[00:00] <asac> Jazzva: meeting is still continuing
[00:00] <Jazzva> ok, will buzz in 10 minutes :)
[00:09] <asac> give me 5 more
[00:09] <asac> or 10
[00:09] <asac> still running :(
[00:12] <Jazzva> sure thing
[00:17] <asac> bug 254117
[00:20] <Jazzva> asac, I noticed that today
[00:21] <Jazzva> on one flash, but I can't remember where
[00:21] <asac> Jazzva: is that the same in nspluginwrapper?
[00:21] <fta> same on google maps
[00:21] <fta> without flash
[00:21] <asac> oh its just some sites?
[00:21] <asac> fta: what?
[00:22]  * asac goes to maps
[00:22] <fta> ff context-menu covers javascript overlay menus
[00:22] <Jazzva> asac, happened only once, I'll check maps now
[00:22] <asac> fta: for me the maps page doesnt open a context menu on its own for me
[00:22] <asac> fta: or is that it just covered?
[00:23] <Jazzva> ah, it happens here
[00:23] <fta> hm, at work, i get two menus, here, it looks fine
[00:24] <asac> fta: i only get the firefox one
[00:24] <asac> if you get two, it cant be right to get just that
[00:24] <asac> Jazzva: what happens for you?
[00:24] <fta> the right click is supposed to show a google menu to set directions
[00:24] <Jazzva> let me take a screenshot
[00:24] <asac> ha ... i think i saw it ;)
[00:24] <asac> no need to
[00:24] <asac> strange
[00:25] <asac> wtf
[00:26] <Jazzva> ok
[00:27] <Jazzva> asac, 3 minutes
[00:29] <asac> Jazzva: ok let me get that thing ;)
[00:30] <Jazzva> now, that's just-in-time ;)
[00:42] <asac> Jazzva: rev 36?
[00:43] <Jazzva> yep
[00:43] <Jazzva> asac, ^
[00:45] <asac> Jazzva: ok ... pushed your branch to ubuntu-dev
[00:45] <Jazzva> asac, yay
[00:45] <Jazzva> thanks :)
[00:45] <asac> Jazzva: ok. next time try to be more precise in changelog ;)
[00:45] <asac> especially no typos ;)
[00:46] <asac> uploaded
[00:46] <Jazzva> i had a typo?
[00:46] <asac> Jazzva: let me know when the binaries are built
[00:46]  * Jazzva is ashamed
[00:46] <Jazzva> sorry :)
[00:46] <asac> we need to push some admin to NEW the i386 one
[00:46] <Jazzva> ok, I'll check to see when it gets built
[00:47] <Jazzva> ah, 'depenedency'
[00:55] <asac> Jazzva: the branch is still in knit format ;)
[00:56] <Jazzva> yeah, it reported that... didn't know if I should upgrade
[00:56] <Jazzva> (if maybe you/mt have a branch which is still in that format)
[00:57] <asac> Jazzva: you can upgrade
[00:57] <asac> though now i have to do that for ubuntu-dev ;)
[00:57] <asac> but maybe try ... if it works i can do it there too
[00:57] <Jazzva> ok
[00:57] <asac> urgh
[00:58] <asac> nsplugintwrapper build is in state CHROOTWAIT
[00:58] <Jazzva> what's that state?
[00:58] <asac> https://edge.launchpad.net/ubuntu/+source/nspluginwrapper/1.1.0-0ubuntu1
[00:58] <asac> not sure what that means
[00:58] <asac> trying to figure out now
[00:59] <Jazzva> E: Could not get lock /home/buildd/build-702715-1742497/chroot-autobuild/var/cache/apt/archives/lock - open (11 Resource temporarily unavailable)
[00:59] <Jazzva> E: Unable to lock the download directory
[00:59] <Jazzva> usually happens when two apt process are called at the same time
[00:59] <Jazzva> *processes
[01:01] <Jazzva> can you rerun the build? maybe it will help
[01:02] <asac> Jazzva: yellow has a problem
[01:02] <asac> and since its idle its unlikely that we will get a slot on "crested"
[01:02] <asac> we will always get bounced to that broken buildd
[01:02] <asac> https://edge.launchpad.net/+builds/crested/+history
[01:03] <Jazzva> sigh
[01:03] <asac> https://edge.launchpad.net/+builds/yellow/+history
[01:03] <asac> https://edge.launchpad.net/+builds/
[01:05] <asac> Jazzva: yeah. most likely a left over. i hope archive admins will at least disable that buildd
[01:05] <Jazzva> let's hope for that
[01:05] <Jazzva> can we notify them somehow?
[01:05] <asac> hmm ... i386 appears to be ACCEPTED ... not NEW
[01:05] <asac> strange
[01:05] <asac> now DONE
[01:06] <asac> maybe it goes into NEW after done?
