=== jamesh__ is now known as jamesh [00:31] apw, ogasawara: would be nice to announce such changes ... https://launchpadlibrarian.net/201982759/buildlog_ubuntu-vivid-amd64.armhf-cross-toolchain-base_1.107_BUILDING.txt.gz [00:42] doko_: They didn't change anything, I slimmed down the chroots a bit and there's a missing build-dep, apparently. [00:44] doko_: Looks like the kernel correctly build-deps on cpio, FWIW, it's just missing for *-toolchain-base, because our chroots used to have initramfs-tools in them (because of upstart), and do no longer. [00:45] infinity, ahh, ok. can you fix these? (and maybe apply multilib-stage1.diff before) [00:46] doko_: If it turns out to be a more widespread problem in the rebuild, I'll re-add cpio to vivid's chroots (and pull it out again in vivid+1)... I already readded the non-Essential procps because too many things seem to expect it. [00:47] doko_: And yeah, I can play with *-toolchain-base. I should have done so right after I got glibc 2.21 in anyway. [01:04] darkxst are you not going to release the fix for bug 1315442 for vivid? [01:04] bug 1315442 in Ubuntu GNOME "Extra "fi" in /etc/init.d/gdm" [Undecided,New] https://launchpad.net/bugs/1315442 [01:06] aeoril: looks like 0ubuntu3 was released 57 minutes ago: https://launchpad.net/ubuntu/+source/gdm [01:07] sarnold oh, foolish me - I just saw that! Thanks! [01:07] sarnold and now the paperwork ... :) [01:10] sarnold I saw this - "I have uploaded to trusty and utopic, please add the SRU paperwork when you can." and only now realize those two were in there and not vivid because they are backports, I guess [01:11] darkxst ^ [01:11] aeoril: right, though as vivid gets closer to release, at some point I think they'll start requiring the full sru dance for uploads for it, too [01:12] ..maybe it's just an okay from the release team. I don't normally work on the devel branch, much there is new to me :) [01:13] sarnold lots of fun, though - I am digging this! It feels good to be starting to get used to some of the tools - feeling more oriented [01:13] aeoril: woot \o/ [01:14] sarnold \o/ :) [01:14] aeoril, its already released in vivid [01:14] see the bots comment, above mine [01:15] darkxst yes, I saw that - I was confused by that one email, but got it now. Thanks! [01:15] SRU's only start after the final release [01:16] oh, thanks [01:22] darkxst also, now I know when I reply to the generated e-mails, they make comments, so I will watch that as well. I am a little shell-shocked tonight from a long day - going on 18 hours now today. Time for some relaxation - the SRU paperwork looks like tomorrow afternoon, but it will get done. [03:26] darkxst I did the SRU "paperwork", but in doing so am a little worried whether I always called the script with "sudo" in my testing. I think so, but am a little nervous about that - I did not put "sudo" in the test descriptions as I wrote them when I uploaded the debdiff patches and made the comments, so I am a little worried about that [03:26] darkxst I think not putting "sudo" was an oversight, but just not absolutely sure [03:27] (in the comments when I uploaded the debdiff patches) [03:28] darkxst anyway, please let me know if the sru info seems good to you [03:28] aeoril, wouldnt worry about, anyone reviewing the sru, is likely going to know when or not to use sudo [03:30] darkxst ok, thanks. I'll be interested what you think of how I wrote the SRU stuff - I hope I did it right [03:30] its good, way more verbose than needed, but thats fine [03:31] darkxst unfortunately, throughout my career I have fought the tendency to be highly verbose and almost always lost :( [03:32] darkxst anyway, I will keep that in mind next time and try to keep it shorter ... [03:34] darkxst basically, when writing, words flow out of me like a sieve lol [03:37] aeoril, I noticed! [03:38] darkxst lol I have always been (in)famous for that (inequity)quality ;) [03:38] darkxst I'll try better to put a cap on it [03:40] darkxst when I am not a computer guy, I do writing and editing on the side, so it can actually be good that I a lot easily for those pursuits [03:40] write a lot* [03:40] I don't suppose it matters too much, just makes it take longer for your to write it! [03:41] darkxst the problem is it also makes it longer for the reader to slog through it! [03:41] (or, just give up and not read it at all) [03:42] yes, I barely read it, just skimmed over and picked out the relevant bits [03:43] most of my SRU's are around 20lines ;) [03:43] darkxst yes, that