| *** radiofree has quit IRC | 02:05 | |
| *** jjardon has quit IRC | 03:21 | |
| *** jjardon has joined #baserock | 03:24 | |
| *** radiofree has joined #baserock | 03:28 | |
| *** gtristan has quit IRC | 03:28 | |
| *** gtristan has joined #baserock | 03:57 | |
| gtristan | SotK, nod, while I am in favor of just using bugzilla (painless easy install, mature software, all important features for serious development are there), I'm not against using something else just because it's "different" | 06:07 |
|---|---|---|
| gtristan | So far what I've seen from Storyboard (current baserock.org installed version), it's just not up to the task of handling serious development | 06:07 |
| gtristan | Not being able to "watch a product" and receive emails when someone creates a new bug on the product you're interested in, or get email notifications on bugs which you are "on CC" for specifically, are both IMO a blocker for any bug tracker | 06:09 |
| gtristan | s/Not// | 06:09 |
| gtristan | Bug status and resolution reasons are also important to a lesser degree, but important nonetheless | 06:11 |
| gtristan | Inter bug dependencies (this bug blocks that bug) are also very, very nice to have | 06:12 |
| gtristan | Opening the link pointing to the bug, and having the bug show up the first time - is also nice to have | 06:13 |
| gtristan | I.e.; imagine you could just open https://storyboard.baserock.org/#!/story/65... see the bug report right away instead of a "no such story error" appearing before you refresh the page | 06:14 |
| gtristan | Attachments - I think are very important, hey so what if it doesnt do it quite as *well* as bugzilla does, recognizing all the mime types and such, displaying stack traces beautifully, etc... but at least plain text | 06:17 |
| gtristan | at least patches right ? | 06:17 |
| gtristan | Ok so patch status is not very important feature, that can be read through the comments, it's nice to have but meh | 06:18 |
| gtristan | seriously though, page load time and refresh time is *very* important; you are usually in the middle of something - maintainer comments on your patch; you receive email | 06:19 |
| gtristan | then, you're like, OK this fix I'm working on in another project will take 30 seconds to compile; I happen to know what I want to reply to the maintainer who commented on my bug report; that's plenty of time | 06:20 |
| gtristan | if the page takes a long time to load in your bug tracker, that can make the difference of whether you follow up on that bug or not | 06:21 |
| *** VLetrmx has left #baserock | 08:04 | |
| paulsherwood | SotK: seriously, use ybd. | 08:22 |
| *** toscalix has joined #baserock | 08:26 | |
| *** toscalix has quit IRC | 08:27 | |
| *** toscalix has joined #baserock | 08:27 | |
| *** ctbruce has joined #baserock | 08:47 | |
| tiagogomes_ | <SotK> grumbles about morph downloading every artifact in the system, then downloading the system artifact too: that would be a one change in morph and I've decided to not do anything about it as in the future we probably will get rid of system artifacts | 08:51 |
| *** jjardon has quit IRC | 08:51 | |
| tiagogomes_ | s/one change/two lines change | 08:54 |
| *** jjardon has joined #baserock | 08:57 | |
| *** jjardon has quit IRC | 09:02 | |
| *** bashrc has joined #baserock | 09:06 | |
| *** jjardon has joined #baserock | 09:11 | |
| *** jjardon has quit IRC | 09:15 | |
| *** franred has quit IRC | 09:39 | |
| *** jjardon has joined #baserock | 09:41 | |
| pedroalvarez | hm.. I understand one of the points of why provision and testing were in the same step in CIAT | 09:42 |
| SotK | gtristan: current master of StoryBoard is somewhat more usable than the about year old version on storyboard.baserock.org, we've recently merged initial support for email notifications for example | 09:42 |
