[buildiso] dbus in chroot/machine-id again #58

Closed
opened 2015-02-17 19:03:49 +00:00 by philm · 12 comments
philm commented 2015-02-17 19:03:49 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

I just open it as a reminder for us to possibly investigate.

Something is still strange with machine-id.

(172/246) installing libgksu                                          [#######################################] 100%
process 18251: D-Bus library appears to be incorrectly set up; failed to read machine uuid: Failed to open "/etc/machine-id": No such file or directory
See the manual page for dbus-uuidgen to correct this issue.
  D-Bus not built with -rdynamic so unable to print a backtrace
usr/sbin/gconfpkg: line 18: 18251 Aborted                 GCONF_CONFIG_SOURCE=`/usr/bin/gconftool-2 --get-default-source` /usr/bin/gconftool-2 --makefile-install-rule /usr/share/gconf/schemas/${pkgname}.schemas > /dev/null
error: command failed to execute correctly

Profile is openrc based.

*Created by: udeved* I just open it as a reminder for us to possibly investigate. Something is still strange with machine-id. ``` (172/246) installing libgksu [#######################################] 100% process 18251: D-Bus library appears to be incorrectly set up; failed to read machine uuid: Failed to open "/etc/machine-id": No such file or directory See the manual page for dbus-uuidgen to correct this issue. D-Bus not built with -rdynamic so unable to print a backtrace usr/sbin/gconfpkg: line 18: 18251 Aborted GCONF_CONFIG_SOURCE=`/usr/bin/gconftool-2 --get-default-source` /usr/bin/gconftool-2 --makefile-install-rule /usr/share/gconf/schemas/${pkgname}.schemas > /dev/null error: command failed to execute correctly ``` Profile is openrc based.
philm commented 2015-02-17 19:18:48 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: Kirek

Maybe you need to initialize it manually when mounting the chroot "dbus-uuidgen > /etc/machine-id"?

*Created by: Kirek* Maybe you need to initialize it manually when mounting the chroot "dbus-uuidgen > /etc/machine-id"?
philm commented 2015-02-17 19:21:17 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

its nothing disturbing the build process, and I also don't get some dbus process killed like in some systemd profiles.
Its just curiosity to track it, and probably approach the uuid creation differently in chroot.
Atm, we do uuid at livesession boot up.

*Created by: udeved* its nothing disturbing the build process, and I also don't get some dbus process killed like in some systemd profiles. Its just curiosity to track it, and probably approach the uuid creation differently in chroot. Atm, we do uuid at livesession boot up.
philm commented 2015-02-17 19:23:05 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: manjaro

So this openrc related only?

*Created by: manjaro* So this openrc related only?
philm commented 2015-02-17 19:25:54 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

I dunno, the error in not new.
But this is perhaps reason why systemd profiles need dbus process killed.

its no problem with openrc, because dbus script runs uuid-gen when it starts and no uuid file is present.

*Created by: udeved* I dunno, the error in not new. But this is perhaps reason why systemd profiles need dbus process killed. its no problem with openrc, because dbus script runs uuid-gen when it starts and no uuid file is present.
philm commented 2015-02-19 16:08:00 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

I did a test.

Profile: lxqt-openrc

without supplying a machne-id file in chroot/var/lib/dbus, no dbus process is killed when building.

If I supply a machine-id on custom image, dbus processes are killed when building, but the above error message is gone.

A new PrivateDevices= switch has been added to service units
      which allows running a service with a namespaced /dev
      directory that does not contain any device nodes for
      physical devices. More specifically, it only includes devices
      such as /dev/null, /dev/urandom, and /dev/zero which are API
      entry points.

http://rpmfind.net/linux/RPM/opensuse/updates/13.2/x86_64/libgudev-1_0-devel-210-25.8.1.x86_64.html

*Created by: udeved* I did a test. Profile: lxqt-openrc without supplying a machne-id file in chroot/var/lib/dbus, no dbus process is killed when building. If I supply a machine-id on custom image, dbus processes are killed when building, but the above error message is gone. ``` A new PrivateDevices= switch has been added to service units which allows running a service with a namespaced /dev directory that does not contain any device nodes for physical devices. More specifically, it only includes devices such as /dev/null, /dev/urandom, and /dev/zero which are API entry points. ``` http://rpmfind.net/linux/RPM/opensuse/updates/13.2/x86_64/libgudev-1_0-devel-210-25.8.1.x86_64.html
philm commented 2015-02-23 11:54:09 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: philmmanjaro

hmm. really weird issue, so ...

*Created by: philmmanjaro* hmm. really weird issue, so ...
philm commented 2015-03-01 16:30:39 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

I notice that this error is related to gtk/gnome whatever packages.
I won'tget this error with pure qt apps, but as soon as I add gtk apps, the dbus process spawned in chroot needs to be killed.

*Created by: udeved* I notice that this error is related to gtk/gnome whatever packages. I won'tget this error with pure qt apps, but as soon as I add gtk apps, the dbus process spawned in chroot needs to be killed.
philm commented 2015-03-03 16:22:59 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: philmmanjaro

I noticed it once again during building KDE5 images.

*Created by: philmmanjaro* I noticed it once again during building KDE5 images.
philm commented 2015-03-05 18:44:49 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

You mean the kde5 big profile?
Would stillbe nice to have a list what packages cause this.
No issue with the minimal one.

*Created by: udeved* You mean the kde5 big profile? Would stillbe nice to have a list what packages cause this. No issue with the minimal one.
philm commented 2015-05-11 12:22:09 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: philmmanjaro

Well, it happens also with XFCE. You can follow it within my build-logs ...

*Created by: philmmanjaro* Well, it happens also with XFCE. You can follow it within my [build-logs](https://gist.github.com/philmmanjaro) ...
philm commented 2015-06-05 06:30:40 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

(205/279) installing libgksu                            [##############################] 100%
process 28255: D-Bus library appears to be incorrectly set up; failed to read machine uuid: Failed to open "/etc/machine-id": No such file or directory
See the manual page for dbus-uuidgen to correct this issue.
  D-Bus not built with -rdynamic so unable to print a backtrace
usr/sbin/gconfpkg: line 18: 28255 Aborted                 (core dumped) GCONF_CONFIG_SOURCE=`/usr/bin/gconftool-2 --get-default-source` /usr/bin/gconftool-2 --makefile-install-rule /usr/share/gconf/schemas/${pkgname}.schemas > /dev/null
error: command failed to execute correctly

That' with systemd build host.

*Created by: udeved* ``` (205/279) installing libgksu [##############################] 100% process 28255: D-Bus library appears to be incorrectly set up; failed to read machine uuid: Failed to open "/etc/machine-id": No such file or directory See the manual page for dbus-uuidgen to correct this issue. D-Bus not built with -rdynamic so unable to print a backtrace usr/sbin/gconfpkg: line 18: 28255 Aborted (core dumped) GCONF_CONFIG_SOURCE=`/usr/bin/gconftool-2 --get-default-source` /usr/bin/gconftool-2 --makefile-install-rule /usr/share/gconf/schemas/${pkgname}.schemas > /dev/null error: command failed to execute correctly ``` That' with systemd build host.
philm commented 2015-11-26 15:56:44 +00:00 (Migrated from gitlab2.manjaro.org)

Created by: udeved

Just adding #150 to keep track of the never ending dbus adventure.

*Created by: udeved* Just adding #150 to keep track of the never ending dbus adventure.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
tools/manjaro-tools#58
No description provided.