is what most people do with my writing lol - I am quite aware of it after almost 30 years of experience with myself [03:43] darkxst I can keep a cap on it if I remember to watch out for it, though - just my natural self if I do not watch out is very verbose [03:43] the human minds seems pretty good at sifting out the important bits from the faff! [03:44] faff? [03:45] "wheat from the chaff" [03:45] all the extra details that arent really required [04:03] aeoril, 3 lines ;) http://pastebin.com/PJ63JbKv [04:04] but the extra details don't hurt [04:06] darkxst thats too funny! :) But, there are some gotchas that need to be considered in the testing that I cover that you don't [04:06] darkxst I can still learn! :) [04:08] I don't actually get why there would even be a difference between running /etc/init.d/gdm and ./gdm from that path (there shouldnt!) [04:09] though if its failing silently, then the stderr is probably redirected to some log in that case [04:09] darkxst I actually did some extensive probing into that and found out some of the reasons, but that was over a month ago. Nevertheless, it is real and happens. It should not, but does. Parsing command line stuff badly, IIRC or something like that [04:10] (pulling commands out of the command line, but not doing it robustly, etc) [04:10] dash can be a weird beast sometimes, but this doesnt like one of them [04:10] "doesn't like one of them"??? [04:11] ^look like] [04:12] poor parsing technique in the actual sysv scripts, I think, but I do not think I ever completely nailed down exactly what was going on [04:13] techniques* [04:13] aeoril, there a lots of corner cases, when it fails strangely on bash syntax [04:14] darkxst it was quite a tree of scripts when I was delving into it! [04:14] ("it" being dash) [04:14] darkxst I meant "it" being the sysv scripts [04:16] darkxst there was some fairly elaborate scripting going on in those scripts beneath gdm [04:19] darkxst: 20 lines? You ovrachiever. Most of my SRU boilerplate looks like "[Justification] See original report [Test Case] Try to run it. Does it work? You friggin' win. [Regression Potential] Might set your house on fire." [04:20] aeoril, I don't know where that script came from, but if it has bash-isms, its likely to have problems [04:20] infinity, see my 3 line version about 20 lines up in the scrollback! [04:21] darkxst: Ahh, indeed. That's more like it. :) [04:22] I probably should have said 20 lines would be the max length ;) [04:23] darkxst: Well, I've had some elaborate ones where the test case is really brutal or the regression potential is actually worth a discussion, but most SRUs are really quite simple. [04:23] (THankfully) [04:24] infinity, agreed, but the point was aeoril's is about 200 lines longs ;) [04:25] darkxst: I accepted his SRUs before he updated the bugs. I'm now afraid to go read. ;) [04:25] Oh wow. [04:25] aeoril: Have you considered a career as a tech writer? [04:26] infinity are you being facetious? My effort seems to have fallen flat ... [04:26] aeoril: A little bit, perhaps. :) [04:27] aeoril: But that level of verbosity could come in handy, say, for help with the release notes in a couple of weeks... [04:27] *cough* [04:29] I have done a lot of tech writing during my career - probably hundreds of pages, if not thousands overall, as part of my duties as a software engineer during all phases of development, including requirements, test cases, bug reports - just about every part of where writing needs to be done. Most of my time has been spent coding and debugging, but tech writing was still a big part of [04:29] things [04:30] infinity hook me up! I think I might be a good fit for that, and like writing as well as editing other peoples writing. I write and edit on the side, sometimes for money anyway. That may be a very good fit for me actually at Ubuntu [04:32] infinity I worked in enterprise, government and military, so tech writing was an oft used skill for any developer in those environments [04:32] aeoril, ubuntu GNOME release notes are generally sparse [04:32] no one has started on the final ones yet [04:32] darkxst: That's okay, he can help write both yours and mine, I don't mind. We'll share him. [04:33] infinity, perfect! [04:33] lol I feel like the kid claimed by two women King Solomon said to cut in half ... [04:34] aeoril, don't take it so seriously then ;) [04:34] aeoril: The sawing in half comes after the release. [04:34] darkxst I was joking ... [04:35] infinity, is half a person going to be eligible