=== ePierre_ is now known as ePierre [10:52] Hello everybody. I am aware of the recent build farm issues, but we have a build stuck for a couple of days now. It is not urgent, so we can live with this for now. Still, is there a page somewhere to monitor how deep the build queue is? It would help us getting an idea about the timelines of these builds. [10:56] I'll answer myself, https://launchpad.net/builders. Have a nice day everybody. [16:00] when will launchpad get fixed? my packages have been bouncing between 2 to 6 hours, for the last 20hours [16:04] patdk-lap: It's not currently particularly broken (aside from not having Spectre-mitigation microcode for a chunk of our x86 farm, which is in the hands of Intel and AMD), just a longish queue. I can score up individual builds if they're urgent. [16:06] Your build times keep jumping up because new builds arrive with a higher score and jump the queue. [16:07] Which happens when we can't quite service the rate of incoming builds quickly enough. [16:34] ya, just attempting to get this dovecot cve patched, but ppa's are kindof low priority so they take awhile [16:34] I don't mind 12hour waits, but 24 is heh [16:35] times like this I should just buildup my own internal systems to do it, but it is a lot of overhead [16:36] patdk-lap: like I say, happy to score up things people are actually waiting for; just need to know the build URLs [16:36] https://launchpad.net/~patrickdk/+archive/ubuntu/testing/+packages [16:36] just waiting on the amd64's [16:39] patdk-lap: all building now [16:40] nice :) [16:40] once I test them and verify, I'll copy don't rebuild to production repo [16:40] that shouldn't cause a wait again so all should be good [16:41] indeed [16:42] think the only other time I had this issue was during an lts release [16:43] if it wasn't for that auth memleak cve issue I wouldn't really mind at all [16:45] Yes, pre-Spectre we hadn't had much of a queue for a while. [16:48] ah ya, I didn't think of that, wonder how much the builders slowed down from their patches [16:50] looks good, a new unit test failed for xenial, but otherwise fine, xenial can wait [18:39] in short, I can reach https://login.ubuntu.com/ and i am said to be logged in, but when I visit https://bugs.launchpad.net/ubuntu/+source/libtasn1-6/+filebug/4ffd812c-02c3-11e8-9dd0-d485646cd9a4? i get "Oops! Sorry, something just went wrong in Launchpad. We’ve recorded what happened, and we’ll fix it as soon as possible. Apologies for the inconvenience. [18:39] Ubuntu bug 4 in Launchpad itself "Importing finished po doesn't change progressbar" [Medium,Fix released] [19:11] doublehp: Hopefully you got an OOPS ID? If so, please quote it [19:12] cjwatson: ... was my line trunckated ? [19:12] (Error ID: OOPS-99109a4ff4a3fe6a44f99c058fff1690) [19:12] https://oops.canonical.com/?oopsid=OOPS-99109a4ff4a3fe6a44f99c058fff1690 [19:12] Apparently so [19:13] OK, just a minute [19:23] doublehp: fixed now, sorry about that. you ran into https://bugs.launchpad.net/launchpad/+bug/1607242 due to an old comment from your email address that we imported from another bug tracker. [19:23] Ubuntu bug 1607242 in Launchpad itself "Logging into a placeholder person OOPSes if email address already on another person" [Critical,Triaged] [19:24] cjwatson: i will eat soon; i will be back in 30-45mn; I will go to bed in 90-120mn - after I left, either send me a memo via memoserv, or an email benoit@demaine.info [19:24] doublehp: uh, there should be nothing more to do, just try the login again [19:24] ok ... [19:24] like I say your account should be fixed now [19:29] first time i use that site, got bugs ... will i receive email notifications when people update my report ? [19:29] Yes [19:29] thanks, bye [19:29] np [21:37] heyyy guys [21:37] what is the version of debhelper available on xenial? [21:40] !info debhelper xenial [21:40] debhelper (source: debhelper): helper programs for debian/rules. In component main, is optional. Version 9.20160115ubuntu3 (xenial), package size 721 kB, installed size 1300 kB [21:40] jfmcarreira: there is also a more recent one in xenial-backports [21:45] nacc: i have debhelper (>= 9.0.0), and it is not being installed [21:46] jfmcarreira: where? do you have logs [21:46] jfmcarreira: it might mean you have other deps that break it [21:48] https://launchpadlibrarian.net/354633756/buildlog.txt.gz [21:49] nacc: as in the past I wanted version 10 I copied from backports to my ppa. but it should not be a problem. it could also use version 10 [21:50] jfmcarreira: have you tried to reproduce it in a chdist? [21:50] jfmcarreira: or build it locally to see? [21:50] debuild -uc -us -b?? [21:51] jfmcarreira: with -backports enabled, presumably [21:51] i'd use an sbuild or a lxd to do it [21:57] whats the command to auto download source? [21:57] trying to use sbuild [21:57] jfmcarreira: how do you mean? [21:58] the comand to download the source based on info on debian folder [21:59] jfmcarreira: if you have the debian folder, you have the source package already? [21:59] jfmcarreira: sorry, i'm not following [22:00] uscan [22:00] i mean uscan [22:00] fetch the source doe [22:00] code* [22:00] sbuild requires [22:00] uscan and uupdate [22:00] they get the orig tarballs [22:00] nacc: yeah thats