[00:58] <bobby> Why do I get this when trying to go to different sites in FF4.0b3pre?
[00:58] <bobby> Your browser sent a request that this server could not understand.
[00:58] <bobby> Size of a request header field exceeds server limit.
[01:00]  * gnomefreak never seen that before
[01:00] <gnomefreak> or heard of it for that matter
[01:00] <gnomefreak> bobby: do you have a link
[01:01] <bobby> Of the siite? It happens when I try to go to YouTube or CNET
[01:01] <bobby> I have a Cookie if you want
[01:01] <gnomefreak> bobby: im gonna guess and say its the server not you, but i would like to test it first
[01:01] <bobby> No
[01:01] <bobby> It happens when going to multiple sites
[01:01] <bobby> They work in Chrome
[01:01] <gnomefreak> i dont have an issue in youtube
[01:01] <bobby> FF4.0b3pre
[01:02] <gnomefreak> 4.0~b3~hg20100730r48391+nobinonly-0ubuntu1~umd1
[01:02] <bobby> yup...
[01:02] <bobby> Here is the cookie I get:
[01:02] <bobby> Cookie: tempSessionId=Cg+ND0xKAmtGtRK1kIY; arrowLat=1280446397983; arrowSpc=20; mad_rsi_segs=ASK05540_10274&ASK05540_10018&ASK05540_10019&ASK05540_10066&ASK05540_10161&ASK05540_10166&ASK05540_10195&ASK05540_10205&ASK05540_10206&ASK05540_10207&ASK05540_10208&ASK05540_10214&ASK05540_10219&ASK05540_10221&ASK05540_10225&ASK05540_10236&ASK05540_10241&ASK05540_10242&ASK05540_10258&ASK05540_10264&ASK05540_10267&ASK05540_10268&ASK05540_102
[01:02] <bobby> 83&ASK05540_10288&ASK05540_10289&ASK05540_10290&ASK05540_10314&ASK05540_10322&ASK05540_10333&ASK05540_10339&ASK05540_10342&ASK05540_10343&ASK05540_10349&ASK05540_10389&ASK05540_10432&ASK05540_10483&ASK05540_10521; XCLGFbrowser=Cg+IKExKAmxOAQAAXgA; MADUCAT=1&0729&EX; cnet_joinCallout=true; purs_1=12333b15569cc78cce36ca1b76af9e564c4a027923154810881174864815860519346955!e00!01lWfGxQM1VCryQZeE!wQhb43tUc5EtJ58r8oRvrdtOenwgf2wDLSefK%2FKE
[01:02] <bobby> b60hh39R66BioA%3D%3D; MADCAPP=0a3klD=4&3klE=5&3jlu&3hlt=3; arrowFdCounter=-1; globid=1.lWfGxQM1VCryQZeE; cnet_rvpCallout=3; ZZFLSH=29; nemo=mvt:8:14:1; s_vsn_diggcomsyndication_1=9563975783937; arrowLat=1280502295879; arrowSpc=8; mad_rsi_segs=ASK05540_10274&ASK05540_10018&ASK05540_10019&ASK05540_10066&ASK05540_10161&ASK05540_10166&ASK05540_10195&ASK05540_10205&ASK05540_10206&ASK05540_10207&ASK05540_10208&ASK05540_10214&ASK05540_102
[01:02] <bobby> 19&ASK05540_10221&ASK05540_10225&ASK05540_10236&ASK05540_10241&ASK05540_10242&ASK05540_10258&ASK05540_10264&ASK05540_10267&ASK05540_10268&ASK05540_10283&ASK05540_10288&ASK05540_10289&ASK05540_10290&ASK05540_10314&ASK05540_10322&ASK05540_10333&ASK05540_10339&ASK05540_10342&ASK05540_10343&ASK05540_10349&ASK05540_10389&ASK05540_10432&ASK05540_10483&ASK05540_10521; MADUCAT=1&0730&EX; arrowLat=1280516712752; arrowSpc=1; mad_rsi_segs=ASK
[01:02] <bobby> 05540_10274&ASK05540_10018&ASK05540_10019&ASK05540_10066&ASK05540_10161&ASK05540_10166&ASK05540_10195&ASK05540_10205&ASK05540_10206&ASK05540_10207&ASK05540_10208&ASK05540_10214&ASK05540_10219&ASK05540_10221&ASK05540_10225&ASK05540_10236&ASK05540_10241&ASK05540_10242&ASK05540_10258&ASK05540_10264&ASK05540_10
[01:02] <gnomefreak> i could have done without that in here
[01:02] <bobby> oh... sorry :P
[01:03] <bobby> Let's see...
