If you're using Intrepid, then there are few other people reporting the same libc6 coredump in #ubuntu+1, I'm not sure if any bugs have been logged yet, and I'm not going to try the upgrade myself until I have physical access to my computer.
[17:16] <salty-horse> Pici, I am using intrepid
[17:17] <seb128> salty-horse: does that break the upgrade?
[17:18] <salty-horse> no. it continues. but I'm not sure if things are ok
[17:18] <seb128> salty-horse: did you open a bug about that yet?
[17:18] <salty-horse> nope, because I'm not sure what the problem is
[17:19] <salty-horse> should I just open a bug with the error and my specs?
[17:19] <seb128> well, bug reports are there to report issues, they are useful even if you are not able to figure what the bug is ;-)
[17:19] <salty-horse> ok, I will
[17:19] <seb128> just copy the upgrade log and the error
[17:19] <salty-horse> ok
[17:22] <kees> init:io.c:724: Assertion failed in nih_io_message_send: message != NULL
[17:22] <kees> Aborted (core dumped)
[17:22] <kees> freaky
[17:23] <kees> it didn't do that when I used dpkg locally
[17:23] <salty-horse> kees, I'm reporting it now
[17:24] <kees> salty-horse: thanks.  I'm moving to #u-devel to see if doko or Keybuk know what's happening
[17:29] <kees> salty-horse: looks like a bug in upstart
[17:43] <salty-horse> kees, https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/275958
[18:42] <mrooney> pedro_: regarding that rhythmbox bug, hadess isn't interested; his preferable patch uses gnome-keyring instead of using the md5
[18:42] <mrooney> which I guess I can agree with, although doing that would take a lot more research on my part
[18:49] <copyofjohan> bug 147464
[18:50] <copyofjohan> what could i do help testing?
[18:57] <mrooney> is there a way to trigger a remote bug watch update?
[18:57] <bdmurray> mrooney: nope, just wait
[18:57] <Hamra> what does the "denied for gutsy by ..." mean?
[18:58] <mrooney> you mean declined?
[18:58] <Hamra> yes :P
[18:59] <mrooney> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/147464/comments/80 might help
[18:59] <Hamra> lol, that's where i noticed this message
[19:00] <mrooney> Right I know you noticed it in the bug
[19:00] <mrooney> but that specific comment explains it
[19:00] <Hamra> ah, you're pointing to comment 80? ok thanks
[19:03] <Hamra> this is a really old bug, with loads of people experiencing it, so why is it still marked new?
[19:48] <mrooney> Hamra: I don't know which package to confirm it against
[19:49] <mrooney> Is everyone in Intrepid getting a logout dialog on shutdown like in bug #252795
[19:51] <lfaraone> mrooney: I think that'd be expected behavoir, it's controlled in the ACPI preferences.
[19:53] <jdong> mrooney: yeah that's expected behavior though I'm not convinced either it's the right behavior.
[19:53] <jdong> when I push the power button I expect to be greeted with options controlling the power state of my system
[19:53] <jdong> such as shutting down, restarting, suspending, hibernating, or turning off the display
[19:57] <lfaraone> jdong: So, wishlist, or low?
[19:58] <mrooney> jdong: yeah, it definitely seems like a power button should allow you to alter the power state
[19:58] <mrooney> especially since Hardy worked that way
[19:59] <jdong> lfaraone: I'd say leave untriaged, continue discussion on ubuntu-devel-discuss list
[19:59] <jdong> lfaraone: and basd on initial developer opinions triage it
[20:07] <Hamra> i use KDE here, this can happen if i uncheck "offer shutdown options" in session managment, maybe there's something similar in GNOME?
[20:09] <mrooney> Not that I can find
[21:14] <chrisccoulson> anyone here speak a bit of french? looking through the log of someone with a failed upgrade, and I see this line: "ERROR got an error from dpkg for pkg: 'php5-cli': 'sous-processus post-installation script tué par le signal (Interrupt)"
[21:14] <chrisccoulson> basically saying the post-installation script was interrupted right?
[21:14] <chrisccoulson> by the user?