for ubuntu membership in the future though, that seems a little problematic! [04:35] infinity if you are serious about me doing some technical writing, where would I get started with that? [04:36] infinity I am actually very interested in that [04:36] (I really enjoy writing) [04:36] aeoril: Well, we have a doc team that probably has all sorts of docs on how to contribute. But none of them seem to contribute to my release notes, so those end up written by frustrated release managers 20 minutes before we ship. :P [04:37] And, while I'm clearly no slouch with this whole English business, writing release notes is about my least favourite thing in the world. [04:37] infinity, we don't even have that, the doc team refuses to recognise flavours [04:37] infinity you are talking specifically not about Ubuntu but more gnome? [04:38] darkxst: I feel your pain, and a beer is in the mail. [04:38] aeoril, infinity is Ubuntu, I am GNOME [04:38] aeoril: I'm talking about all the Ubuntu flavours, though my bias is to the "main" release notes that cover all the core bugs and new features in Ubuntu in general. [04:39] infinity I might give it a shot. Definitely a different direction than where I was going, but I might like it [04:40] aeoril: Anyhow, I'll nudge you about it closer to release week. I'm sure we'd all appreciate some help from someone who actually enjoys writing and editing, rather than what most of the flavour release managers do, which is copy and paste some bug reports, scream "THAT'S TOTALLY GOOD ENOUGH, LA LA LA" and run off to the pub. [04:40] Which is a natural response after the week of pain we go through to get there. [04:41] infinity whould this be steering me into the arena of a "release manager" then? [04:42] aeoril: Nah, that's an entirely different role. [04:42] good [04:42] aeoril, community members can't really do that [04:42] darkxst: That's not true, lots of the release team is community (well, all of them, depending on your world view). [04:43] anything with the word "manager" in it strikes me with fear and loathing [04:44] aeoril: Depends on the flavour, but most release management is really just technical steering. It's a position of trust (and holds many keys to the kingdom), though, so comes well after upload permissions and earning trust from your peers, etc. [04:44] infinity, I meant the actual releases, not the release team [04:44] that requires access to secret canonical servers no ? [04:45] darkxst: Well, yeah, the final button mashing is something we still have to do internally, because of how the infrastructure is laid out, but we've got that down to mostly a no-brainer. [04:45] infinity sounds like an honorable thing then [04:45] darkxst: Really, that bit takes me only a few minutes per flavour, so I don't really think about it anymore. It's all the other crap we (as a team) do leading up to it that's work. :) [04:47] infinity thanks for recognizing some value in what I did and offering me something I may be really interested in. A great manager can match the right talent with the right job [04:47] darkxst: Say, why are you not core-dev? [04:50] infinity, I probably dont have enough time for that! still be meaning to put my -desktop team application in [04:51] darkxst: Well, core-dev doesn't imply uplading every package in main tomorrow, just the ability to do so without being sponsored. :P [04:51] darkxst: Which, I'd guess you've got enough history for at this point. [04:52] darkxst: (And my motivation for asking was because I was pondering if you'd make a good addition to ~ubuntu-release, but because of the permissions that team gives, I won't invite non-core-devs, since it allows you to bypass upload rights) [04:54] infinity, most of my work has been within ubuntu-desktop sure there would be some core-dev packages in there though [04:55] I don't really check which are the difference, since I don't have upload rights for either [04:56] but my MOTU application was knocked back with your too GNOME! [04:56] darkxst: Hah. Well, that's curious. Maybe I should have stood for a DMB position this term. [04:57] darkxst: I mean, it's fair that if all of someone's contributions can be covered by a package set, then that's the safe set of permissions to give them. But people maintaining major DEs like xfce and gnome tend to need to mangle packages that aren't theirs quite often, IME. [04:57] Since we all share a bunch of stuff in core. [04:58] infinity, none of that is in the flavour packagesets though [04:59] ours is