[01:06] <asac> Jazzva: best channel to inform them is probably launchpad
[01:06] <asac> i asked on a company internal channel ... so lets hope
[01:06] <Jazzva> ok, I'll go to launchpad
[01:08] <Jazzva> sent a notice
[01:09] <Jazzva> I'll wait to see if someone will react
[01:09] <asac> yeah. i think they should know now
[01:09] <asac> lets move on ;)
[01:12] <Jazzva> asac, looks like nspluginwrapper is published https://edge.launchpad.net/ubuntu/+source/nspluginwrapper
[01:13] <asac> thats what i am wondering
[01:13] <asac> is DONE really "released" ... or will NEW come on next publisher run for binary packages?
[01:13] <Jazzva> I'll see if I can upgrade to it now
[01:15] <Jazzva> does ddebs exist only for dbg packages?
[01:15] <asac> no
[01:15] <asac> only for non-dgb package ;)
[01:15] <asac> every package with binary or binary-lib should have one
[01:15] <asac> everything else is a bug
[01:15] <Jazzva> aha...
[01:15] <asac> (like what we have now for firefox/xul)
[01:15] <Jazzva> so, -dbgsym is a ddeb, and normal package is deb?
[01:15] <asac> some packages might produce bad dbgsym
[01:16] <asac> that is the case when they strip the symbols during package build
[01:16] <asac> which is why we pass --disable-strip and CFLAGS=-g to firefox/xul builds
[01:16] <asac> -g to enable debug symbols at all
[01:16] <Jazzva> I see
[01:16] <asac> dh_strip then does the magic rumbling and removes those symbols from the binaries while putting them in to symbol .so files
[01:17] <asac> ;)
[01:17] <Jazzva> asac, it doesn't show up in the upgrade
[01:17] <asac> Jazzva: well. publisher runs every 20-30 minutes i think
[01:17] <Jazzva> yeah, I was just checking if it somehow finished here :)
[01:18] <asac> Jazzva: this quetsion just was discussed in -devel :)
[01:18] <asac> 02:14 < cjwatson> ogra: the cron job is 0-50/5; the 55 slot is used for security uploads
[01:18] <asac> 02:14 < cjwatson> ogra: so an upload at :51 would wait nine minutes or so for processing, but would  still be safe to get into the :03 publisher run
[01:18] <Jazzva> aha, ok
[01:18] <asac> so every 5 minutes
[01:19] <asac> not sure where that goes to though
[01:19] <asac> maybe its a more or less internal master mirror
[01:19] <asac> from where the main tier mirrors sync from
[01:26] <asac> splitting fun here
[01:26]  * asac wonders when this whole freenode world will fall apart
[01:27] <Jazzva> because of netsplits?
[01:27] <asac> yeah
[01:27] <asac> ;)
[01:32] <asac> Jazzva: so did you upgrad ethe remote branches properly?
[01:32] <Jazzva> not right now... phone :(
[01:32] <asac> sure
[01:32] <Jazzva> yeah, it's pak-0.92 :)
[01:32] <Jazzva> *pack
[22:31] <fta> jemalloc doesn't mix with fakeroot
[22:33] <asac> fta: so we kill PPA builders?
[22:33] <asac> or was that just your local builder that blocks?
[22:34] <fta> both
[22:35] <fta> i killed 6 builders
[22:36] <fta> rebuilding to get dbgsyms
[22:39] <fta> fakeroot-ng would probably help here
[22:41] <asac> since when did that start?
[22:41] <fta> last 2 days
[22:41] <asac> did you enable the jemalloc things or what?
[22:41] <fta> i pointed you the commit yesterday
[22:41] <fta> no, bs did
[22:41] <fta> but hgweb/mxr are dead now
[22:42] <fta> mozilla bug 452682
[22:42] <asac> fta: i must have missed that
[22:46] <asac> fta: https://code.edge.launchpad.net/~asac/firefox/firefox-3.0.head.awesome
[22:51] <fta> looks crazy ;)
[22:51] <fta> i hate the new Loggerhead
[22:52] <asac> Loggerhead=?
