/srv/irclogs.ubuntu.com/2010/05/21/#ubuntu-mozillateam.txt

fta*sigh* i trashed my session while debugging and now the snaps are gone00:15
BUGabundo_NCISeheh00:29
BUGabundo_NCISthe perfect solution: start fresh, or else...00:29
BUGabundo_NCISnite00:37
=== BUGabundo_NCIS is now known as BUGabundo_Sleep
=== BUGabundo_Sleep is now known as BUGabundo_with
=== BUGabundo_with is now known as BUGabundo_gouki
ftahttp://arstechnica.com/web/news/2010/05/google-offers-web-designers-hosted-open-source-web-fonts.ars00:44
ftaso much for the mysterious new ubuntu font. google did it 1st ;)00:45
BUGabundo_goukillo00:46
micahgddecator: did you have a chance to check songbird to see if it still builds?05:33
ddecatormicahg: built for me either last night or the night before just fine :)05:35
micahgddecator: k, I'll check the merge sat night and if there are no issues, push it up to PPA05:35
ddecatormicahg: sounds good. let me know if there are any issues or if i need to fix anything05:36
micahgchrisccoulson: I'm back, might not make 11:00 UTC though...05:38
ddecatormicahg: btw, i pushed the latest branch from when i built to my lp account, but the only thing that has changed since i requested the last merge was updates from upstream, nothing i had to change, so i didn't know if you wanted me to bother with another merge request05:44
micahgddecator: nah, that's ok as long as nothing was rebased05:45
ddecatormicahg: alright, that's what i thought05:45
BUGabundo_remoteWelcome to Ibiza Summer Time!09:17
asacBUGabundo_remote: you are on ibiza?09:50
BUGabundo_remoteasac: I wish :)09:51
BUGabundo_remoteasac: still at work.. but ill be taking a week vacation to Algarve next month, if you want to meet up09:52
asacgood. otherwise i would have been unhappy about you having holidays ;)09:52
asacheh09:52
asacunfortunately i wont have time till august from what i see09:52
BUGabundo_remoteshame09:53
asacalways travelling and wrestling ;)09:53
BUGabundo_remoteas long as you are happy, hacking and drunk09:53
BUGabundo_remoteerk, that didn't came out right ;P09:54
asachehe09:59
asacno its fine.09:59
cyrus82Hi. Is anybody working on #507089 (thunderbird segfault with ldap auth in nsswicht)?10:52
ftaasac, why is the n-m ppa stuck?12:38
ftashould i stop the bot?12:38
ftalast update was 3 months ago12:38
asacfta: cyphermox has to open the changelog up again12:44
asacnt sure why that hadent happened12:44
asaci thought he did12:44
asaci will try to remember to ask him when he comes back12:44
ftaok12:48
ftahttp://www.osnews.com/story/23335/Patent_Troll_Larry_Horn_of_MPEG-LA_Assembling_VP8_Patent_Pool12:51
ftaasac, (another chromium beta incoming)12:59
ftaamazon.de seems to snap quite a lot :(13:00
ftaBUGabundo_remote, do you know where all the videos of the google io sessions are? i just see the intro and old stuff13:41
BUGabundo_remoteI saw a link for it today13:42
BUGabundo_remotebut don't have it on hand13:42
BUGabundo_remotethe wave intro should have all the links13:42
ftai read some minutes in wave but no video :(13:42
BUGabundo_remoteif I see it again, ill ping you13:43
BUGabundo_remotereally busy today13:43
BUGabundo_remote150 hosts to _fix_ dns13:43
BUGabundo_remotebut I've seen more ppl complain that not all videos had been published13:44
ftaoh, in the wave modal box, i see a broken puzzle piece, maybe that's why13:46
ftaBUGabundo_remote, grrr "We will not be streaming Google I/O live via the Internet. However, we will be recording most sessions and making them available on http://code.google.com/io and on YouTube after the event. Please check back after June 1st to view the recorded sessions or follow us on @googleio where we will announce when recordings are available."13:53
BUGabundo_remotelol14:03
BUGabundo_remotethere you go14:04
=== yofel_ is now known as yofel
micahg_Tsk_: so, I think I might have an idea about the 64 bit crash for TB, I think it just started me as well, I'm retracing a coredump locally for it18:28
