[11:40] greetings all, i need access granted for http://pad.ubuntu.com/eiEDBfo63Q please [11:41] or tell me what i am doing wrong [11:41] i get Authorization is required to access http://pad.ubuntu.com/eiEDBfo63Q [11:44] Kilos: You need to be an Ubuntu member, or be in the ubuntu-etherpad team. [11:44] https://launchpad.net/~ubuntu-etherpad [11:45] Kilos: Although since you're an Ubuntu member according to your cloak, you ought to have access already. [11:46] You do need to go through single-sign-on. [11:46] yeah i dont know why im struggling, but i joined that team now and will try again, thanks alot cjwatson [12:02] thank you its working now [12:04] OK, cool. Very odd that it didn't work already, since you should already have been an indirect member. [12:05] lol some glitch somewhere [16:18] thanks for the help guys, enjoy your day [18:43] Hi all! :) Can someone help me a bit with launchpad PPA? I have a error while building a single binary package :/ [18:48] fix the error :) [18:49] building locally with sbuild should let you debug what is failing [18:50] Locally works like a charm... [18:50] it builds perfectly [18:51] here is the log [18:51] https://launchpadlibrarian.net/206213032/buildlog_ubuntu-trusty-amd64.gpuplotter-x64_4.0.3.2-0ubuntu2_BUILDING.txt.gz [18:54] not locally, build it in a clean chroot that doesn't already have everything installed [18:55] mmmmm kk i understand now, just installed sbuild, any guide to use it for debugging? [18:55] https://wiki.ubuntu.com/SimpleSbuild [18:55] is this in a commercial ppa? [18:56] Commercial ppa? i think no, its my personal ppa [18:57] you need to upload a source package which includes source and compiles the binaries from that, then [18:57] what is "gpuplotter" exactly? [18:59] the problem is the original source code makefile isnt multiarch (need to change a variable in makefile to set it 64 or 32), and dont know very well about the makefile and rules file to make it multi-arch [18:59] is it this? https://github.com/bhamon/gpuPlotGenerator [18:59] yes! [18:59] ah, well then it should be patched [18:59] and the correct package naming and such used [19:01] hmm, manual makefiles are annoying, but doesn't look to hard to fix it [19:01] sure?? [19:02] i ll try to do it [19:02] want to learn how to and headhaches :P [19:02] thanks for all the info! [19:07] yeah, i'm sure. :) [19:32] now just set up sbuild in ram and do the build, its fails like the previous log [19:33] how can i search the dependecy? [19:33] and where to place the dependency package? in debian/control? [19:34] yes, dependencies are specified in debian/control [19:35] but like i said, you need to build from source, not just upload pre-build binaries [19:35] pre-built [19:35] the same for burst-wallet [19:37] mmmh, i get it, thanks dobey! [19:39] Homepage: [19:39] wow, i'm surprised launchpad even accepted that upload [19:40] description is also the :( [19:42] where did you get burst-wallet from exactly? [19:43] yes im surprised too hahaha, well it 's for a study project, a live cd hat includes a PPA with all packages related about a new cryptocurrence, like bitcoin, dont have time to convert all the sources to .deb packages, but learn a lot how to do for future [19:44] How NOT upload/make pakcages [19:45] its from https://github.com/BurstProject/burstcoin [19:49] ah, java. so annoying with including the .jars in source repositories [20:11] Errr. Sounds like this channel is not about novation launchpad :-) [20:46] cjwatson: remember we talked about signing messages for the email interface? i just sent one that got bounced back with a bad signature, then copied and pasted the original text and verified it with gpg and all is well. seems like there is indeed something wrong with launchpad. === mwhudson_ is now known as mwhudson [21:39] wxl: I'd need to see a sample of the text that you *have not copied and pasted*. Copying and pasting can cause whitespace differences, which will matter. [21:42] cjwatson: that's the weird thing is that copying and pasting actually seems to work :) [21:42] wxl: Do you have the message-id? [21:43] cjwatson: of the sent message? [21:43] er - yes? what else? [21:43] here's the message id of the message i sent: Message-ID: [21:44] wxl: what address did you send that to? [21:44] cjwatson: <1246906@bugs.launchpad.net> [21:45] back in a bit, kids [21:49] wxl: hm, I see it in the logs but there isn't much to go on there [21:49] well that's unfortunate cjwatson. should i just file a bug against launchpad itself then? [21:49] wxl: do you still have the original message? can you forward it with full headers to cjwatson@canonical.com ? [21:50] cjwatson: sure, on it [21:50] it's very important that whatever you do to forward it preserves it byte-for-byte as it was sent [21:50] I don't want one of those "forwards" that just tacks some body text on the front [21:51] ohhhhh [21:51] hm [21:51] because I need to be able to see it just as Launchpad did [21:51] crap [21:51] i'll have to think about how to bounce this easily [21:51] might havbe to install mutt and all [21:52] yeah, I believe you can save it as a file [21:52] going to have a think abut it whilst grabbing coffee [21:52] oh [21:52] at which point you could send it as an attachment in mutt [21:52] ok well brb [21:54] wgrant: is there any reason we don't run process-mail with DEBUG logging, other than usual awfulness? [21:54] this might be less painful to debug if we did ... [21:56] actually, you know what, that's just obviously wrong, will prepare a branch to crank the debug level [22:35] cjwatson: i think you