[01:03] <bobby> CENT doesn't work... YouTube
[01:03] <bobby> Facebook works fine though...
[01:03] <bobby> So does Google
[01:04] <bobby> Apache server at www.youtube.com Port 80...
[01:05] <bobby> Any suggestions?
[01:05] <gnomefreak> bobby: any place in Cnet or any of it?
[01:05] <bobby> Nope, none of it works
[01:05] <bobby> Same with YouTube
[01:05] <gnomefreak> deolete your cookies work?
[01:05] <gnomefreak> delete even
[01:06] <gnomefreak> 64bit was the next question
[01:06] <bobby> Okay, that worked...
[01:06] <bobby> Funny, as I cleaned my cache/cookies about 10 minutes ago...
[01:06] <bobby> Thanks
[01:06] <gnomefreak> :)
[01:06] <gnomefreak> np
[01:07] <bobby> That was strange, I've never had that happen before
[01:07] <bobby> Any idea why?
[01:08] <gnomefreak> bobby: nope
[01:08] <bobby> Well, still waiting for Jager engine to be completed
[01:08] <gnomefreak> size of cookie maybe
[01:08] <bobby> I haven't been anywhere with epic cookies?
[01:08] <bobby> Just Facebook...
[01:08] <bobby> owait...
[01:09] <gnomefreak> im seeing that every site you are mentioning is flash releated
[01:09] <gnomefreak> ill be back need to reboot and hope it works
[01:09] <bobby> kk cya
[01:14] <bobby> Hey does anybody know a date for the UI update?
[11:16] <ta_bu_shi_da_yu> hey folks
[11:17] <ta_bu_shi_da_yu> is Micah about?
[11:17] <ta_bu_shi_da_yu> micahg, hey, you about?
[15:56] <micahg> ta_bu_shi_da_yu: hi
[15:56] <ta_bu_shi_da_yu> micahg, hey!
[15:57] <ta_bu_shi_da_yu> you sent me an email...
[15:57] <ta_bu_shi_da_yu> bug 610039
[15:57] <ubot2> Launchpad bug 610039 in firefox (Ubuntu) (and 2 other projects) "firefox-bin crashed with SIGSEGV in sqlite3VdbeExec() (affects: 2) (dups: 1) (heat: 22)" [Medium,Triaged] https://launchpad.net/bugs/610039
[15:57] <ta_bu_shi_da_yu> wooo....
[15:58] <micahg> ta_bu_shi_da_yu: yes, so we use mostly in source libraries for Firefox to match the upstream build as close as possible, sqlite is one of those
[15:58] <micahg> ta_bu_shi_da_yu: so we build a copy of sqlite in Firefox, which is why I said that Mozilla needs to patch their copy
[15:58] <ta_bu_shi_da_yu> ah! sorry, didn't realise
[16:00] <ta_bu_shi_da_yu> doesn't Michael Terry work for Canonical?
[16:00] <micahg> ta_bu_shi_da_yu: if there's a bug in the Ubuntu sqlite3 package, we add a task for it
[16:00] <micahg> ta_bu_shi_da_yu: yes, I think he does
[16:01] <ta_bu_shi_da_yu> I was wondering if he could verify that is indeed the same issue... there's an easy test case
[16:01] <micahg> ta_bu_shi_da_yu: but in this case, the bug doesn't actually affect the Ubuntu package (well, it might, but no one's reported it)
[16:01] <ta_bu_shi_da_yu> well... sqlite3 is definitely affected
[16:01] <ta_bu_shi_da_yu> if the index corrupts, then it segfaults
[16:02] <micahg> ta_bu_shi_da_yu: right, but sqlite3 is used by a lot of apps, so an SRU is higher risk
[16:02] <ta_bu_shi_da_yu> sorry, I'm kind of new to this... SRU?
