[01:18] lamont: Hi. I revived that branch: https://code.edge.launchpad.net/~wgrant/launchpad/multi-arch-builders. [01:18] This also conveniently lets us later change the master to dispatch multiple archs to a single builder, of course. [01:55] wgrant: neat-o stuff :) [03:54] hi guys. please, how to use space in the bazaar? [03:55] Ursinha, ôlas [04:53] spm: I feel like I haven't bugged you in at _least_ 4 hours [05:01] ~pythoneers are officially off my christmas card list. [05:01] srsly. [05:03] It seems like those builds should be demoted. [05:03] But they should be almost done now. [05:03] Are you kidding? There's still >2000 left. [05:04] Ah, it's about half done. [05:04] Sorry, 1095. [05:04] Yeah. [05:04] ...but there's more than 2000 in the i386 queue. [05:04] I guess the lack of builders last night made it take a while. [05:05] I've been waiting more more than 24 hours now, and I doubt I'm the one worst off. [05:05] Oh, certainly. [05:05] This situation is completely ridiculous. [05:05] But nobody who matters cares :) [05:07] You know what the worst part is? [05:07] Those times where you look at the /builders page and like 80% of the i386 ppa builders are idle. [05:08] Like /right/ now. [05:08] That's because buildd-manager takes around 15 minutes to go through all of the builders at times like this. [05:08] So it will iterate through all the builders... processing completed builds. [05:08] Then it will iterate through them all again... dispatching builds to them. [05:09] And because it's so slow, most of the builds have finished by the next cycle. [05:09] So it takes ages to complete that cycle too. [05:09] etc, etc. [05:09] How long it takes to go through builders depends on how many builds are in the queue? [05:09] ...or what's making it take so long right now? [05:11] Mostly how many builds have finished since the last cycle. I also suspect slowness in the query to determine which build is to be dispatched next, but I have no evidence of this theory. [05:12] ok. [05:14] I hope one day I'll work up sufficient motivation to add the capability to Launchpad to provide one's own builder. [07:05] what's the matter? lp isn't generating diffs for branch merges. I have been waiting for 5 minutes now. [07:05] kiko: you there? [07:08] bilalakhtar: which branch? [07:10] https://code.edge.launchpad.net/~bilalakhtar/gwibber/fix-585892/+merge/29952 [07:10] spm: ^^ and https://code.edge.launchpad.net/~bilalakhtar/ubuntu/maverick/selinux-basics/merge-603595/+merge/29547 [07:10] ta, lookin' [07:12] spm: the first one among tese links is just a one-line change and the gwibber branch is only 7mb [07:15] spm: seen? [07:21] hyperair: welcome [07:21] why, thank you =) [07:21] guys, for several days, none of my dailies got a chance to build. i thought the scheduler was fixed after the full rebuild nightmare.. === Ursinha is now known as Ursinha-afk [07:22] seems that pythoneers/py27stack2 rebuild is taking all the slots [07:22] fta2: python 2.7 test build [07:22] yep [07:23] fta2: also, the builders were yanked for part of the day [07:23] i'm sick of this. i'll shut the dailies for good [07:23] they're clearly unwelcome [07:24] fta2: occasionally things block them, I think overall they build [07:24] fta2: I believe the Launchpad team is having a sprint which means a little less love for the builders [07:26] when do the builders recognize if something's should be superceded [07:27] bilalakhtar: the job that generates those had gone gaga, being a technical term; I gave it some lovin' via kill, and things should start happening again, [07:27] micahg: I believe at the point just before they would otherwise start to build it [07:27] maxb: ah, so it'll show queued, but not actually build? [07:28] yes [07:28] k, I hope so :) [07:29] ah, I see one where it happened, good :) [07:29] micahg, i killed umd, feel free to host the bot on your side [07:30] spm: thanks [07:30] fta2: I'm not ready to do that yet, can you please keep if going for another couple weeks? [07:30] micahg, it's useless, no slots to build anything. we're losing our time [07:31] fta2: the LOSAs offered to score up the builds before, I think we should call in that favor maybe :) [07:31] only if cake is involved [07:31] spm: next UDS? [07:31] i stop grinning. it's not worth it [07:32] bye [07:32] spm: any chance we can get the chromium and mozilla builds scored up so they build? [07:33] micahg: sure, can you give me a link - just makes it easier tracking stuff down [07:33] oh yeah, this reminds me, the recipe build I requested 2 days ago and