UbuntuIRC / 2010 /05 /04 /#ubuntu-mozillateam.txt
niansa
Initial commit
4aa5fce
[00:13] <mahfouz> we need new dreams tonight
[00:26] <LLStarks> MOJIRA!
[02:36] <mahfouz> http://www.google.com/reader/next?go=noitems
[03:55] <micahg> ddecator: ff37 is good to know if something's fixed or not, but as far as Ubuntu goes, it's triaged until 3.7 reaches the devel release
[03:55] <micahg> ddecator: it also means most likely there's an open upstream issue
[03:56] <ddecator> micahg: alright, i'll take a look upstream then
[04:10] <ddecator> micahg: huh, i can't find an upstream report, or a dupe downstream. i'm guessing i should still make an upstream report even though it's fixed in 3.7?
[04:11] <micahg> ddecator: depends how bad it is
[04:11] <micahg> ddecator: which bug?
[04:12] <ddecator> micahg: bug 573365, just some text overlapping in svgs
[04:12] <ubottu> Launchpad bug 573365 in firefox "SVG fonts rendered wrong" [Undecided,New] https://launchpad.net/bugs/573365
[04:20] <micahg> ddecator: sorry, too tired to make sense of it right now
[04:21] <ddecator> micahg: haha, it's alright. it's just an appearance issue, so i'm not sure the upstream devs would care too much about it.
[04:22] <micahg> ddecator: well, rendering issues are important, but if it's not an issue in 3.7, then idk if they'd fix it
[04:22] <ddecator> micahg: right, that's what i meant. if it wasn't fixed in 3.7, i would definitely report it. should i report is just in case or wait until we're more awake? haha
[04:24] <micahg> ddecator: well, there are quite a few SVG bugs, I just can't think of what the proper search terms are ATM
[04:24] <ddecator> micahg: yah, i tried multiple searches and couldn't find anything. i'll mark it confirmed for now and try some more searches later
=== almaisan-away is now known as al-maisan
[08:39] <BUGabundo_remote> all HAIL the mighty Summer... oh wait!
[10:14] <fta2> BUGabundo_remote, ch fixed
[10:17] <BUGabundo_remote> really?
[10:17] <BUGabundo_remote> let me test
[10:17] <BUGabundo_remote> I must admit, it was nice to return to Firefox
[10:17] <BUGabundo_remote> its an wonderful browser
[10:18] <BUGabundo_remote> fta I confirm. Ch fixed
[12:02] <BUGabundo_remote> wth http://www.omgubuntu.co.uk/2010/05/canonical-licenses-h264.html
[12:02] <BUGabundo_remote> o
[12:22] <Dimmuxx> that's pretty crazy
[13:47] <fta2> BUGabundo_remote, i can't stand ff anymore. it's really too slow. My own html5 webapps (using canvas) are at least 5 times slower in ff
[13:48] <BUGabundo_remote> its still fast
[13:48] <BUGabundo_remote> just not as fast as Ch
[13:48] <BUGabundo_remote> but its really nice... plus I have there the single addon I can't find for chromium
[13:49] <BUGabundo_remote> shame it takes so long to open/load
[13:53] <fta2> which addon?
[13:55] <BUGabundo_remote> nosquint
[14:03] <fta2> BUGabundo_remote, tried https://chrome.google.com/extensions/detail/jgfonhdeiaaflpgphemdgfkjimojblie ?
[14:06] <BUGabundo_remote> no
[14:06] <BUGabundo_remote> let me see
[14:07] <BUGabundo_remote> fta I have this https://chrome.google.com/extensions/detail/ombpcpigmndepfckcifdblemkabaoihk
[14:07] <BUGabundo_remote> same guy
[14:07] <BUGabundo_remote> but stop working a few builds ago
[14:09] <BUGabundo_remote> nope, doesn't work as nosquint :(
[14:13] <BUGabundo_remote> http://mozillalinks.org/wp/2010/05/bug-in-ubuntu-10-04-causing-firefox-crashes-with-noscript/
[14:13] <BUGabundo_remote> boooo
=== JanC_ is now known as JanC
=== al-maisan is now known as almaisan-away
=== almaisan-away is now known as al-maisan
[15:36] <chrisccoulson> hey micahg - i'm going to get FIREFOX_3_6_4_BUILD3 in to the u-m-s PPA once i've built and tested it. are there any changes you want to get in?
