Discussion:
systemd-197-r1 starts gdm-3.6.2
(too old to reply)
Stefan G. Weichinger
2013-01-30 17:02:20 UTC
Permalink
Would someone mind and take a look?

As mentioned in the former thread (I decided to start a new one) I can't
login to Gnome via gdm right now ... when running the system with
systemd (it worked before with openrc, could check back, yes ...).

I created a brand new user to rule that out, same behavior.

use-flags:

[I] gnome-base/gdm
Available versions: 2.20.11-r1 [M](~)2.32.1-r1 (~)3.4.1-r3
(~)3.6.2 **9999[1] {accessibility afs audit branding +consolekit debug
dmx +fallback fprint gnome-keyring +gnome-shell +introspection ipv6 ldap
pam plymouth remote selinux smartcard systemd tcpd test xinerama
+xklavier ELIBC="glibc"}
Installed versions: 3.6.2(22:49:57 28.01.2013)(fallback
gnome-shell introspection ipv6 systemd tcpd xinerama ....

[I] gnome-base/gnome-session
Available versions: 2.32.1-r3 (~)3.4.2.1 (~)3.6.2 (~)3.6.2-r1
**9999[1] {debug doc gconf ipv6 systemd ELIBC="FreeBSD"}
Installed versions: 3.6.2-r1(17:40:19 30.01.2013)(ipv6 systemd ....


The following log is from /var/log/messages ... I start gdm.service, try
to log in, get thrown back to the user selection (-> this is the
problem...) and stop gdm.service.

What I tried:

* downgraded accountsservice (found that mentioned:
https://bbs.archlinux.org/viewtopic.php?pid=1198455)

* fixed bluetooth.service (threw some errors, now ok afaik)

* rebuilt all around dbus, gdm, pulseaudio ... etc

* when I stop gdm, and start xdm.service I am able to log in and use
Gnome ...

-

Canek, maybe you have an idea?

Thanks in advance, Stefan

And here the log:

Jan 30 17:46:30 hiro root: START gdm
Jan 30 17:46:36 hiro dbus[4643]: [system] Rejected send message, 1
matched rules; type="method_call", sender=":1.9" (uid=0 pid=5095
comm="/usr/bin/gdm --nodaemon ")
interface="org.freedesktop.DBus.Properties" member="GetAll" error
name="(unset)" requested_reply="0" destination=":1.10" (uid=0 pid=5097
comm="/usr/libexec/gdm-simple-slave --display-id /org/gn")
Jan 30 17:46:36 hiro acpid: client 4884[0:0] has disconnected
Jan 30 17:46:36 hiro acpid: client connected from 5101[0:0]
Jan 30 17:46:36 hiro acpid: 1 client rule loaded
Jan 30 17:46:36 hiro acpid: client connected from 5101[0:0]
Jan 30 17:46:36 hiro acpid: 1 client rule loaded
Jan 30 17:46:37 hiro gdm-simple-slave[5097]: WARNING: Failed to give
slave programs access to the display. Trying to proceed.
Jan 30 17:46:37 hiro dbus[4643]: [system] Activating via systemd:
service name='org.freedesktop.Accounts' unit='accounts-daemon.service'
Jan 30 17:46:37 hiro accounts-daemon[5117]: started daemon version 0.6.22
Jan 30 17:46:37 hiro dbus[4643]: [system] Successfully activated service
'org.freedesktop.Accounts'
Jan 30 17:46:37 hiro gdm-launch-environment][5115]:
pam_unix(gdm-launch-environment:session): session opened for user gdm by
(uid=0)
Jan 30 17:46:37 hiro gdm-launch-environment][5115]:
pam_ck_connector(gdm-launch-environment:session): nox11 mode, ignoring
PAM_TTY :0
Jan 30 17:46:37 hiro dbus[4643]: [system] Activating via systemd:
service name='org.freedesktop.UPower' unit='upower.service'
Jan 30 17:46:37 hiro dbus[4643]: [system] Successfully activated service
'org.freedesktop.UPower'
Jan 30 17:46:38 hiro kernel: ata1.00: configured for UDMA/133
Jan 30 17:46:38 hiro kernel: ata1: EH complete
Jan 30 17:46:38 hiro pulseaudio[5191]: [pulseaudio] authkey.c: Failed to
open cookie file '/var/lib/gdm/.config/pulse/cookie': Datei oder
Verzeichnis nicht gefunden
Jan 30 17:46:38 hiro pulseaudio[5191]: [pulseaudio] authkey.c: Failed to
load authorization key '/var/lib/gdm/.config/pulse/cookie': Datei oder
Verzeichnis nicht gefunden
Jan 30 17:46:38 hiro dbus[4643]: [system] Activating via systemd:
service name='org.freedesktop.RealtimeKit1' unit='rtkit-daemon.service'
Jan 30 17:46:38 hiro kernel: ata2.00: configured for UDMA/133
Jan 30 17:46:38 hiro kernel: ata2: EH complete
Jan 30 17:46:38 hiro kernel: ata6.00: configured for UDMA/133
Jan 30 17:46:38 hiro kernel: ata6: EH complete
Jan 30 17:46:38 hiro dbus[4643]: [system] Successfully activated service
'org.freedesktop.RealtimeKit1'
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully called chroot.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully dropped privileges.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully limited resources.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Running.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Watchdog thread running.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Canary thread running.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully made thread 5194
of process 5194 (/usr/bin/pulseaudio) owned by '104' high priority at
nice level -11.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Supervising 1 threads of 1
processes of 1 users.
Jan 30 17:46:38 hiro kernel: EXT4-fs (sdc2): re-mounted. Opts: commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (sdc3): re-mounted. Opts:
user_xattr,commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-0): re-mounted. Opts: commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-9): re-mounted. Opts:
user_xattr,commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-6): re-mounted. Opts:
user_xattr,commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-2): re-mounted. Opts: commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-5): re-mounted. Opts:
user_xattr,commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-4): re-mounted. Opts: commit=0
Jan 30 17:46:38 hiro kernel: EXT4-fs (dm-1): re-mounted. Opts: commit=0
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully made thread 5237
of process 5194 (/usr/bin/pulseaudio) owned by '104' RT at priority 5.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Supervising 2 threads of 1
processes of 1 users.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully made thread 5238
of process 5194 (/usr/bin/pulseaudio) owned by '104' RT at priority 5.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Supervising 3 threads of 1