was to build 1 hour after the request, have not been build even now [07:33] spm: https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa/ and the PPAs for this team: https://edge.launchpad.net/~chromium-daily, thanks [07:35] ew. 44 in the first one? any particular sets that you can live without for the timebeing? eg lpia? [07:35] spm: sure, don't need lpia :) [07:36] spm: and if I had to choose, I"d say hardy, lucid, and maverick should come first [07:36] oki [07:38] spm: oh, and you can skip firefox 3.6.8 and xulrunner 1.9.2.8 on lucid and maverick [07:42] micahg: oki, so latest chromiums are rescored [07:43] spm: k, hopefully that'll make fta happy [07:43] spm: thanks [07:44] np, the ui isn't conducive to mass re-prioritising. no criticism/complaint, just saying. [07:44] spm: if you can do anything for the mozilla daily PPA, that would be great too [07:44] spm: should I file a bug for that? [07:44] micahg: heh, getting the firefox ones done next. [07:45] bug, nah. the bug is the slow build scheduling which aiui, is being looked at with some urgency [07:45] spm: k, thanks, I think I'm going to drop off to sleep then, have a good day [07:45] you too, have a good one! [07:48] spm: oh, it seems like only the chromium daily has any builds BTW [07:49] micahg: ? not sure I follow? I'm seeing heaps for mozilla here? about 96. [07:50] spm: no, mozilla does have 96, but you can skip firefox368 and xulrunner1928 for lucid/maverick, lpia builds, [07:50] oh yes, I was. :-) === Ursinha-afk is now known as Ursinha [07:51] spm: but I remember that I asked you for all the chromium PPAs and only the daily one has builds [07:52] Oh! right. sorry - my bad. [07:52] fwiw, I'm only rescoring the most recents. Unf you have a few older ones :-( [07:53] spm: oh, also skip Firefox 4.0 [07:53] heh, I think I've already done those. just did the xulrunner ones [07:53] spm: it won't install anyways [07:54] micahg: thunderbird? [07:54] spm: the thunderbird builds should be ok to rescore [07:54] oki [07:56] micahg: I think that's the lot now... [07:57] spm: awesome, thanks, can I ping you tomorrow night if the python stuff isn't done yet? [07:57] depends if cake arrives tomorrow or not. let me just dig up the address of a local cake delivery agent.... :-P [07:58] micahg: certainly :-) [07:58] spm: thanks [09:00] hi all.. Need a small info on launchpad .. iam a registered user on launchpad. i have completed the "code of conduct" on the page. hOwever, the status still shows that i didnot do it. [09:00] Any help/info on that, from anyone please [09:02] stanley_robertso, Do you have a link to your Launchpad profile page? [09:03] hi mrevell .. here it is : https://launchpad.net/~roopesh-majeti [09:04] * mrevell looks [09:04] stanley_robertso, Can you try signing it again? [09:04] sure [09:05] mrevell, but how .. should i do it with differe key ? [09:05] stanley_robertso, What exactly happened when you tried last time? [09:06] it went fine ... when i click on "code of conduct" link now.. it is saying : "It appears you have already done this. The key XXXXX is registered on your account. You can skip to the next step if you are not intending on signing with a different key. " [09:07] stanley_robertso, Have you tried skipping to the next step? [09:08] mrevell, i did .. actually iam registering myself for the bug squad team .. and the approval is getting declined ... as "code of conduct" is not shown as completed/signed [09:09] And what happened when you skipped to the next step? [09:10] nothing.. there is no next step mentioned [09:10] i just uploaded my fingerprint and thatz it.. didnot do anything else .. ofcourse.. followed the instruction mentioned in the fingerprint email we get [09:13] stanley_robertso: There should be another step on the Code of Conduct page. [09:13] Under the message telling you that the key is already registered. === yofel_ is now known as yofel [09:14] wgrant, mrevell .. i got it [09:14] i forgot the final step.. bad on me.. did it just now ... [09:14] thanks to all [10:38] hi guys [10:38] deleted ppas still appear on my personal page [10:39] any way to make it disappear? [10:39] or re.enable the PPA again? [10:45] falktx: I don't think you can reenable them yet. [10:45] But they only show on your page to you. Others can't see them. [10:45] oh, ok [10:45] didn't knew that [10:46] i guess it's ok then [10:46] thanks for the info [10:47] has someone an idea what happened here? http://launchpadlibrarian.net/51966250/upload_1871365_log.txt [10:49] Looking at the build log it seems to come from "Installed-Size: 4460. [10:49] But I don't have an idea why [10:49] geser: The package is bad; Installed-Size has a tab at the end. [10:49] but why? the other arch built fine [10:49] Where's the build log? [10:49] pkgbinarymangler started mangling Installed-Size a few hours ago. [10:50] I wonder if that's related. [10:51] http://launchpadlibrarian.net/51966247/buildlog_ubuntu-maverick-sparc.gnome-power-manager_2.30.1-1ubuntu2_BUILDING.txt.gz [10:51] https://edge.launchpad.net/ubuntu/+source/gnome-power-manager/2.30.1-1ubuntu2/+build/1871365 [10:51] might be, gnome-session on i386 is also affected [10:52] Yeah, so, the package is bad. [10:52] * wgrant checks pkgbinarymangler. [10:52] Woooooah. [10:53] http://bazaar.launchpad.net/~ubuntu-core-dev/pkgbinarymangler/ubuntu/revision/85 [11:37] Hi there when I upload my PPA I always get a warning that my key is not trusted. Is there anything I can do about that? [11:37] I tried signing, but that doesn't work [11:39] can you us the error message you got? [11:41] hi wgrant [11:41] Hi poolie. [11:47] gpg: WARNING: This key is not certified with a trusted signature! [11:47] gpg: There is no indication that the signature belongs to the owner. [11:54] wgrant: nice! [12:13] I thought the servers that compile ppa's were back? [12:14] sluimers: They were... but they're gone again. [12:14] hi! whats wrong with the builders? I am waitung since 27h hours for a 6h job - now it tells me to wait 23h! why this? this is a inpossible situation... [12:14] I still get to see that I have to wait till the next day, just like yesterday [12:15] I got the same thing dnjl :(. [12:15] yo, i see [12:16] It would be nice if the remaining 1011 Python 2.7 rebuilds were to be deprioritised... [12:16] lol, I see.. [12:16] Where's the python ppa? [12:16] https://edge.launchpad.net/~pythoneers/+archive/py27stack2/+builds?build_state=pending [12:18] After that, is the worst over? [12:19] Maybe. [12:19] That's around half the remaining jobs. [12:20] Er, a third, sorry. [12:22] X_X [12:22] 3000 builds? [12:22] Idea: It would be nice if ppa users would able to provide ressources for builders: the lp would send a secured buildd-vm to the provided machines which do the job(s)... - e.g. with libvirt migration [12:23] dnjl: It's impossible to allow users to provide builders in a secure fashion, unfortunately. [12:23] sluimers: 3200, but yes, X_X indeed. [12:25] Guess I'm not helping with 64 builds ehehehehe ^_^;; [12:25] why? if the lp dispatcher would build a ro squashfs-image secured with zertificates to send to the "known registered ppa user", this should a way to think about [12:26] dnjl: The user can still tamper with the VM... [12:26] wgrant, Where do you see the list of builds? [12:26] sluimers: I can't really see a list, but https://launchpad.net/builders provides a summary. [12:27] wgrant: yes, sure - but its easy to say fastly "no" - does someone really thought about it? [12:29] dnjl: builds are insecure by design. === Ursinha is now known as Ursinha-afk [12:33] jpds: yes i know, but what if we are sending to the users host a complete precreated "ro/live" buildd image which is secured with lp certs and this will connect through e.g. openvpn to lp to get the rest. [12:34] dnjl: The user still has access to the hardware. [12:34] And if you have access to the hardware, you have access to the VM. [12:34] I thing there are several ways to think about to get the required savety [12:36] If you can identify a virtualisation technique which does not involve trusting the hardware, I'm sure everyone involved in cloud computing will be most interested. [12:39] but is it possible to manipulate the process if the buildd system is certified and crypted so that you will get i only run if its not manipulated - the vm itself only needs ram and network and network is secured via vpn to lp [12:40] It's possible for the VM to trust LP, sure. But it's not possible for LP to trust the VM. [12:41] yes, this would tricky - i will thing about... [12:41] think [12:44] Hey http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/maverick/sugar-flipsticks-activity/maverick/revision/3/debian/install is down === flacoste_afk is now known as flacoste === Ursinha-afk is now known as Ursinha [14:54] wgrant, hello [14:55] wgrant, any solutions to solve the problem with builds which are waiting for dependencies like forever [15:00] would anyone know why its taking an abnormally long time to build? like, its been saying waiting to build for 7 hours+ [15:03] wrinkliez: a python2.7 test build of all python-packages