dismall, http://people.canonical.com/~ubuntu-archive/packagesets/vivid/ubuntugnome [04:59] darkxst: Exactly, nothing in core can be in flavour packagesets, so once a flavoury person finds they've had to touch those packages often enough and have proven competence in doing so, I'm all for them being a bit less flavourful in their permissions. [05:00] infinity, and nothing in -desktop either, though pretty sure I will get that now, if I can actually make a meeting [05:00] its like 5am or so here [05:03] darkxst: Just learn to live timezone-free, like I do. [05:03] (don't do what I do, it's apparently really bad for you) [05:04] infinity, I need my sleep, had chronic fatigue syndrome, days don't go so well If i don't sleep at night! [05:05] darkxst: I'm pretty sure I need sleep too, I just rarely get any. [05:13] infinity what is your main role in the Ubuntu community? [05:13] infinity, I get all sick and stuff If I do that ;( [05:14] aeoril: Jack of all trades, master of two or three. [05:15] infinity yes, I saw you are a member of a heck of a lot of things on your launchpad page ... do you work for canonical? [05:15] aeoril: I do, yes. [05:15] infinity do they pay you to work there? [05:15] They keep me fed, yes. :) [05:16] Ah so it was a choice between food and sleep. Good choice, you made. [05:16] Heh. [05:16] darkxst I also need my sleep now - not so much when I was younger [05:18] aeoril, are you saying I am young? that is unlikely [05:22] Unit193 remember when you tried to get me interested in vala? [05:23] Sounds like me to mention it at least, yep. [05:23] Unit193 I ran into some of it in a bug I was trying to fix (gnome stuff) [05:24] It reminded me of you and the ubuntu beginners team [05:25] Heh. [05:26] darkxst no, nothing meant about you - just me being older now [05:30] aeoril, quite a few gnome apps are [05:30] written in vala [05:30] none of the core though, thats all pretty much C [05:31] with a touch of JS, for the shell [05:37] darkxst yes, I remember you mentioning I might be a good fit for parts of Gnome because of my Javascript background ... [05:37] I have C and Javascript, so might be a good match (I could learn Vala) [05:38] its not really standard JS, somewhat customised [05:39] its the firefox spidermonkey engine with a bunch of custom hooks [05:39] darkxst I was just about to ask the question to your answer :) [05:39] aeoril, there you go, upstream want me to maintain gjs [05:41] darkxst hmmm ... I had aspirations of Kernel/Driver, then tech writer, now gjs - oh me oh my, so many choices! What to do! :D [05:41] gjs is 100% C [05:42] but it enable the custom JS bindings, most of the code if enababling gobject-instrospection [05:42] s/if/is/ [05:43] aeoril, just do what you enjoy! [05:44] darkxst I am very fortunate - I enjoy just about everything technical! [05:44] (I've been pulled from that a little, given the lack of manpower in Ubuntu GNOME ;( ) [05:54] darkxst infinity I'm off to slumberland. Thanks again for all you help yesterday and today. Good night. [05:54] your* [06:40] Just installed the new Ubuntu, but something should be changed in the installer imo. [06:40] It asks crypt password BEFORE it asks keyboard layout [06:41] So you need to enter a password before your keyboard is set correctly (ok you can change it manually in the menu bar .. but not that user-friendly) [07:43] dupondje, Since not everyone is around at the moment, could you please file a bug asking for that to be changed? === PaulW2U_ is now known as pw2 === pw2 is now known as Paulw2U === JethroTux is now known as JethroTux`afk [13:11] could someone please merge this fix for a critical bug before the 15.04 release? It's been waiting for quite some time: https://code.launchpad.net/~psusi/ubuntu/vivid/grub-installer/fix-efi-multi-disk-installs/+merge/247772 [13:52] hm no psusi.. uploaded grub-installer with that fix === JethroTux is now known as JethroTux`aFk === ximion is now known as ximion-afk [19:09] psusi: i've uploaded the fix [20:17] tjaalton, thanks [20:20] now if someone more familliar with ubiquity could take a look at this release critical blocker of a bug, that would be great: https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1418706 [20:20] Launchpad bug 1418706 in ubiquity (Ubuntu) "Vivid: UEFI: blank drive incorrectly detected as existing BIOS-mode install" [Critical,Triaged] [20:20] I have a strong suspicion that the underlying breakage is more far reaching than the simple use case in which it was originally detected, which alone is very serious