[21:15] <hggdh> sub process post-installation script killed by the signal interrupt
[21:15] <Hamra> i guess it means there was an interrupt signal, which could possibly mean a ctrl+c
[21:15] <chrisccoulson> which is basically CTRL+C isn't it?
[21:15] <chrisccoulson> you beat me to it hamra;)
[21:16] <hggdh> probably
[21:16] <bdmurray> chrisccoulson: which bug?
[21:16] <chrisccoulson> bug 220066
[21:16] <chrisccoulson> one of the ones from the hug day last week
[21:21] <bdmurray> bug 138535 sounds similar
[21:22] <chrisccoulson> well spotted
[21:23] <afflux> chrisccoulson: 220066: user pressed ctrl+c when being asked for what to do with differing config files.
[21:23] <afflux> I'll look at the german one
[21:24] <chrisccoulson> which bit of the log tells us that they interrupted when being asked what to do with the conffiles?
[21:25] <afflux> apt-term.log, line 2046 (long line with a lot of control characters) is very likely the dialog for the conffiles
[21:25] <afflux> immediatly followed by "sous-processus post-installation script tué par le signal (Interrupt)"
[21:26] <afflux> looks suspicious in my opinion. Same goes for the 138535 by the way.
[21:27] <chrisccoulson> thats strange. it seems unlikely that two users would interrupt it exactly the same point
[21:28] <afflux> true
[21:28] <hggdh> it is not really strange: a dialog is shown there, so that gives the user a lot of time to decide what to do
[21:29] <chrisccoulson> thats true. but why haven't we seen more users interrupt conffile prompts for other pacakges?
[21:29] <chrisccoulson> or maybe we have?
[21:29] <hggdh> *that* I cannot answer ;-)
[21:29] <afflux> I haven't
[21:30] <afflux> maybe bdmurray has ;)
[21:31] <bdmurray> Not I, but I have some lists of post-installation script apport-package bugs someone could look at
[21:32] <afflux> sounds like a good target for a hugday, after intrepid release maybe
[21:32] <chrisccoulson> the description in bug 138535 actually says that the upgrade hung at installing php5-cli, which would explain why the user interrupted it
[21:32] <chrisccoulson> maybe the same happened for bug 220066
[21:33] <chrisccoulson> unfortunately, the logs do not say why it hung
[21:33] <chrisccoulson> i think the only way to work that out is probably to do a gutsy -> hardy upgrade
[21:34] <afflux> I can imagine that the users did not read the last lines. They just noticed that it stopped there for quite some time and that's why they try a ctrl-c
[21:34] <james_w> I believe you can get a package hang if a package misuses ucf
[21:34] <hggdh> bug 221856 sounds identical
[21:34] <james_w> ucf displays debconf like prompts as well
[21:34] <chrisccoulson> its funny you should say that james_w. i've just looked in the postinst script for php5, and there is a workaround for legacy php5 packages which use ucf
[21:34] <chrisccoulson> to take care of the upgrades from dapper -> hardy
[21:35] <james_w> chrisccoulson: can you stick it in a pastebin please?
[21:35] <chrisccoulson> http://paste.ubuntu.com/52224/
[21:35] <james_w> fanks
[21:35] <chrisccoulson> no problem;)
[21:36] <james_w> hmm, doesn't have the same code as I've seen fixed, but I don't know if that means it doesn't have the bug
[21:37] <afflux> sticking to my theory (ctrl-c at the prompt): we probably don't have too many reports for that, because most users don't use the readline frontend.