_Tsk_cool !18:28
micahg_Tsk_: I'm wondering if it's the xulrunner-1.9.1 + cairo 1.8.10 issue18:29
_Tsk_micahg: so it would be specific to the linux distri build right ?18:29
micahg_Tsk_: yes18:29
micahg_Tsk_: looks like it's something different18:35
chrisccoulsonmicahg - i noticed that seahorse in hardy builds a plugin for epiphany18:58
chrisccoulsoni'm not sure what to do with that, as that plugin won't work in the webkit world18:58
micahgchrisccoulson: SRU it away?18:58
chrisccoulsonyeah, that might be the only option18:58
chrisccoulsonwhat does asac think?18:58
chrisccoulson;)18:58
asacmy brain is empty ;)19:00
asacchrisccoulson: drop that plugin ... e.g. ship an empty package19:01
chrisccoulsonasac - ok, that's easy enough19:01
chrisccoulsonthanks19:01
asacunless seahorse plugin now exists for webkit19:01
asaccould be that it exists19:01
asacor maybe its built-in now19:01
chrisccoulsoni don't think so. i think the plugin has gone away19:01
chrisccoulsoni'll have a look though19:01
micahgchrisccoulson: oh, one thing I forgot to mention in our chat is about pushed beta full releases to the devel release as well19:08
micahg*pushing19:08
chrisccoulsonmicahg - oh, you mean like pushing build4 now to maverick?19:09
micahgchrisccoulson: yep19:09
micahgchrisccoulson: I can take care of that if you want once I have upload rights19:09
chrisccoulsonthat makes sense. i need to do an upload to maverick this weekend to disable CSD anyway19:09
micahgchrisccoulson: it gives us added beta/pre release testing19:09
micahgchrisccoulson: BTW, there will be a BUILD5 soon19:10
ccheneymicahg, ping19:30
micahgccheney: pong19:31
ccheneymicahg, the dictionary symlink cleaning done by debian i believe will only be the removal of the '-' symlinks, the backwards compat will still be around, but we probably need to change the internal symlink of firefox/thunderbird/etc to use /usr/share/hunspell19:31
micahgccheney: k, that's no problem19:32
ccheneymicahg, i talked to rene and he is not planning on dropping the /usr/share/myspell dir until after the debian release since they are too close now19:32
ccheneymicahg, so the symlinks in it for the '_' dicts will still be around until after debian releases19:32
micahgccheney: as long as we're sure that all the languages are in the hunspell dir, we can switch the symlink for Maverick at any time19:32
ccheneymicahg, they should be afaict but double checking probably wouldn't hurt19:33
micahgccheney: I have a work item for the beta release to check that out19:34
ccheneymicahg, i checked them before lucid release and they looked ok, but another set of eyes wouldn't hurt19:34
ccheneymicahg, well actually there were a few bugs in lucid but i fixed all the ones i found :-)19:34
ccheneymicahg, ok, wasn't sure if the item about symlinks encompassed that too19:34
micahgccheney: k, I'll ping you if I have any questions later with that19:34
ccheneymicahg, ok19:34
micahgccheney: well, the item really should be to make sure that the mozilla apps no longer have dupes :)19:35
ccheneyyea, no dupes will be good too :)19:35
ftaasac, jcastro, jdstrand: I'm preparing yet another version of chromium for maverick, that will be the 3rd update that lucid missed. users keep asking me why the upgrades stopped, should i tell them to use a PPA instead?19:39
jcastrofta: yeah19:40
jcastrofta: asac had a plan for it but I totally forgot19:40
jcastromicahg: do you remember?19:41
jcastrodid we write it down?19:41
ftajcastro, he said the security team was supposed to do it19:41
ftathat has yet to happen19:41
chrisccoulsonright, that's something i'll probably be involved with19:41
asacthe plan is to just do it19:41
asaci dont know who is doing it though19:41
micahgjcastro: yeah, the security team was supposed to do it :)19:41
jcastronow would be a good time to figure it out!19:41
ftashould i just push and wait?19:42
ftaor will it be rejected?19:42