[16:02] <micahg> ta_bu_shi_da_yu: my guess is Firefox corrupted the index somehow, which is why it's more visible
[16:02] <micahg> !sru | ta_bu_shi_da_yu
[16:02] <ubot2> ta_bu_shi_da_yu: Stable Release Update information is at http://wiki.ubuntu.com/StableReleaseUpdates
[16:02] <ta_bu_shi_da_yu> ah... thanks :-)
[16:03] <ta_bu_shi_da_yu> yeah, not sure how Firefox did this!
[16:03] <micahg> ta_bu_shi_da_yu: Firefox gets regular updates though, so we can easily get the fix in if they take the patch in their source
[16:03] <micahg> ta_bu_shi_da_yu: well, if it crashed at the wrong time, it might have corrupted the index
[16:04] <ta_bu_shi_da_yu> well... I'm not sure that an SRU is necessary then... the issue is fixable by a reindex
[16:05] <ta_bu_shi_da_yu> however... the problem is that when the index corrupts it means you can never run certain queries till you do a reindex
[16:05] <ta_bu_shi_da_yu> and it's fairly tough to work out what's causing the issue...
[16:05] <ta_bu_shi_da_yu> anyway... not a huge issue :-) all those apps that will be affected should really patch their sqlite3.c file
[16:06] <ta_bu_shi_da_yu> I do have a question though... it's about PPAs
[16:06] <ta_bu_shi_da_yu> am I allowed to upload deb files where I've recompiled Firefox but have turned off compiler optimizations?
[16:06] <micahg> ta_bu_shi_da_yu: well, the other apps in teh archive should be using the sqlite3 package
[16:07] <micahg> ta_bu_shi_da_yu: if you disable the official branding, Mozilla is very specific about who can use their branding
[16:07] <ta_bu_shi_da_yu> no branding changes
[16:07] <ta_bu_shi_da_yu> just rebuilt the firefox package with noopt
[16:07] <ta_bu_shi_da_yu> it helps with getting to the bottom of segfaults
[16:08] <micahg> ta_bu_shi_da_yu: yeah, I think you'd have to disable the branding since that could affect performance
[16:08] <ta_bu_shi_da_yu> er... what branding?
[16:09] <ta_bu_shi_da_yu> the package I'm running right now has the same branding... and I've turned off the optimization flags
[16:09] <micahg> ta_bu_shi_da_yu: in debian/rules, I think you have to comment out the BUILD_OFFICIAL line
[16:10] <ta_bu_shi_da_yu> didn't seem to need to...
[16:10] <micahg> ta_bu_shi_da_yu: it's not needed to compile, it's needed to upload
[16:11] <ta_bu_shi_da_yu> ah... sorry, I'm following :-)
[16:11] <ta_bu_shi_da_yu> so if I do this, then I can upload?
[16:12] <ta_bu_shi_da_yu> it might help others with segfaults
[16:12] <ta_bu_shi_da_yu> I couldn't make any headway on that bug till I change the compilation option because for some reason the p variable was optimized out...
[16:13] <micahg> ta_bu_shi_da_yu: well, what happens, you get less segfaults with the compiler optimizations off?
[16:13] <ta_bu_shi_da_yu> nope... when I get a stacktrace I can see more relevant variables
[16:13] <ta_bu_shi_da_yu> so for instance, the sql that was causing the crash got optimized out
[16:14] <ta_bu_shi_da_yu> when I found the sql statement that caused the problem, I loaded places.sqlite into the sqlite3 shell, ran the query and reproduced the segfault
[16:14] <micahg> weird, but maybe I don't understand compiler optimizations well enough
[16:14] <micahg> chrisccoulson: around?  do you see the above as a bug in Firefox?