[15:39] <micahg> chrisccoulson: what are we allowed to take?
[15:41] <micahg> chrisccoulson: make sure you add asac's ARM patch
[15:41] <micahg> s/patch/fix
[15:42] <chrisccoulson> i'm just looking at the changes in lp:firefox/3.6 now to see if there are any that should go in lucid
[15:42] <chrisccoulson> i think we have all the important ones already
[15:42] <chrisccoulson> micahg - do you mean asac's change for the optimisation on arm?
[15:42] <chrisccoulson> that's already in lucid
[15:42] <asac> ack ... at least i think the one we needed should already be in the lucid upload
[15:42] <micahg> chrisccoulson: yeah
[15:42] <chrisccoulson> yeah, that one's in the current release already
[15:43] <micahg> ah, it's in there
[15:44] <micahg> don't we have to SRU the apparmor updates rather that throw them in with the USN?
[15:45] <chrisccoulson> yeah, we would do. i don't think the apparmor updates are important enough to SRU though, as the profile is not enabled by default anyway
[15:45] <chrisccoulson> although
[15:45] <chrisccoulson> any profile changes related to the new plugin-container binary should probably go in
[15:46] <jdstrand> I believe the apparmor updates are SRUable
[15:46] <micahg> chrisccoulson: I thought the profile was enabled in Lucid
[15:46] <jdstrand> micahg: no
[15:46] <jdstrand> still disabled by default
[15:46] <micahg> ok
[15:46] <jdstrand> but it is a supported profile
[15:47] <chrisccoulson> jdstrand, oh, ok. i think the original question still stands though
[15:47] <chrisccoulson> (do we do a separate SRU or update them with the security update)
[15:47] <micahg> chrisccoulson: the plugin-container change is for a ff/xul split build
[15:47] <jdstrand> technically, yes, they need an SRU. that said if the new firefox requires an apparmor update to work, then no SRU
[15:48] <jdstrand> I'd rather not have the archive churn though
[15:48] <jdstrand> maybe we can get pitti to ack it as SRU-worthy and just shove them into the security update
[15:49] <chrisccoulson> yeah, that might be possible
[15:49] <jdstrand> chrisccoulson: which ones are we talking about?
[15:49] <chrisccoulson> jdstrand, i've not looked at them in detail yet, but i just noticed the last 4 commits are yours ;)
[15:50] <chrisccoulson> so, those are the ones i'm referring to
[15:50] <jdstrand> chrisccoulson: in 3.6.head?
[15:50] <chrisccoulson> jdstrand, yeah
[15:51] <jdstrand> chrisccoulson: let me look at them
[15:51] <chrisccoulson> thanks
[15:54] <jdstrand> chrisccoulson: I'll take this to #ubuntu-devel and ask pitti's opinion
[15:56] <micahg> chrisccoulson: I have to run, but we should probably chat later
[16:53] <asac> bug 580173
[16:53] <ubottu> Error: Launchpad bug 580173 could not be found
[16:53] <asac> debian bug 580173
[16:53] <ubottu> Debian bug 580173 in xulrunner-1.9.2 "xulrunner-1.9.2: libmozjs3d dependency not tight enough" [Normal,Open] http://bugs.debian.org/580173
[16:53] <asac> chrisccoulson: more on mozjs being not good as sys lib ;)
[16:55] <fta> == Scope of Canonical's acquired ffmpeg patent licenses for derivatives ==
[16:55] <fta> * This is believed to be strictly an OEM business, and not apply to Ubuntu in general
[16:55] <fta> * [ACTION] Martin to confirm with Canonical's legal department and follow up
[16:55] <fta> BUGabundo_remote, ^^
[16:55] <BUGabundo_remote> thanks
[16:56] <BUGabundo_remote> OEMs
[16:56] <asac> wsa that TB meeting?
[16:56] <BUGabundo_remote> I thought that as well
[16:56] <chrisccoulson> asac - heh, interesting
[16:56] <chrisccoulson> thanks
[17:30] <fta> the builders are once again gone
[17:32] <BUGabundo_remote> :(
[17:36] <asac> fta: for how long?