asacfta: push and wait? try to get the process actively started .folks will not find what to push etc.19:42
asacfta: jdstarnd or chrisccoulson should push that to the security ppa19:42
asacjdstrand: ^^19:42
asacfta: you cant push anyway. but having a branch they can use and clear instructions how to upload might help. ask chrisccoulson to help you getting it up19:43
asache knows how to upload security fixes19:43
asacremember to document the security issues fixed19:43
asacotherwise its hard to push out19:43
asacif you dont know what was fixed, then pretend that you know and just document whatever was fixed on stable in between19:43
ftawhat about the exception thingy for browsers?19:44
asaconce we have stable we will push everything19:44
asacfta: not needed. it will just happen19:44
asacfta: there was never an exception to push stuff without security issues ;) ... the idea was to make clear that we will hav eto push securtity with other stuff together19:44
asacbut thats understood. tech board said, its not a tech board issue, so its security team decision how to do that19:44
asactalk to rickspencer319:45
asache will help you19:45
asache wants to push chromium to main ;)19:45
ftathe thing is, each time i'm late for an update, i loose users in favor of chrome19:45
asacright. so it needs to be sorted19:45
asacbut its not a big deal imo atm19:45
asacyou will gain all default netbook installs ;)19:45
asacthats the big cake ... ;)19:46
ftahttp://people.ubuntu.com/~fta/popcon-20100521.png20:10
ftaasac, ^^20:10
jdstrandfta, asac, chrisccoulson, micahg: what did I miss? a chromium update?20:43
jdstrandI was not told I needed to push anything20:43
micahgjdstrand: well, now it's number 3 that Lucid didn't get20:43
jdstrandit is in universe, so it isn't on our team20:43
jdstrandif a community person or someone else wants it updated, I can sponsor it20:44
micahgfta: ^^^20:44
jdstrandI have not been approached to do that20:44
jdstrandI figured it would be dealt with in the same way that seamonkey is-- ie, it is built in a ppa somewhere, gets testing, then I am asked to push through -security20:45
micahgjdstrand: well, we were talking about that a few minutes ago, would you prefer that to just sponsoring?20:45
jdstrandmicahg: if it is a debdiff, I can spnsor. if it is a full release like ff and seamonkey, it should really have a team/person maintaining it like we have for mozilla20:46
jdstrandmicahg: and a process in place to test it20:46
jdstrandlet me put it this way20:46
jdstrandI don't get paid to test seamonkey updates, so the mozilla team/community needs to do all the work on that, and then ask me to push it into -security20:47
jdstrandat that point I just do that-- I don't test it cause the mozilla team/community already did it20:47
jdstrandwe need the same thing to happen with chromium20:48
micahgjdstrand: k, should we use the mozilla-security PPA for that or something else?20:48
ftajdstrand, what i send to the repos already lives in a ppa beforehand20:48
jdstrandmicahg: I don't care-- ubumtu-mozilla-security is nice cause it is already there and non-virtual so I can do a direct pocket copy20:48
jdstrandfta: what kind of ppa is it? virtual or native?20:49
ftajdstrand, atm, it's the chromium beta ppa. but i advise to take either what i send to maverick, or the last tag in the bzr branch20:49
jdstrandfta: so that is a virtual ppa20:50
ftai don't have native ppa unfortunately20:50
ftamaverick is native :)20:50
jdstrandfta: that will have to be rebuilt in -security ppa or corrdinated with the mozillateam ppa20:50
ftajdstrand, can't you take it from maverick/universe?20:50
jdstrandI can't pocket copy it, no20:51
ftaoh20:51
fta*sigh*20:51
micahgfta: does the target for maverick just need to change?20:51
ftamicahg, yes, that and the version number20:51
jdstrandit needs to be built with the same toolchain and libraries for which it is going to land20:51
micahgfta: can you prepare the bzr branch for Lucid and then tell chrisccoulson what to do to generate the upload so he can push to mozilla-security PPA?20:52