| SotK | you can also actually see more than 20 events on story timelines, and its a lot less buggy in general | 09:43 |
| * tiagogomes_ wonders whether is going to spam me as gerrit | 09:44 | |
| SotK | hopefully your experience with it will improve somewhat when we finally update | 09:44 |
| SotK | tiagogomes_: the notifications default to off for the moment, we aim to make the level of notification you receive usefully configurable | 09:44 |
| SotK | gtristan: also thanks for the comments on what you find important, they are useful :) | 09:45 |
| *** franred has joined #baserock | 09:46 | |
| pedroalvarez | so, if you provision on a different step, the cleanup will be more tricky | 09:46 |
| SotK | paulsherwood: is that likely to improve my chances of successfully upgrading, or just likely to be less annoying with its downloading habits? | 09:47 |
| paulsherwood | SotK: the latter | 09:47 |
| tiagogomes_ | I also didn't change that behavior as I found to be odd that if you build a system, the only thing that would happen was the system artifact to be downloaded | 09:49 |
| pedroalvarez | which sometimes may be faster | 09:49 |
| pedroalvarez | I like that when deploying it fetches thes system artifact though | 09:51 |
| pedroalvarez | I mean, I like that we produce those artifacts, given that you can deploy them directly, forgetting about building. | 09:52 |
| persia | tiagogomes_: Why was that odd? | 09:52 |
| *** locallycompact has joined #baserock | 09:56 | |
| tiagogomes_ | persia, if I run `morph build`, I expect something to be built and have all pieces used in the system stored locally, not just the system artifact | 09:57 |
| persia | Ah. I understand. I personally find that behaviour sufficiently annoying to not want to use the tool anymore, because it adds hours to days to the process of grabbing something that someone else built. | 09:58 |
| persia | But your network connectivity to cache may differ from mine :) | 09:58 |
| tiagogomes_ | If we still care about the deploy phase to be arch-independent, I think the best solution is to construct the system artifact as some sort of overlay file system | 10:06 |
| pedroalvarez | I believe we care | 10:07 |
| pedroalvarez | I care | 10:08 |
| *** ssam2 has joined #baserock | 10:10 | |
| *** ChanServ sets mode: +v ssam2 | 10:10 | |
| *** edcragg has joined #baserock | 10:16 | |
| *** ssam2 has quit IRC | 10:17 | |
| *** ssam2 has joined #baserock | 10:19 | |
| *** ChanServ sets mode: +v ssam2 | 10:19 | |
| benbrown_ | hmm, git.baserock.org's delta/git.git hasn't had any changes for 6 months, that seems a little off? | 10:48 |
| pedroalvarez | benbrown_: I'll have a look | 10:49 |
| ssam2 | ha, I remember a bug where it was pushing to .git or something instead :-) | 10:49 |
| ssam2 | thought that was fixed though | 10:50 |
| pedroalvarez | here we go: http://paste.baserock.org/ijurufuyis | 10:54 |
| pedroalvarez | those maint-* branches don't exsit anymore | 10:56 |
| persia | Does Lorry produce any sort of event stream for exceptions (e.g. lorry failures) that can trigger alerts of some sort? | 11:00 |
| paulsherwood | not afaik | 11:01 |
| * paulsherwood gets at least one email a week about git, from Perforce :/ | 11:03 | |
| pedroalvarez | k, this patch should fix it: https://gerrit.baserock.org/1653 | 11:03 |
| pedroalvarez | benbrown_: thanks for raising the issue | 11:04 |
| benbrown_ | pedroalvarez: np | 11:04 |
| ssam2 | persia: lorry controller has that info and could be extended to send email alerts | 11:14 |
| ssam2 | i think right now it would generate more alerts than we have time to fix, though | 11:14 |
| pedroalvarez | benbrown_: fixed now | 11:15 |
| benbrown_ | pedroalvarez: \o/ | 11:15 |