[17:37] <fta> no idea, it started a few hours ago, most are red tagged as "disabled"
[17:37] * asac checks
[17:41] <chrisccoulson> jdstrand - did you manage to speak to pitti about updating apparmor profiles?
[17:42] <chrisccoulson> i'm nearly in a position to upload build3 to the u-m-s PPA
=== al-maisan is now known as almaisan-away
[18:18] <fta> asac, we really need the i386 builders. 34h of q :(
[18:18] <asac> fta: some of that is low prio rebuilds
[18:19] <asac> fta: so there currently is a bug in soyuz and the team is fighting it
[18:19] <asac> some xmlrpc timeout issue causes the builders to get auto disabled
[18:19] <asac> so thats fine ;)
[18:19] <asac> just stay tune
[18:19] <asac> d
[18:27] <jdstrand> chrisccoulson: I asked, he didn't answer
[18:32] <chrisccoulson> jdstrand - ok, thanks.
[18:33] <chrisccoulson> micahg - did you do any work on gjs/gnome-shell SRU ?
[18:33] <micahg> chrisccoulson: not yet
=== almaisan-away is now known as al-maisan
[18:34] <chrisccoulson> micahg - that's ok. do you think you'll have time this week? if not, then i can look at that
[18:34] <chrisccoulson> i'd like to fix that before UDS really, so that it's ready for the security update
[18:34] <fta> linux/beta (5.0.342.9 -> 5.0.375.29)
[18:34] <fta> so lucid already needs a bump
[18:34] <fta> asac, ^^
[18:35] <micahg> chrisccoulson: yeah, I might be able to do that
[18:35] <chrisccoulson> micahg - cool, thanks
[18:36] <micahg> chrisccoulson: sounds good, I'd like to chat with you in about 5 hours if you'll be available
[18:37] <chrisccoulson> i shall probably still just be available in 5 hours
[18:39] <asac> fta: any security issues fixed in that bump?
[18:40] <asac> i would suggest to only roll bumps when that happens. otherwise you go crazy. also that makes it easier to justify
[18:40] <fta> asac, http://googlechromereleases.blogspot.com/2010/05/beta-channel-update.html
[18:41] <asac> so no security issues
[18:41] <asac> usually they seem to do that
[18:41] <asac> hmm.
[18:41] <asac> not sure if for dev channels
[18:42] <asac> can you find that out?
[18:45] <fta> asac, what do you mean?
[18:47] <asac> how we can track whether there were security issues fixed in a beta release.
[18:47] <asac> do they include that in the announcement as they do for stable
[18:48] <asac> or do we need to assume if a stable security update was rolled that this one also includes those fixes?
[18:48] <asac> maybe you can ask one of your chromium contacts how they communicate security fixes in beta?
[18:49] <asac> or whether they could include that info to make it easier for us to maintain this in stable rleeases?
[18:53] <fta> asac, check your mails
[18:55] <asac> fta: in your mail i dont see anything about beta channel announces
[18:55] <asac> for stable what i see on releaseblog looks good enoug ... of course CVEs would be better
[18:55] <fta> re-read then
[18:56] <asac> i dont see anything there ;)
[18:56] <asac> explain to me here
[18:56] <fta> it was about our last update of ffmpeg with tons of CVEs, but nada in the release blog
[18:56] <asac> i see that you refer to full release blog ... but i am not sure they undersand that includes beta channel announces
[18:56] <fta> as i said, it's all the same for them
[18:56] <asac> right. but most likely the same went out to stable
[18:57] <asac> ok. so that means if in between our last release from beta and next release there was a security announce for stable, we need to update?
[18:57] <asac> thats all i wanted to understand
[18:58] <fta> hm, no
[18:58] <fta> between the two beta, dev baked long enough and proved to bring lots of improvements to justify to promotion to new beta target
[18:58] <asac> assuming they announce all security fixes for stable ... whats the right procedure to identify the beta uploads then? the last beta channel update on release blog is from apr 7
[18:59] <asac> hmm ok
[18:59] <fta> -to+the
[18:59] <asac> so you say that if they announce a stable security updates there always is a beta update?
[19:00] <fta> sort of
[19:00] <fta> unless it's too late
[19:01] <fta> http://paste.ubuntu.com/427803/
[19:01] <asac> so they skipped a few on beta channel?