should have got a redirected mail from me [22:44] wxl: only your first one; but I'm waiting for increased logging to be applied on our servers [22:44] cjwatson: the forward? not the redirect? i sent that latter one just a few minutes ago [22:44] cjwatson: well 9 minutes ago :) [22:46] I haven't seen it yet, though my mail goes through a few twisty paths [22:46] okie dokie [23:00] wxl: can you try sending that (or similar) to Launchpad again? [23:02] sent cjwatson [23:03] huh, it didn't even try to check the signature [23:04] that would explain the problem [23:04] i have three different programs telling me the signature is good [23:07] wxl: I really do need the full message though, and your attempted redirect hasn't come through. Is there any way you can save it to a file and send it as an attachment or even put it on a pastebin or something? [23:07] cjwatson: annoying the redirect didn't go through [23:08] (because I'm fairly sure that I need to look at the fine details of LP's signed message parsing) [23:09] cjwatson: http://paste.ubuntu.com/11086415/ [23:09] i am using sha2. hopefully it's not choking on THAT. :) [23:11] thanks, will look in a bit [23:11] ping me if you need anything else from me cjwatson and thanks again [23:12] thanks for the information, it's much more tractable with an example message [23:17] wxl: so, I get a bad signature here; but it could be whitespace differences of some kind [23:17] wxl: (I had to convert it to Unix line endings before mutt/gpg would accept it as a signed message at all) [23:18] cjwatson: did you just download the paste? [23:18] yes [23:19] and then stuffed into a new Maildir so that I could look at it in mutt [23:20] ah interesting [23:20] there is some sort fo error with it [23:20] * wxl sighs [23:21] i think it's choking on the html content [23:21] gpg --verify gives me an invalid armor header that ends with /a>\r\n [23:22] no, I'm getting this even from just the text part [23:22] and I know that LP walks part by part [23:23] there's also a CRC error of some kind [23:23] same error if I leave DOS newlines in place [23:23] i wonder if it's not pastebin that causes the problem [23:23] 1s [23:24] must not be [23:24] that was why my first option was to send as an attachment [23:24] same md5 [23:24] did you copy and paste into pastebin, or use a program? [23:24] ok [23:24] so it could be the way thunderbird saves the message but that seems unlikely [23:24] still, it's not a straight signature verification failure on LP's side [23:25] right [23:25] well, i'll keep exploring options i guess [23:25] maybe i should try thunderbird/enigmail [23:25] a signature failure would (now) leave a debug message [23:26] so where do you think the problem lies? [23:27] my best theory so far is that it's in the code which walks through MIME messages looking for signatures - that is, it simply doesn't think the message is signed at all, good or bad [23:31] but when I run that by hand it seems fine ... [23:31] well let's see what enigmail does [23:31] just sent one [23:31] i have a strong suspicion it won't fail [23:39] still no reply [23:41] wxl: worked, though, https://bugs.launchpad.net/ubuntu/+source/lxterminal/+bug/1246906 [23:41] Ubuntu bug 1246906 in lxterminal (Ubuntu) "changing terminal color in lxterminal causes display of htop to partially disapper" [Undecided,Incomplete] [23:41] yeah bizarre [23:42] i wish there was a way to test messages directly with lp [23:43] so i could test things without having to have something to do it to :) [23:44] i note there's no mime [23:44] You could probably use staging for this [23:44] oh that's a thought [23:45] Mail to same address but @bugs.staging.launchpad.net [23:45] Assuming the bug exists there, otherwise pick another one [23:45] well i could probably make a new bug too [23:45] well not really make one, but you get the idea :) [23:45] to new@ [23:46] wxl: Actually, could you try again with your previous setup? Because I've just looked more closely and it would appear that the crontab change I requested was live briefly, then reverted, and is now live again [23:46] wxl: Which means that a good part of the debugging I did earlier was on the wrong track :-( [23:46] ahhh [23:46] (Because debugging was in fact switched off) [23:46] so you want me to do the original message? [23:46] Yeah [23:46] k 1sd [23:47] But maybe to something on staging now to avoid confusion [23:47] should i send it to staging [23:47] yeah ok [23:47] bug 1 maybe XD [23:47] bug 1 in Ubuntu Malaysia LoCo Team "Microsoft has a majority market share" [Critical,In progress] https://launchpad.net/bugs/1 [23:47] Oh [23:47] Hang on [23:47] I bet staging doesn't have my debugging change [23:47] So it would actually be more helpful in this case to pick a different bug on production [23:47] ok lemme dig something up [23:48] If I really have to, I'll set something up on dogfood, but that won't be tonight [23:49] i can't imagine others are having this problem otherwise you'd hear more griping [23:49] then again maybe people don't really use the email interface [23:51] It gets some use, but I suspect most people who use it aren't using gmail [23:51] sent [23:51] Haven't actually checked, difficult to mine it out of logs due to spam [23:51] yeah well such is the case with email :( [23:52] ok, that cron job is */3 so wait a couple of minutes [23:55] ah, there we go [23:55] 2015-05-11 23:54:14 DEBUG attempt gpg authentication for [23:55] 2015-05-11 23:54:14 DEBUG Signature couldn't be verified: (7, 8, u'Bad signature') [23:58] cjwatson: and indeed that's the message i get :)