[22:52] <fta> the bzr web interface on lp
[22:54] <asac> cant tell ;) ... still have to find everything i previously knew ;)
[22:54] <asac> wow there is a wierd bug
[22:54] <asac> flickering
[22:54] <asac> disappearing and reppearing column
[22:54] <asac> http://bazaar.launchpad.net/~asac/firefox/firefox-3.0.head.awesome/changes
[22:55] <asac> when i unfold one item an dhover the second column
[22:55] <asac> s/dhover/hover/
[22:55] <asac> which reads: "/~asac/firefox/firefox-3.0.head.awesome/static/images/treeExpanded.png"
[22:55] <asac> (was hard to select because of thef lickering
[22:56] <asac> thats must be a bug
[22:56] <asac> ;)
[22:56] <asac> should probably not replace that toggle ;)
[22:57] <fta> ok, fakeroot uses LD_PRELOAD and fakeroot-ng doesn't, that's why
[22:57] <[reed]> fta: file a bug? I dunno
[22:58] <fta> i discussed a bit with bs, he wants a trace
[23:06] <fta> [reed], do you get audible sound in ff3.1 <video/> ? i hear statics even with your official nightly builds
[23:07] <[reed]> hmm
[23:08] <fta> for ex, here: http://quetzalcoatal.blogspot.com/2008/08/thanks-dbaron.html
[23:11] <fta> damn, bzr bd changed. it now moves the debs on level above.. but not the dbgsym obviously
[23:11] <fta> asac, ^^
[23:11] <fta> ^on^one
[23:12] <asac> fta: i have a custom --build-dir
[23:12] <asac> so i probably wont notice
[23:13] <fta> me too. so you will
[23:13] <asac> or is --result=... now required
[23:13] <asac> yeah. then --result ;)
[23:13] <asac> (which was broken for me before ... lets hope its fixed)
[23:13] <fta> dpkg-genchanges: including full source code in upload
[23:13] <fta> dpkg-buildpackage: full upload (original source is included)
[23:13] <fta> Placing result in ..
[23:13] <fta> the last line is new
[23:13] <asac> fta: any new options in bzr help builddeb
[23:13] <asac> ?
[23:17] <fta> difficult to say
[23:17] <asac> maybe there is --compat-1.x ;O)
[23:17] <asac> =
[23:17] <asac> ?
[23:19] <fta>   You can also specify directories to use for different things. --build-dir
[23:19] <fta>   is the directory to build the packages beneath, which defaults to
[23:19] <fta>   '../build-area'. '--orig-dir' specifies the directory that contains the
[23:19] <fta>   .orig.tar.gz files , which defaults to '..'. '--result-dir' specifies where
[23:19] <fta>   the resulting package files should be placed, which defaults to '..'.
[23:19] <fta>   --result-dir will have problems if you use a build command that places
[23:19] <fta>   the results in a different directory.
[23:20] <asac> ok ... so --orig-dir is new
[23:21] <asac> and --result-dir as well
[23:21] <fta> why would they change a default like that. it breaks my scripts
[23:21] <asac> well, new default value ;)
[23:21] <asac> fta: they do that to get a more "original" feeling
[23:21] <asac> to build bridges from the "old" tools (e.g. apt-get source debuild -b; or dpkg-buildpackage)
[23:21] <asac> so when you build bzr branches you feel more at home
[23:21] <fta> what if ".." is not writable
[23:22] <asac> fta: we... same couulod have been true for ../build-area in the past
[23:22] <asac> just specify your custom ones
[23:22] <asac> (should remind us that depending on non-explicit defaults ends up causing maintenance work)
[23:32] <fta> http://paste.ubuntu.com/41378/
[23:36] <asac> fta: look what debian is doing
[23:36] <asac> fta: i think they had wierd issues
[23:36] <fta> ?
[23:36] <asac> and mh introduced an env or something to turn jemalloc off during maintainer scripts
[23:36] <fta> they have 3.1 ?
[23:40] <fta> asac, pointers ?
[23:43] <asac> no
[23:43] <asac> they have 1.9
[23:43] <asac> but had similar issues with jemalloc ;)
[23:43] <asac> i just remember all the bugs
[23:44] <asac> fta: http://packages.debian.org/changelogs/pool/main/x/xulrunner/current/changelog
[23:44] <asac> * memory/jemalloc/jemalloc.c: Allow memory to be allocated from a small pool of static memory during initialization. Some LD_PRELOADed libraries such as libaoss and libtrash can divert some functions used by jemalloc initialization, themselves needing malloc or calloc. This used to lead to a deadlock. Closes: #487614.
[23:45] <fta> damn, someone pushed a new appart, by branch merge is doomed
[23:45] <fta> apport
[23:45] <asac> hmm
[23:45] <asac> fta: i doubt
[23:45] <asac> fta: merging should be easy
[23:45] <asac> unless he touched the same
[23:47] <fta> that was it: http://hg.mozilla.org/mozilla-central/rev/12616044efd376d65d505b25c096c86007a9a13c
[23:48] <fta> mike said "set MOZILLA_NO_JEMALLOC=1 for now"
[23:48] <asac> yeah
[23:48] <asac> well ... he has its own wrappper script
[23:48] <asac> and uses LD_PRELOAD
[23:48] <asac> thats what i see now
[23:49] <asac> and the other patch is about something else
[23:49] <asac> e.g. use small pool of mem to not conflict with aoss malloc overloading
[23:49] <fta> but that's just a workaround, not a fix
[23:51] <asac> the small pool thing?
[23:51] <asac> well ...  the fix is probably impossible with fakeroot overloading malloc too
[23:51] <asac> but i have no real clue i have to admit. i just can imagine thats a problem ;)
[23:52] <asac> i am still fighting with diversions and major version upgrades :(
[23:52] <asac> where the pkglibdir changes
[23:53] <asac> but i hope this code is now good for that ;)