[19:02] <asac> is our current one from apr 7?
[19:02] <fta> http://paste.ubuntu.com/427804/
[19:03] <asac> seems so... so question still stands if that means that all the security fixes announced/released for stable in the meantime have been fixed on beta too :(
[19:03] <fta> yes, i've requested a more informative changelog wrt security fixes, but that have yet to happen
[19:17] <fta> asac, so what should i do then?
=== al-maisan is now known as almaisan-away
[20:18] <chrisccoulson> asac - are we going to be supporting security updates for seamonkey in a similar manner to firefox?
[20:18] <chrisccoulson> (ie, should i get them hosted in the u-m-s PPA)
[20:32] <asac> yes. thats the idea
[20:32] <asac> just no USNs
[20:32] <asac> asking folks to test there would be great also to get more testers again etc.
[21:19] <chrisccoulson> asac - cool, i will get 2.0.5 in to the PPA later then
[21:43] <micahg> chrisccoulson: was 2.0.5 tagged?
[21:43] <micahg> yeah, I see they just tagged it
[21:43] <micahg> chrisccoulson: I have one fix before you do that
[21:43] <micahg> the sparc FTBFS
[21:44] <chrisccoulson> micahg - ok, no worries
[21:44] <asac> chrisccoulson: for 1.x to 2.0.x we should maybe use an intermediate -proposed step though ... and carefully evaluating feedback before pushing to -security/-updates
[21:44] <asac> chrisccoulson: howver, procedure is still the same ... push to security ppa with -security targettted
[21:44] <asac> let jdstrand put that to -proposed for baking and have tracker bugs for that update so community can comment on regresssions there
[21:44] <asac> (and have that in changelog)
[21:45] <asac> there will be regressions, but we should understand how bad they are etc.
[21:45] <chrisccoulson> asac - we've already done the 1.x to 2.0.x transition (unless you're referring to older releases)
[21:45] <asac> chrisccoulson: in stable releases?
[21:45] <asac> != lucid?
[21:45] <ubottu> Error: I am only a bot, please don't think I'm intelligent :)
[21:45] <micahg> asac: same for FF36 update?
[21:46] <chrisccoulson> asac - it's only in lucid for now
[21:46] <micahg> asac: and TB3?
[21:46] <asac> definitly ... FF36 needs communication with jdstrand on the way forward. TB3 i would hope we can keep on doing backports
[21:46] <chrisccoulson> asac - also, do you think we should get ff3.6.4 in hardy this week? (even if we don't actually switch the default just yet)
[21:46] <asac> usually there is one or two mailnews patches 3 month that needs backporting
[21:47] <asac> thats doable, but chrisccoulson has to decide on the way forward. i wouldnt do TB3 before FF36 is out
[21:47] <micahg> asac: can I do the thunderbird stable PPA now though?
[21:47] <asac> chrisccoulson: not sure what you mean with "not switching the default"
[21:47] <micahg> I waited until we got the builders back
[21:47] <asac> chrisccoulson: we should get it in -securirty ppa and ask for testing everywhere
[21:47] <asac> maybe ven -devel announce as this
[21:48] <chrisccoulson> asac - ok, i'll get it in to the PPA first. the only thing blocking that currently is it needs a newer mozilla-devscripts
[21:48] <asac> right. i dont know if its ready
[21:48] <asac> what about all the extensions in hardy?
[21:48] <asac> ubufox clearly needs to be staged too
[21:48] <asac> not sure if ew should do all the staging outside of the main security ppa
[21:49] <chrisccoulson> yeah, i suspect nearly all of the hardy extensions will need updating. i need to build a list of those though so i know how much work is involved
[21:49] <asac> just in case there isa firedrill ... but i guess in that case we could also use a different ppa
[21:49] <asac> i would make it dependent on how well i feel prepared for that
[21:50] <asac> e.g. put everything in a ppa ... test ... if all the bits are there, push to seucirty ppa or something
[21:50] <asac> chrisccoulson: in hardy the app-install-data should be quite complete
[21:50] <asac> so we are lucky ... but intrepid/karmic etc. need to get 3.6 too
[21:50] <asac> problem there is that we offer upgrade to intrepid ... but that hsa an older versoin the ... so be careful
[21:50] <asac> to be safe the transition has to bubble from top to bottom
=== almaisan-away is now known as al-maisan
[21:51] <asac> and we need to test 3.0 -> 3.6 upgrade path ;)
[21:51] <asac> pain
[21:51] <micahg> asac: s/intrepid/jaunty
[21:51] <micahg> intrepid = EOL
[21:51] <asac> micahg: well. i think we dont offer upgrades to jaunty. so interesting question. maybe lucid is the only choice we offer still ... or intrepid (which is broken)
[21:52] <chrisccoulson> i assume we need to rename the source for hardy to firefox-3.6 don't we?
