/srv/irclogs.ubuntu.com/2009/01/27/#ubuntu-arm.txt

doko_somebody aware of the hal install failure on armel? causes build failures on the buildds (openjdk-6)10:03
ograinstall failure ?10:05
ograit installed fine yesterday here in a local bootstrap10:05
doko_ogra: https://edge.launchpad.net/ubuntu/+source/openjdk-6/6b14-0ubuntu10/+build/84881810:08
ograthats udev, no ?10:11
ogracat: /sys/kernel/uevent_seqnum: No such file or directory10:11
ograinvoke-rc.d: initscript udev, action "restart" failed.10:11
ogradpkg: error processing udev (--configure):10:11
ogra subprocess post-installation script returned error exit status 110:11
doko_maybe10:15
ograwell, i pinged keybuk ...10:15
suihkulokkiogra: that'n nutty, buildd's shouldn't be installing hal/udev to the build chroot.14:50
suihkulokkiogra: check which package is pulling them in14:50
ograsuihkulokki, its already fixed in udev14:51
ograudev shouldnt start if installed in a chroot14:51
ograthere are packages that depend on stuff that in turn depends on hal14:51
ograso you cant really avoid things like that14:51
ograpackages like hal, dbus and udev need special casing14:52
suihkulokkino14:52
Stskeepsshouldn't they "just" fake invoke-rc.d and start-stop-daemon? :P14:52
ograall of xorg depends on hal14:52
suihkulokkiyou build-depend on libhal-dev libdus-dev etc14:52
ograStskeeps, thats what they do, but udev had a bad grep in the posinst before14:53
suihkulokkiand none of those -dev packages should pull their respective daemon in14:53
ografeel free to look at the build log14:53
ograbut i know the fix got into udev already so that might be moot14:53
ograhttp://launchpadlibrarian.net/21658731/buildlog_ubuntu-jaunty-armel.openjdk-6_6b14-0ubuntu10_FAILEDTOBUILD.txt.gz14:53
ograhttps://lists.ubuntu.com/archives/jaunty-changes/2009-January/003929.html is the fix14:55
suihkulokkiogra: that's fixing the symptom, not fixing the root of the problem - ie. "why is udev getting installed in the first place"14:57
ograprobably becuse the package needs a running X env ... we have a bunch of packages that need xvfb for example ...14:59
ogradont ask me, i'm neither maintaining the buildds, nor maintain the openjdk package nor am i resonsible for udev :)14:59
ograand the udev maintainer obviously considered that fis as appropriate14:59
persiasuihkulokki, Probably a side effect of recommends-by-default15:00
suihkulokkipersia: ...which is why recommends-by-default is not used on buildd's ;)15:00
suihkulokkiat least, on debian that is15:00
ograbeyond fixing that particular buildd issue it will also fix chroot building in general (in case you dont have sysfs mounted in your chroot during bootstrapping)15:00
ograwell, ubuntu doesnt use dak :)15:01
persiasuihkulokki, Heh.  Well, that's one way to fix it.15:01
persiaPersonally, I'd rather it did, because many maintainers seem to build packages on their personal local systems, with all sorts of random stuff installed, and certainly recommends.15:01
persiaMind you, this is getting *lots* better, but still.15:02
suihkulokkiogra: I don't claim that udev bug was not worth fixing - merely that there is most likely still a unnecessary depends somewhere15:04
suihkulokkibut if you don't care, whatever.15:04
ograsuihkulokki, well, ask doko_ about it ... :)15:04
ograhe was bringing it up since he tried to build openjdk15:04
ograno idea what pulls it in15:04
doko_suihkulokki: pulseaudio sucks it in15:05
persiaAnd pulse needs to do device analysis if the daemon is running.15:06
persiadoko_, Why does OpenJDK need both libpulse-dev and pulseaudio to build?  Would libpulse-dev itself not be sufficient?15:07
persiaIs this for the test suite?15:07
doko_AC_DEFUN([FIND_PULSEAUDIO],15:09
doko_[15:09
doko_  AC_PATH_PROG(PULSEAUDIO_BIN, "pulseaudio")15:09
doko_  if test -z "${PULSEAUDIO_BIN}"; then15:09
doko_    AC_MSG_ERROR("pulseaudio was not found.")15:09
doko_  fi15:09
doko_  AC_SUBST(PULSEAUDIO_BIN)15:09
doko_])15:09
persiaBlech.  No helping it then, really.15:11
* suihkulokki grumbles15:11
suihkulokki1) i would still check if that configure check can be fixed15:15
suihkulokki2) if pulseaudio does really need to depend on hal15:15
suihkulokkithen again, if the build system is pulling in recommends, this ranting is completly moot15:15
ograthere are more packages like that15:15
ograall the stuff that uses xvfb for their test suites will run into the same now that xorg hard depends on hal15:16
ograi think you should start to consider it a common case that udev *might* show up in buildd chroots15:17
suihkulokkithat depends on how much you care about quality and bloat15:18
ograwell, self tests can improve the quality ... so i'D rather call it a philosophical thing :)15:18
suihkulokkiI doubt even with current xorg xvfb would really hard-depend on hal15:19
suihkulokkiso in other words, someone took the lazy route15:19
ograxvbf runs X15:19
ograand X needs hal to run15:19
ogra*vfb15:20
* ogra humms "times are changin ..."15:21
* suihkulokki humms unverified bullshit15:21
ograheh15:22
ograwell, fix X to not use hal ... but i doubt upstream will accept any patches that revert this behavior ... they made quite a hard cut and dropped much backwards compatibility with this release15:23
suihkulokkiyou are confusing the dependency on libhal and dependency on hald15:28
ograxorg moaning loudly if hald doesnt run doesnt help though :) but right, it doesnt have a dep on hal itself15:33
NCommanderhola16:15
ScriptRipperNCommander: hi21:17
NCommanderhey ScriptRipper23:15
ScriptRipperDid you think about the e-mail for the apt repo layout?23:15
ScriptRipperNCommander: your last comment was: i think about it under the shower, if i remeber correctly :)23:17
ScriptRipperNCommander: what did your shower say :)23:18
NCommanderScriptRipper, "ack, it burns"23:23

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!