[16:14] <ta_bu_shi_da_yu> not a bug....
[16:15] <ta_bu_shi_da_yu> a way of tracking down the bug more easily :-)
[16:15] <ta_bu_shi_da_yu> when you are developing/debugging it's standard practice to compile without optimizations
[16:15] <micahg> ta_bu_shi_da_yu: right, but maybe we can do something more official to help...
[16:15] <ta_bu_shi_da_yu> when you do a final release, you turn on the compiler optimizer
[16:16] <ta_bu_shi_da_yu> :-) well, we provide debug symbols, which is quite cool
[16:16] <ta_bu_shi_da_yu> it would be nice to have packages without optimization, of course only those actively tracking down segfaults where the optimization gets in the way would need it
[16:17] <ta_bu_shi_da_yu> but for firefox, it's a huge package (takes up lots of disk space to compile!) so it would be handy to have a non-stripped, non-optimized version for this sort of thing
[16:17] <ta_bu_shi_da_yu> it was also a pain having to force an uninstall for firefox, then reinstall it again...
[16:17] <ta_bu_shi_da_yu> well... not that much of a pain :-)
[16:23] <ta_bu_shi_da_yu> anyhow, I gotta get going - thanks for your time micahg :-)
[16:24] <micahg> ta_bu_shi_da_yu: no problem, feel free to come back with any questions, and I'll see what we can do about debug builds
[16:25] <ta_bu_shi_da_yu> thanks :-) It's rare I get crashes in Firefox... but I've been looking through the Mozilla bug database seeing if I can repro a few now that I've got a debugging version installed :-)
[16:25] <ta_bu_shi_da_yu> night all!
[16:30] <BUGabundo> that nick scared me
[17:19] <vish> hmm , is there a way for me to assign a shortcut for search in FF ?  like if i highlight a word in say xchat , and do a "alt+s" , it will open a search in FF?
[17:20] <micahg> vish: I would guess jsut append to a goole query
[17:20] <micahg> *google
[17:20]  * vish tires..
[19:33] <fta> micahg, http://people.ubuntu.com/~fta/ppa-dashboard/chromium-daily.html :(
[19:34] <micahg> fta: I gave up after the first 2 rescores, sorry
[19:47] <micahg> fta: I don't know if we'll get them back before alpha 3 is released
[19:47] <micahg> fta: BTW, I just pushed tb3.1
[19:47] <micahg> bzr update
[20:05] <fta> micahg, i'm not blaming you, it's just sad. they claim they want dailies, yet they don't have the resources
[20:06] <micahg> right, and now they have soruce recipies to generate more builds
[20:06] <micahg> at least amd64 is buildign this time
[20:55] <chrisccoulson> grrrrrr, ubuntuzilla
[20:55]  * chrisccoulson rages
[20:58] <micahg> chrisccoulson: I think the .la file issue is a regression by virtual of gdk-pixbuf being repackages
[20:58] <micahg> *repackaged
[20:58] <micahg> *virtue
[20:58] <chrisccoulson> micahg - the la files have been deliberately removed due to a policy change, i think
[20:59] <micahg> chrisccoulson: right, but packages that ship them are supposed to continue to ship them until they are remvoed
[21:00] <micahg> chrisccoulson: http://www.debian.org/doc/debian-policy/ch-files.html#s-libraries, near footnote 71
[21:01]  * micahg needs to read the whole manual one day
[21:02] <chrisccoulson> micahg - ah, ok. i'll try and have  a look at that in a bit. it's g-c-u that is failing isn't it?
[21:03] <micahg> chrisccoulson: yeah
[21:04] <micahg> chrisccoulson: actually, I think I should just build against libgdk-pixbuf, that might fix it, let me test
[21:07] <micahg> or rather libgdk-pixbuf2.0-dev
[21:12] <micahg> nm, that was silly, it's a depends already :(