[21:37] <chrisccoulson> i'm not going to claim i know anything about ucf;)
[21:37] <james_w> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=474587
[21:37] <chrisccoulson> anyway, i'll mark the two bugs as duplicates for now, as it is obvious that something goes wrong
[21:38] <chrisccoulson> thanks james_w
[21:38] <james_w> it may be a red herring of course
[21:39] <chrisccoulson> i just thought actually - you say that the postinst script doesn't look like others that have been fixed previously. i suppose it could have already been fixed, as this bug is from april and i downloaded the latest source package
[21:39] <chrisccoulson> i'll have a look through the change history
[21:41] <james_w> ah, good idea
[21:41] <james_w> --debconf-ok may be a keyword
[21:42]  * afflux bets on the users just missed the dialog ;)
[21:43] <afflux> s/dialog/debconf readline frontend/
[21:54] <chrisccoulson> james_w - i'm just comparing the php5 postinst to the diff in the debian bug you pointed me too
[21:54] <chrisccoulson> the main difference is the addition of the "--debconf-ok" option, which doesn't exist in the php5 postinst script
[21:55] <james_w> yeah, and moving the db_stop
[21:55] <james_w> I don't really know what the cause of the hang without it is
[21:55] <chrisccoulson> me neither, as i don't really know much about ucf
[21:56] <chrisccoulson> just looking at the man page for ucg
[21:56] <chrisccoulson> ucf even
[21:57] <chrisccoulson> http://paste.ubuntu.com/52233/
[21:57] <chrisccoulson> second sentence in
[22:02] <chrisccoulson> bug 243809
[22:02] <chrisccoulson> looks related
[22:13] <chrisccoulson> does the gnome splash screen work for anyone in intrepid?
[22:14] <bdmurray> chrisccoulson: which screen?
[22:15] <chrisccoulson> when i log in from GDM, there doesn't appear to be a gnome splash as the desktop loads
[22:16] <chrisccoulson> i can manually run the helper from the command line after the desktop has loaded, and the splash appears
[22:16] <chrisccoulson> just not when the desktop loads
[22:17] <bdmurray> hmm, I don't recall
[23:10] <murdok> the spanish repository isn't working good... :-/
[23:10] <murdok> W: Imposible obtener http://es.archive.ubuntu.com/ubuntu/dists/intrepid/universe/binary-i386/Packages.bz2  Hash Sum mismatch
[23:13] <Hamra> which reminds me. when the main repo is updated with new packages. how long does it approximately take for the mirrors to get updated?
[23:14] <Nafallo> depends on the mirror
[23:14] <mrooney> anyone know anything about the status of the dawn of ubuntu wallpapers being options for Intrepid since they were relicensed, re bug 214711?
[23:14] <wgrant> Some will take hours, some will take days, some appear to take weeks.
[23:14] <Nafallo> some will be instant
[23:15] <Hamra> so there is real slow mirrors that take weeks? that's just great!
[23:15] <Nafallo> I haven't experienced any.
[23:15] <wgrant> Some particularly pathetic .au ones do.
[23:15] <wgrant> Occasionally.
[23:16] <bdmurray> wgrant: do you know much about powersave or did you just upload it?
[23:16] <wgrant> bdmurray: I just uploaded it a few releases ago, and I've been TIL ever since, and regretting it.
[23:17] <Hamra> a week ago i was wondering why the hell some packages that i'm sure got upgraded were still unavailable, until someone in #kubuntu suggested i use any mirror other than Main Server, and that did the trick
[23:18] <bdmurray> Hamra: https://launchpad.net/ubuntu/+archivemirrors might be interesting
[23:18] <wgrant> Hamra: Those mirrors update instantly. They are the mirrors that mirrors mirror from.
[23:19] <Hamra> thanks :D
[23:19] <bdmurray> wgrant: I ran across bug 256737 and it looked like it might be important
[23:20] <wgrant> bdmurray: No clue.
[23:45] <kees> wow, 273761 is just racking up dups.
[23:45] <greg-g> bug 273761
[23:47] <greg-g> wow
[23:49] <emgent> argh
[23:49] <seb128> something doesn't work correctly, launchpad is supposed to list bugs which have a similar title
[23:50] <mrooney> seb128: maybe users aren't working correctly :)
[23:50] <bdmurray> sbeattie: just committed a bug pattern for that
[23:51] <seb128> so either that's buggy or users just don't read the list in which case launchpad should perhaps force them to enter "yes I read the list and I know there is a bug which has a similar title but I still want to send this one" or something ;-)
[23:52] <sbeattie> yeah, I committed an apport filter pattern for that, but the apport retracers may not be caught up on working through the dupe stacktraces.
[23:53] <bdmurray> ah, that's true
[23:55] <seb128> the retracer queue is empty