| tiagogomes_ | IRC notifications? | 11:17 |
| gtristan | Hey guys... I have to go eat a (very late) lunch... but there is another patch storm | 11:19 |
| * pedroalvarez grabs the umbrella | 11:20 | |
| gtristan | There is one known issue with it... the empathy definition needs to use a different ref than the one in there | 11:20 |
| gtristan | but I cant fix it | 11:20 |
| persia | ssam2: Understood that it requires admin to address things, but I thought having a public log that benbrown_ could have checked quickly to see if git was out-of-date before asking, to ease investigation might be nice. | 11:20 |
| gtristan | cause it has to refer to a branch that points the gitmodules to the right modules | 11:20 |
| pedroalvarez | gtristan: I have some work for you for when you come back | 11:21 |
| ssam2 | persia: sorry, i misunderstood. making lorry-controller's logs public would be great | 11:21 |
| gtristan | I can fix that after https://gerrit.baserock.org/#/c/1626/ gets applied | 11:21 |
| ssam2 | l-c has a web interface, but it hides a lot of useful info at the moment | 11:21 |
| ssam2 | http://git.baserock.org/lc-status.html is all we have | 11:21 |
| persia | ssam2: To my mind, the logs would be great, and a quick dashboard summary somewhere (like with the Mason one) would be doubleplusgood, but yeah, lack of tuits :) | 11:22 |
| ssam2 | but we could make the actual logs public with a bit of tweaking to the code | 11:22 |
| persia | If that is relatively easy, that would probably be a great step. | 11:23 |
| *** gtristan has quit IRC | 11:28 | |
| radiofree | pedroalvarez: is the libvpx build on arm thing fixed? | 11:30 |
| *** jjardon has quit IRC | 11:51 | |
| *** ssam2 has quit IRC | 11:59 | |
| *** jjardon has joined #baserock | 12:00 | |
| radiofree | i switched over the flashing script to use gdisk rather than fdisk, so it should work on systems with gnu fdisk/util-linux <2.25 | 12:02 |
| *** jjardon has quit IRC | 12:11 | |
| *** ssam2 has joined #baserock | 12:13 | |
| *** ChanServ sets mode: +v ssam2 | 12:13 | |
| *** gtristan has joined #baserock | 12:26 | |
| *** jjardon has joined #baserock | 12:32 | |
| pedroalvarez | radiofree: nope, didn't have the time | 12:32 |
| gtristan | pedroalvarez, ... so what is it ? | 12:56 |
| gtristan | pedroalvarez, is this cross compiling ? | 12:56 |
| radiofree | yes | 12:56 |
| radiofree | it assumes you'll be building for arm by cross compiling | 12:56 |
| gtristan | s/it/pedroalvares ? | 12:56 |
| gtristan | hehe | 12:56 |
| radiofree | it does build however, you just need to modify configure.sh | 12:57 |
| pedroalvarez | gtristan: what I wanted to say before was https://mason-x86-64.baserock.org/log/e425c907b0a7a08873493520b2fce0b12e881cb4--2015-12-17%2012:52:04.log | 12:57 |
| radiofree | also, it tries to build it softfp for some reason, the detection is totally buggered | 12:57 |
| radiofree | (which also causes it to fail) | 12:57 |
| radiofree | however if you force hardfp and "gcc" as the name of the compiler in configure.sh it configures/builds fine | 12:58 |
| gtristan | pedroalvarez, I dont get it... https://git.gnome.org/browse/libgd | 13:01 |
| gtristan | that exists | 13:01 |
| gtristan | fatal: Unable to look up git.gnome.org | 13:01 |
| SotK | gtristan: mason can't see the outside world | 13:02 |
| *** jjardon has quit IRC | 13:02 | |
| gtristan | Oh wait | 13:03 |
| *** jjardon_ has joined #baserock | 13:03 | |
| gtristan | Ah | 13:05 |
| gtristan | pedroalvarez, I didnt realize about totem, I have that already done for gedit | 13:05 |
| gtristan | pedroalvarez, alright will fix | 13:05 |
| *** ssam2 has quit IRC | 13:08 | |
| *** jjardon_ is now known as jjardon | 13:23 | |
| * gtristan hands pedroalvarez https://gerrit.baserock.org/#/c/1654/ | 13:23 | |