it. sbuild required it [22:00] jfmcarreira: you can also just download the files from the ppa [22:01] nacc: done. [22:02] now i get an error with debuild: "dh_install: Cannot find (any matches for) "usr/lib/libPlaYUVerLib.so*" (tried in "." and "debian/tmp")" with the pattern because it puts in usr/lib/x86_64-linux-gnu/libPlaYUVerLib.so* [22:03] jfmcarreira: wasn't that the error you got before? [22:04] yeah because before i had usr/lib/*/libPlaYUVerLib.so* [22:04] and I removed the extra "/*" [22:05] cant run sbuild [22:05] http://paste.ubuntu.com/26467048/ [22:06] jfmcarreira: how are you invoking it? [22:06] sbuild only [22:07] jfmcarreira: so you are in the source package directory? [22:08] yeah and i ran uscan and it downloaded the source tar to the directory above [22:09] jfmcarreira: sorry, i'm confused; have you used sbuild before? [22:09] i ran uscan and then sbuild [22:09] jfmcarreira: you need to give it a source package directory. not the debian/ and not just the upstream [22:09] first time using it [22:09] jfmcarreira: yeah, that's not quite how you use it [22:09] jfmcarreira: do you konw what a debian source package is? [22:10] not really... it is a combination of source files and control files to describe the pkg and the build? [22:10] jfmcarreira: right, basically, it's an upstream orig tarball(s), untarred, and a debian tarball (untarred) [22:11] jfmcarreira: so you need to be in that untarred state to build it, or use the dsc [22:12] i have this playuver-0.18.2.tar.gz and playuver_0.18.2-2.debian.tar.xz which were downloaded automatically from github.. at least the first [22:13] this file is empty playuver_0.18.2-2.debian.tar.xz [22:13] jfmcarreira: well, it certinaly shouldn't be empty [22:14] how should i fill it? [22:14] i can run debuild to create a source package [22:14] jfmcarreira: this might be more of a rathole than is nencessary [22:14] jfmcarreira: if you can build a source package, pass that source pacakge to sbuild [22:14] where should i be to build a src pkg? [22:15] source folder? debian folder? [22:16] jfmcarreira: you shulld be in the untarred source packages root directory [22:17] with the debian folder? [22:19] do i need to be in the src folder and have .orig above it? [22:19] jfmcarreira: it's better to be specific here [22:19] src might mean a directory called 'src' or you might mean 'source' [22:20] i mean having the source code folder with a debian folder inside [22:20] you should be in a directory containing the uncompressed orig tarball, a debian/ and the .orig tarball should be in the parent directory, in my experience [22:20] now i tried inside the debian folder with the source in ../XXX.orig [22:21] now it created the .changes, .build, .dsc [22:21] what do i do with it? [22:22] you should pass the .dsc to sbuild === newell is now known as Guest48853 [22:31] nacc: i do need to create a chroot right? [22:34] jfmcarreira: yeah with mk-sbuild [22:34] sbuild-createchroot --include=eatmydata,ccache,gnupg unstable /local/jcarreira.it/unstable-amd64-sbuild http://deb.debian.org/debian [22:35] jfmcarreira: well, if you want to build a debian unstable one... [22:35] jfmcarreira: but i thought you were building xenial? [22:35] i do [22:35] xenial and bionic and other debian distro [22:36] jfmcarreira: those are ubuntu [22:36] jfmcarreira: not debian [22:38] nacc: short term goal are ubuntu xenial and bionic cause are the ones i use in my research lab. but I wanted to deploy it to as much distros ass possible [22:38] the xenial problem were raised becasue of the launchpad build recipes [22:39] jfmcarreira: i don't have any context into what you are doing, but if this is an applilcation, why ont build a snap? [22:40] nacc: why snap? why not debian packages? to be honest I am not big fan of snap [22:41] jfmcarreira: because it avoids these issues? :) [22:41] jfmcarreira: and debian packages only work on debia-based distros; snaps are cross-distro [22:41] nacc: the issue here seems quiet stupid. it is a problem with a dependency which exists on xenial [22:41] jfmcarreira: it's something else, most likely [22:42] jfmcarreira: for whatever reason, due to other dependencies, most likely, you are not able to install debhelper whenn buildig your srcpkg [22:42] jfmcarreira: but hey, in a snap you don't need debhelper at all [22:42] nacc: based on that it makes sense to deploy a snap package. but i would like to be able to deploy debian pkgs for xenial and bionic (next LTS) as I need to have daily builds for it [22:42] jfmcarreira: like i said, if this is an application, snaps feel like a natural fit, esp. if you aren't already committed to doing debs (i mean, you don't have a xeniall build yet) [22:43] jfmcarreira: you cann do daily builds with sanps [22:43] *snaps [22:43] that's waht 'edge' is for [22:43] nacc: i have xenial build for a long time. but in the past days i tried to split the pkgs into multiple apps and thats what raised problems [22:44] jfmcarreira: oh i didn't know that [22:44] jfmcarreira: sorry, i need to refocus on some other stuff before i eod [22:45] jfmcarreira: but i'd seriously look at snaps. you'd then only have one versionn (well, 4 if you do all for tracks) to worry about everywhere [22:45] 9 [22:45] oops, sorry [22:46] nacc: ok thanks