processes of 1 users.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Successfully made thread 5239
of process 5194 (/usr/bin/pulseaudio) owned by '104' RT at priority 5.
Jan 30 16:46:38 hiro rtkit-daemon[5195]: Supervising 4 threads of 1
processes of 1 users.
Jan 30 17:46:38 hiro dbus[4643]: [system] Activating via systemd:
service name='org.bluez' unit='dbus-org.bluez.service'
Jan 30 17:46:38 hiro bluetoothd[5240]: Bluetooth daemon 4.101
Jan 30 17:46:38 hiro dbus[4643]: [system] Successfully activated service
'org.bluez'
Jan 30 17:46:38 hiro bluetoothd[5240]: Starting SDP server
Jan 30 17:46:38 hiro kernel: Bluetooth: Core ver 2.16
Jan 30 17:46:38 hiro kernel: NET: Registered protocol family 31
Jan 30 17:46:38 hiro kernel: Bluetooth: HCI device and connection
manager initialized
Jan 30 17:46:38 hiro kernel: Bluetooth: HCI socket layer initialized
Jan 30 17:46:38 hiro kernel: Bluetooth: L2CAP socket layer initialized
Jan 30 17:46:38 hiro kernel: Bluetooth: SCO socket layer initialized
Jan 30 17:46:38 hiro kernel: Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Jan 30 17:46:38 hiro kernel: Bluetooth: BNEP filters: protocol multicast
Jan 30 17:46:38 hiro kernel: Bluetooth: BNEP socket layer initialized
Jan 30 17:46:38 hiro bluetoothd[5240]: Failed to open RFKILL control device
Jan 30 17:46:38 hiro pulseaudio[5194]: [pulseaudio] authkey.c: Failed to
open cookie file '/var/lib/gdm/.config/pulse/cookie': Datei oder
Verzeichnis nicht gefunden
Jan 30 17:46:38 hiro pulseaudio[5194]: [pulseaudio] authkey.c: Failed to
load authorization key '/var/lib/gdm/.config/pulse/cookie': Datei oder
Verzeichnis nicht gefunden
Jan 30 17:46:39 hiro dbus[4643]: [system] Activating via systemd:
service name='org.freedesktop.UDisks2' unit='udisks2.service'
Jan 30 17:46:39 hiro udisksd[5270]: udisks daemon version 2.0.91 starting
Jan 30 17:46:39 hiro dbus[4643]: [system] Successfully activated service
'org.freedesktop.UDisks2'
Jan 30 17:46:39 hiro udisksd[5270]: Acquired the name
org.freedesktop.UDisks2 on the system message bus
Jan 30 17:46:39 hiro polkitd[5057]: Registered Authentication Agent for
unix-session:c4 (system bus name :1.29 [gnome-shell --mode=gdm], object
path /org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8')
Jan 30 17:46:39 hiro dbus[4643]: [system] Rejected send message, 2
matched rules; type="method_call", sender=":1.29" (uid=104 pid=5255
comm="gnome-shell --mode=gdm ")
interface="org.freedesktop.DBus.Properties" member="GetAll" error
name="(unset)" requested_reply="0" destination=":1.5" (uid=0 pid=4991
comm="/usr/sbin/console-kit-daemon --no-daemon ")
Jan 30 17:46:47 hiro gdm-password][5282]:
pam_unix(gdm-password:session): session opened for user sgw by
(unknown)(uid=0)
Jan 30 17:46:47 hiro gdm-password][5282]:
pam_ck_connector(gdm-password:session): nox11 mode, ignoring PAM_TTY :0
Jan 30 17:47:09 hiro gdm-simple-slave[5097]: GLib-GObject-CRITICAL:
g_object_ref: assertion `object->ref_count > 0' failed
Jan 30 17:47:09 hiro gnome-session[5126]: WARNING: Detected that
screensaver has left the bus
Jan 30 17:47:09 hiro gnome-session[5126]: CRITICAL:
gsm_manager_set_phase: assertion `GSM_IS_MANAGER (manager)' failed
Jan 30 17:47:09 hiro gnome-session[5126]: Gdk-WARNING: gnome-session:
Fatal IO error 11 (Die Ressource ist zur Zeit nicht verfügbar) on X
server :0.
Jan 30 17:47:09 hiro gdm-simple-slave[5097]: GLib-GObject-CRITICAL:
g_object_unref: assertion `object->ref_count > 0' failed
Jan 30 17:47:09 hiro polkitd[5057]: Unregistered Authentication Agent
for unix-session:c4 (system bus name :1.29, object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8')
(disconnected from bus)
Jan 30 17:47:17 hiro root: STOP gdm
Hinnerk van Bruinehsen
2013-01-30 17:36:15 UTC
Permalink
On Wed, Jan 30, 2013 at 06:02:20PM +0100, Stefan G. Weichinger wrote:
>
> Would someone mind and take a look?
> Fatal IO error 11 (Die Ressource ist zur Zeit nicht verfügbar) on X
> server :0.
> Jan 30 17:47:09 hiro gdm-simple-slave[5097]: GLib-GObject-CRITICAL:
> g_object_unref: assertion `object->ref_count > 0' failed
> Jan 30 17:47:09 hiro polkitd[5057]: Unregistered Authentication Agent
> for unix-session:c4 (system bus name :1.29, object path
> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8')
> (disconnected from bus)
> Jan 30 17:47:17 hiro root: STOP gdm