fta(feels weird to use a mozilla-* location)20:53
ftabut well..20:53
micahgfta: I think we're all intertwined anyways ;)20:53
jdstrandfta: the alternative is to create source packages for each release you want to provide them for, then someone from the ubuntu-security team can upload them to ubuntu-security-proposed20:54
jdstrandfta: and then ultimately get them into -proposed20:54
jdstrandfta: that would be a way to use the existing infrastructure and get people to try out the builds before pushing them to -security20:55
jdstrandfta: basically, the process is not defined for chromium, and it needs to be (until it ends up in main, at which point chrisccoulson/micahg or someone will likely start working on it)20:55
jdstrandfta: fyi, if going the ubuntu-security-proposed route: if you want that done early next week, I will be gone. you can ask mdeslaur or kees to do that for you20:57
ftajdstrand, not sure where i should stop then. it's easy to see which version is in maverick/universe, then lucid needs the same thing. the tarball is in maverick, the packaging is in bzr with a tag20:58
jdstrandfta: we need full source packages for each release you want uploaded20:59
jdstrandfta: ie, we need to upload it to build it from source20:59
ftasure, but it's a simple merge, i don't see the difficulty here21:00
ftaor i'm missing something21:00
ftait could also be dget from maverick, update d/changelog, new source package, push21:01
ftathere are plenty of options21:01
micahgfta: don't you want to maintain a .lucid branch in bzr?21:01
jdstrandfta: I don't have anything to upload on your behalf, ie to 'sponsor'. how the security team normally works with security updates for community supported packages is that the debdiff is provided for an existing package *or* the community person supplies signed and 'tested on that release' source packages21:02
ftamicahg, i initially wanted to do that, but since i can't upload21:02
jdstrandfta: chromium needs a process for building and testing. aiui, you are asking me to just do it all. I don't have the resources to do that21:03
ftawhy isn't it like the NEW queue??21:03
ftajdstrand, the process is already there21:03
jdstrandthe process for the devel release is, but not -security21:03
jdstrandwhich is why we are talking about it now21:03
micahgfta: it seems like either we have to push to a native PPA or you or someone else needs to provide signed source builds21:04
ftabut the package is exactly the same, it's flexible enough to be built from +1 down to hardy21:04
jdstrandfta: perhaps it would be easier to say that this is very similar to an SRU where you want to do a microrelease bump21:04
jdstrandfta: sure, I could pull the branch and do all the version changes, etc, etc. but then the build is not tested21:05
jdstrandthe build 'on that release of Ubuntu'21:05
jdstrandfta: would you ask ubuntu-sru to create the source packages for you and upload them on your behalf?21:06
ftait has already lived in the beta PPA for a few days, before i release it to +1.21:06
ftait's not enough?21:06
jdstrandI think we are not communicating effectively21:06
jdstrandI don't know what testing you have done21:06
jdstrandI just heard that there is an update for chromium21:06
jdstrandchromium is in universe21:06
jdstrandit is up to the community to provide updates for universe packages21:07
jdstrandyou are (presumably) that community for chromium21:07
jdstrandif you wanted to do an sru for chromium in lucid, how would you proceed?21:07
jdstrandthe process is not really much different for ubuntu-security21:08
asachd21:09
asacjdstrand: only chriscoulson and you can push to the security ppa21:09
jdstrandasac: yes, and I said that if I were provided tested source packages I would upload them on his behalf21:09
asacfta: cant you do that?21:10
asac;)21:10
asacpoint jdstrand to tested sources to push?21:10