[21:52] <micahg> asac: I'm confused
[21:52] <micahg> chrisccoulson: why not use the firefox-3.0 source?
[21:52] <micahg> chrisccoulson: or just firefox
[21:53] <chrisccoulson> micahg - ok, we could use firefox-3.0. firefox is already used in hardy for shipping the old 2.0 in universe
[21:56] <asac> currently trying to find out if we need to care about hardy -> intrepid->jaunty upgrade path at all still
[21:56] <asac> otherwise we can just push 3.6 to hardy and only test the 3.6 -> 3.6 upgrade path for hardy->lucid
[21:58] <micahg> asac: ah, ok
[22:04] <chrisccoulson> i suppose if we still have to care about the hardy->intrepid->jaunty upgrade path, then we'd need to do the work in jaunty first
[22:04] <chrisccoulson> whereas if we don't care about that path, then we can update hardy first
[22:04] <asac> yeah.
[22:05] <asac> 3.0 is EOL ... 3.5 i am not so sure ;)
[22:05] <asac> was there any final say
[22:05] <asac> ?
[22:05] <chrisccoulson> 3.5 doesn't seem to be EOL just yet
[22:06] <chrisccoulson> there's a 3.5.10 scheduled according to https://wiki.mozilla.org/Releases/
[22:06] <asac> right so we could get away with staged rollout of hardy + lucid ... and then jaunty + karmic later
[22:06] <chrisccoulson> in fact
[22:06] <asac> which would allow better planning and remove a bit of the pressure
[22:06] <asac> what a nice coincident ;)
[22:06] <chrisccoulson> we already have FIREFOX_3_5_10_RELEASE ;)
[22:06] <asac> hmm
[22:07] <chrisccoulson> i suppose i should get that updated tonight really
[22:07] <asac> BUILD1?
[22:07] <asac> or are we late?
[22:07] <chrisccoulson> asac - there's a BUILD1 tag too
[22:07] <asac> right. but it wasnt released yet, i assume? its in beta?
[22:07] <chrisccoulson> the release isn't scheduled until may 13th
[22:07] <asac> right. so its beta most likely or even beta prepration still
[22:07] <chrisccoulson> so, yeah, it's in beta now. so, i should get that in the PPA too
[22:07] <asac> getting BUILD1 would be good i guess
[22:22] <fta> asac, i have 5.0.375.29~r46008-0ubuntu1 ready, i'm not sure what i should do with it now
[22:23] <fta> and maverick is not open yet i assume
=== al-maisan is now known as almaisan-away
[22:33] <micahg> chrisccoulson: the _RELEASE tags are tagged, but moved if needed, so if we use it before release, we use the BUILDX tag, if after release, the RELEASE tag
[22:33] <chrisccoulson> micahg - ok, thanks. i was a bit confused about that
[22:34] <chrisccoulson> i'm just doing the 3.5.10 build1 update now anyway
[22:34] <micahg> chrisccoulson: we've got at least another month on 3.5.x
[23:00] <micahg> chrisccoulson: so, should I push thunderbird-stable tonight?
[23:01] <micahg> chrisccoulson: PPA I mean
[23:04] <asac> fta: i thought it was open
[23:04] <fta> i didn't see anything building except doko's packages
[23:05] <micahg> asac: no, not yet
[23:05] <asac> fta: i would document the security issues supposely fixed in changelog ... then discuss with jdstrand how to get this up
[23:05] <fta> which security issues?
[23:45] <micahg> asac: is there a (spoken) language I should brush up on for Brussels?
[23:53] <mahfouz> french
[23:53] <mahfouz> Bruxelles
[23:56] * micahg will try