| *** franred has quit IRC | 13:25 | |
| gtristan | radiofree, I feel there is a part of the conversation I missed | 13:32 |
| gtristan | or that we're mixing topics, or something | 13:32 |
| gtristan | pedroalvarez, I'm about to get outa here... would you like me to just push that gitmodules fix ? | 13:46 |
| gtristan | or leave it in gerrit ? | 13:47 |
| * gtristan has since built with the fix applied, fwiw | 13:47 | |
| gtristan | SotK, oh, sorry I missed the storyboard story... I look forward to seeing the new-and-improved version :) | 13:49 |
| gtristan | before I run out... leaving this as food for thought: https://gerrit.baserock.org/#/c/1621/ | 13:55 |
| gtristan | in my patch, I add a separate lorry file to lorry a different tarball from the same webkitgtk upstream - this could technically be done from the same lorry, but it would mean A.) The new commit would be an older version from the original, so history would appear backwards... and B.) Further releases could technically happen to both... as 2.4.x is the *last* webkitgtk branch which has support for the older, parallel installable API | 13:57 |
| gtristan | and yeah, we need to build both of them. | 13:57 |
| gtristan | ok ! | 13:57 |
| gtristan | laterz | 13:57 |
| *** gtristan has quit IRC | 14:01 | |
| *** ssam2 has joined #baserock | 14:12 | |
| *** ChanServ sets mode: +v ssam2 | 14:12 | |
| pedroalvarez | ouch, I think this totem fix will trigger some conflicts | 14:12 |
| *** ssam2 has quit IRC | 15:02 | |
| *** ssam2 has joined #baserock | 15:17 | |
| *** ChanServ sets mode: +v ssam2 | 15:17 | |
| *** paulw has joined #baserock | 15:56 | |
| pedroalvarez | openid down? | 16:08 |
| *** paulw has quit IRC | 16:10 | |
| Zara | context: I'd forgotten my password and couldn't reset it. | 16:10 |
| Zara | page loads for ages, and I get a 500 internal server error | 16:11 |
| Zara | hm, doesn't seem to work when I guess a password either, but I think other people have been able to login? | 16:12 |
| pedroalvarez | more context: tried to reproduce the error, 500 internal server error | 16:12 |
| Zara | \o/ | 16:12 |
| pedroalvarez | I mean, I cannot, page doesn't load at all | 16:13 |
| SotK | it took me a while to log in | 16:13 |
| Zara | did I break it by trying to log in? that's the kind of day we're having today. | 16:13 |
| *** ctbruce has quit IRC | 16:21 | |
| *** toscalix has quit IRC | 16:36 | |
| pedroalvarez | found the error | 16:52 |
| pedroalvarez | fixed it thanks to ssam2 | 16:52 |
| pedroalvarez | and now hating ops more :) | 16:52 |
| *** mwilliams_ct has left #baserock | 16:54 | |
| pedroalvarez | Zara: you can try to recover the password now | 16:55 |
| Zara | okay, thanks | 16:56 |
| Zara | oh, I think I guessed it right anyway | 16:57 |
| Zara | it wasn't working before so I thought I had the wrong one, but I didn't | 16:57 |
| Zara | cool, thanks for getting things working again :) | 16:57 |
| pedroalvarez | the database was down, so maybe it wasn't working because it couldnt check your password | 16:57 |
| pedroalvarez | makes sense | 16:58 |
| Zara | ah, right | 16:58 |
| Zara | yeah | 16:58 |
| *** tiagogomes_ has quit IRC | 17:19 | |
| *** CTtpollard has quit IRC | 17:19 | |
| *** ssam2 has quit IRC | 17:51 | |
| *** bashrc has quit IRC | 18:02 | |
| *** gtristan has joined #baserock | 18:04 | |
| *** rdale has quit IRC | 18:05 | |
| jjardon | pedroalvarez: maybe this is of your interest: https://gerrit.baserock.org/#/c/1655/ | 19:20 |
| *** edcragg has quit IRC | 19:33 | |
| pedroalvarez | Wow! Tested? | 19:52 |
| jjardon | pedroalvarez: yes, weston without parameters runs fine | 19:53 |
| *** locallycompact has quit IRC | 22:04 | |
| *** edcragg has joined #baserock | 22:07 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!