asacafter adjusting target to lucid-security i guess21:10
asacso i give up making a real script in upstart .conf21:10
asacthats really painful21:10
micahgasac: I think keybuk did that on purpose :)21:11
ftawhat should i happened after to the version?21:15
fta-to21:15
micahgubuntu0.10.04.1?21:16
ftanothing less ugly? ;)21:18
ftai see other packages don't add anything21:18
micahgfta: other packages generally don't have the same version released to multiple Ubuntu releases AFAIK21:18
ftaGet:5 http://archive.ubuntu.com/ubuntu/ lucid-updates/main apt-transport-https 0.7.25.3ubuntu9 [80.7kB]21:19
ftaGet:6 http://archive.ubuntu.com/ubuntu/ lucid-updates/main libpangomm-1.4-1 2.26.2-0ubuntu1 [62.2kB]21:19
ftaGet:7 http://archive.ubuntu.com/ubuntu/ lucid-updates/main software-center 2.0.4 [278kB]21:19
ftathose are in all versions21:19
asacfta: new upstream versions in security is rare21:19
asacso what micahg said is right21:19
asacubuntu0.9.10.1 ubuntu0.10.04.1 etc.21:19
jdstrandfta: https://wiki.ubuntu.com/SecurityTeam/UpdatePreparation#Update%20the%20packaging has tips on versioning21:20
fta-chromium-browser (5.0.375.38~r46659-0ubuntu1) maverick; urgency=low21:20
fta+chromium-browser (5.0.375.38~r46659-0ubuntu0.10.04.1) lucid-security; urgency=low21:20
fta?21:20
jdstrandfta: which is basically what micahg and asac said when you have the same base version21:21
jdstrandfta: looks good21:21
asacfta: yeah thats correct21:21
asacthen you have to do a trick to push that to lucid in some ppa .. you have to directly upload to the lucid/ folder21:21
jdstrandfta: the thing to remember is a user needs to be able to upgrade to the next release of Ubuntu21:21
ftathat i already know :) i'm not new to packaging21:22
jdstrandfta: heh21:24
jdstrandfta: I didn't mean to imply that-- versioning can be tricky with new sources is all21:24
ftajdstrand, ok so lp:~chromium-team/chromium-browser/chromium-browser.lucid has it. what do you need now?21:26
ftaa source package somewhere?21:26
jdstrandfta: signed and tested, yes21:26
ftajdstrand, can't i just have a native ppa to push everything i need there and let you copy afterwards? because here, it's just the browser, but some other time, it will be also the codecs and/or gyp and/or libvpx21:30
jdstrandfta: current policy is that only canonical employees can have a native ppa (I didn't make that policy)21:31
ftathen a canonical employee should do all that21:32
jdstrandfta: the ubuntu-security team is happy to upload packages into ubuntu-security-proposed and then pocket copy to -proposed on your behalf21:32
ftai can just sit on my hands and watch the dailies build21:32
jdstrandfta: a canonical employee should do what?21:32
micahgfta: if you don't want to, I can make the source package and test locally over the weekend21:33
micahgfta: s/if you don't want to/if you want me to/21:34
ftamicahg, it's not the problem, it only one brz bd --merge away for me, it's just that there are so many barriers preventing good wills like myself to properly maintain released stuff21:34
jdstrandfta: I'm not sure what the problem is. I asked for a signed source package that is tested, that is all21:34
micahgfta: well, most packages don't need updates like this, generally the ones that allow micro-releases are maintained in main by canonical21:35
ftajdstrand, you'll have it once it's uploaded somewhere21:35
jdstrandfta: that is no more or less than I ask of any other sponsored security upload21:35
BUGabundoo/21:35
micahghi BUGabundo, welcome to the Chromium browser saga :)21:35
BUGabundocome again?21:36
micahgBUGabundo: check the logs in 25 minutes :)21:36
jdstrandfta: I'm sorry if there is some frustration here. I appreciate that you are doing the work. keeping a browser up to date with new point releases is hard, and is why there is a whole team dedicated to it for mozilla21:36
BUGabundodon't tell me your vacations made you swicht sides too?21:36
jdstrandfta: I hope that such a team will exist for chromium, especially if it is going to main...21:37