takes some PPA resources [15:04] oh is that whats going on [15:04] cool [15:07] wrinkliez: and for a reason I don't know, PPA builders come and go as they like [15:08] huh. [15:08] the additional builders are not dedicated to the build farm, they're simply lent to it when someone else is not using them [15:09] FE FI FO FUM [15:09] we're getting 7 new dedicated builders soon [15:10] And a fixed buildd-manager? [15:41] FE FI FO FUM, I smell a hunam [15:47] sluimers: Yourself? [15:50] Just quoting an old game [15:51] I see that there's one ppa armel builder on https://launchpad.net/builders basically doing nothing [15:51] it's not for general use [15:51] ah, okay === beuno is now known as beuno-lunch [16:52] Launchpad is teasing me with an error message saying that my browser don't send a HTTP Referres Header. Why? [16:54] yotta911: https://answers.edge.launchpad.net/launchpad/+faq/1024 [16:56] micahg: I know how it's work and i guess a privacy violation [16:57] micahg: and more: If i woud like attack launchpad, i will forge this header easily. [16:58] But Ok. Thanks for helping. [16:59] yotta911: I just know where the FAQ is, I don't work on the code (yet) [16:59] The Referer header check is not intended to protect against direct forged requests, it protects against cross-site request forgery [17:00] What's the problem with cross-site request? [17:02] micahg: I understood and i will open a bug. Thanks again.] [17:24] yotta911: For example, some website presents you with a form that asks you to vote on something. But secretly (in the html source), it actually directs your browser to ask Launchpad to delete your PPA. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === beuno-lunch is now known as beuno === dendrobates is now known as dendro-afk [19:33] mneptok: o/ === dendro-afk is now known as dendrobates [19:38] pmatulis: ahoy! [19:38] mneptok: all good on your end? [19:39] pmatulis: AFAIK, yes. do you know something i don't know? ;) [19:39] * mneptok is busy prepping for OSCon [19:40] pmatulis: anyone from the office going to DebConf in NYC or LinuxCon in Boston? [19:40] mneptok: etienne mentioned it a while back [19:41] mneptok: debconf [19:41] pmatulis: i should call him with travel tips i have accrued. [19:41] mneptok: i'm not sure if he's going though [19:43] let's find out .... :) [19:47] is loggerhead down? http://bazaar.launchpad.net/~sproaty/whyteboard/development/changes [19:49] aand back :P === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [20:47] bueno had a session on doing merge proposals on lp but is now sick while sprinting, the session is in 15 minutes, can anyone possibly substitute teaching a session for an hour for Ubuntu Developer Week? [20:52] rockstar: help? [20:53] jcastro, forewarning: I am frakkin' tired right now... [20:53] ok, I'll cancel it [20:53] jcastro, I can probably field questions or something. [20:53] jcastro: We could just turn it into a Q&A session if we don't find anyone [20:53] ok [20:53] rockstar: want to do that? [20:54] jcastro, sure, but can I cut out before the whole hour is up? [20:54] yeah [20:54] jcastro, for context, I woke up at 330 this morning, couldn't get back to sleep, and it's getting late here. [20:54] wait, are you sprinting? [20:55] jcastro, I am indeed in Prague. [20:55] (I'll see you next week) [20:55] I'd rather cut the day early, disregard. [20:55] I wasn't aware you were sprinting! [20:55] jcastro, as long as it doesn't cause problems. I'm happy to help. [20:56] it's all good, go to bed! === chrisccoulson__ is now known as chrisccoulson [21:42] guys, launchpad is timing out for me [21:42] I'm on this url: https://launchpad.net/ubuntu/+search?text=landscape-client [21:42] is it a general issue or just a beef with that particular search? [22:10] hi. Can anybody here help with password reset on launchpad? [22:21] Can anybody here help with password reset on launchpad? [22:29] I've a question, does someone here now how to target a bug for a milestone ? (The milestone doe support bug target.) [22:31] iwontbecreative: you mean with the API? [22:34] lifeless: http://pypi.python.org/pypi/testr_recipe/0.1 === dendrobates is now known as dendro-afk [22:38] No i meant in launchpad itself ? [22:38] I know it's possible a stupid question but couldn't find [22:41] iwontbecreative: you need a certain level of access to the project to be able to do that apparently [22:43] Oh... [22:44] Seems bad for some projects like onehundredpaper cuts :/ [22:44] Thank you for the answer :) [23:33] hey spm === fta_ is now known as fta