micahgjdstrand: team ATM = myself + chrisccoulson :)21:37
BUGabundojdstrand: I already said several times: rename ubuntu-mozilla-team, Ubuntu Browser Team21:37
jdstrandmicahg: currently, but your team has a clearly defined process for updates in stable releases21:37
BUGabundomicahg: forgetting Jon ?21:37
micahgBUGabundo: asac and I talked about that actually21:37
micahgBUGabundo: he's not doing packaging ATM21:37
BUGabundo:(21:38
* micahg doesn't know how fta does it21:38
BUGabundohe is a super machine21:39
BUGabundowith a lot of scripts and bots21:39
BUGabundo*if* one day he looses his hard drive21:39
BUGabundohe will go baserk21:39
ftajdstrand, http://people.ubuntu.com/~fta/lucid/21:39
jdstrandfta: my plan is to upload to ubuntu-security-proposed and then pocket copy to lucid-proposed for wider testing. is this acceptable?21:40
ftajdstrand, sure, whatever you want, it's better than no upload at all21:41
ftaupgrade21:41
ftajdstrand, and thanks21:41
jdstrandsure, np21:42
ftajdstrand, and you can expect another update in a few days, i already have 5.0.375.53~r47719 in progress for maverick21:42
jdstrandfta: ok. I won't be around the beginning of next week, so get mdeslaur or kees to do it21:43
ftaok21:43
ftamicahg, the golden rule is automation. if i have to some something once, it's fun, twice, it's boring, three times, it's unacceptable, i write a script21:45
BUGabundome too21:45
fta-some+do21:47
ftahttp://ubuntu-virginia.ubuntuforums.org/showthread.php?p=9338769 :(21:49
ftai should stop this google alert, it's making me sad21:49
BUGabundowow21:50
BUGabundowhat DID YOU DO?21:50
BUGabundochromium is SLOOWWWW loadong21:50
BUGabundo*loading21:51
ftanothing21:51
BUGabundoit used to be *instantenous21:51
ftadon't blame me21:51
BUGabundoI have to 'blame' someone21:52
micahgfta: will I have profile isssues if I switch between beta and daily?21:52
BUGabundoand you are in ping reach21:52
ftamicahg, beta->daily, no, daily->beta, more than probable21:52
micahg:(21:52
ftadepends on how big the gap is21:53
micahgfta: I guess I"ll use a temp profile dir21:53
ftai remember in had to trash my session once21:53
fta-in+i21:53
fta(i hate this new keyboard)21:53
BUGabundoahaaha21:54
ftacurve my *s21:54
BUGabundomicahg: alias chromiumnew='chromiumdatadir="$(mktemp -d)";chromiumdiskcache="$(mktemp -d)";chromium-browser --user-data-dir=$chromiumdatadir --disk-cache-dir=$chromiumdiskcache21:54
BUGabundohere's something I learned from fta21:54
BUGabundothen I automated it :D21:54
ftaapparently, --user-data-dir is enough21:55
BUGabundoDOH21:55
BUGabundoyou fail21:55
micahgBUGabundo: chromiumnew started pacman :)21:57
BUGabundoI like that to test our servers and sites.21:58
BUGabundoalways brand new profiles21:58
BUGabundono cache,  passwords, logins, cookies21:58
BUGabundoits AWESOME21:58
* micahg should do that for firefox too :)21:59
BUGabundoalso kewl to login to a site more then once with 3 or more accounts21:59
BUGabundocause private navigation only allows one window21:59
BUGabundomicahg: it will... once it does PID separation21:59
ftaonly one?22:02
ftai can start several22:02
BUGabundofta: private??22:02
ftayes22:02
BUGabundobut they all share the same PID22:03
BUGabundoand logins/cookies are tied22:03
ftahttp://www.osnews.com/story/23334/The_In-depth_Technical_Analysis_of_VP8_Counterpoints22:06
ftahttp://www.osnews.com/story/23335/Patent_Troll_Larry_Horn_of_MPEG-LA_Assembling_VP8_Patent_Pool22:07
BUGabundoheh22:08
BUGabundohow is that going?!22:08
BUGabundowill webm really be patent free?22:08
jdstrandfta: fyi-- uploaded to ubuntu-security-proposed: https://launchpad.net/~ubuntu-security-proposed/+archive/ppa/+packages22:29
fta\o/22:30
jdstrand:)22:31
jdstrandfta: I'll keep an eye on it and pocket copy it to -proposed. I'll also file a bug so it can be tracked (as per SecurityTeam/SponsorsQueue)22:32

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