I've just installed systemd on one of my systems to give it a test and I
had similar problems due to the systemd useflag on policykit being
hardmasked (it also pulled in consolekit because of that).
Since the errors are very similar you may check your useflags on
policykit and - if necessary remove the use-mask of systemd for
policykit.

WKR
Hinnerk
Stefan G. Weichinger
2013-01-30 17:48:48 UTC
Permalink
Am 30.01.2013 18:36, schrieb Hinnerk van Bruinehsen:

> I've just installed systemd on one of my systems to give it a test and I
> had similar problems due to the systemd useflag on policykit being
> hardmasked (it also pulled in consolekit because of that).
> Since the errors are very similar you may check your useflags on
> policykit and - if necessary remove the use-mask of systemd for
> policykit.

Let me get that straight:

I have:

# cat profile/package.use.mask

media-sound/pulseaudio -systemd
net-misc/networkmanager -systemd
sys-auth/polkit -systemd
sys-fs/udisks -systemd
sys-power/upower -systemd

because of some older thread or the gentoo wiki for systemd (can't
remember right now).

This gets me:

[I] sys-auth/polkit
Available versions: 0.107-r1 0.110 {examples gtk +introspection
kde nls pam selinux systemd}
Installed versions: 0.110(18:09:30 30.01.2013)(gtk introspection
nls pam systemd -examples -kde -selinux)

while I have USE= ... -consolekit systemd ... in make.conf.

I also get consolekit installed here:

[I] sys-auth/consolekit
Available versions: 0.4.5_p20120320-r1 {acl debug doc pam
policykit selinux test KERNEL="linux"}
Installed versions: 0.4.5_p20120320-r1(18:13:50 30.01.2013)(acl
pam policykit -debug -doc -selinux -test KERNEL="linux")

What exactly do you suggest now?

Thanks!

Stefan
Canek Peláez Valdés
2013-01-30 17:52:39 UTC
Permalink
On Wed, Jan 30, 2013 at 11:48 AM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 18:36, schrieb Hinnerk van Bruinehsen:
>
>> I've just installed systemd on one of my systems to give it a test and I
>> had similar problems due to the systemd useflag on policykit being
>> hardmasked (it also pulled in consolekit because of that).
>> Since the errors are very similar you may check your useflags on
>> policykit and - if necessary remove the use-mask of systemd for
>> policykit.


As Hinnerk said, it could be a PolKit problem, but you said that you
had unmasked the systemd USE flag from PolKit. The new information I
see in this mail is that you have =gnome-base/gnome-session-9999
installed; why do you have a live version? Did you use --autounmask to
install GNOME?

I think that's the problem: gnome-session has no live version in the
tree; therefore you are installing it from the GNOME overlay. The live
version of gnome-session in the GNOME overlay doesn't use a specific
version, tag or branch to checkout, so depending on when you installed
it, it's possible you are running gnome-session-3.7.x.

I would keep gnome-session keyworded, but unmasked; that would force
the install of the 3.6.2 version. Also, if you have more live
versions, I would recommend downgrading them to the last 3.6.x
version. GNOME 3.6 is not masked in Gentoo, just keyworded.

> Let me get that straight:
>
> I have:
>
> # cat profile/package.use.mask
>
> media-sound/pulseaudio -systemd
> net-misc/networkmanager -systemd
> sys-auth/polkit -systemd
> sys-fs/udisks -systemd
> sys-power/upower -systemd
>
> because of some older thread or the gentoo wiki for systemd (can't
> remember right now).
>
> This gets me:
>
> [I] sys-auth/polkit
> Available versions: 0.107-r1 0.110 {examples gtk +introspection
> kde nls pam selinux systemd}
> Installed versions: 0.110(18:09:30 30.01.2013)(gtk introspection
> nls pam systemd -examples -kde -selinux)
>
> while I have USE= ... -consolekit systemd ... in make.conf.
>
> I also get consolekit installed here:
>
> [I] sys-auth/consolekit
> Available versions: 0.4.5_p20120320-r1 {acl debug doc pam
> policykit selinux test KERNEL="linux"}
> Installed versions: 0.4.5_p20120320-r1(18:13:50 30.01.2013)(acl
> pam policykit -debug -doc -selinux -test KERNEL="linux")
>
> What exactly do you suggest now?

If you have -consolekit, why it's still installed? What is pulling it
into your system? Can you do a "equery depends consolekit"?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-30 17:56:11 UTC
Permalink
Am 30.01.2013 18:52, schrieb Canek Peláez Valdés:
> On Wed, Jan 30, 2013 at 11:48 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>> Am 30.01.2013 18:36, schrieb Hinnerk van Bruinehsen:
>>
>>> I've just installed systemd on one of my systems to give it a test and I
>>> had similar problems due to the systemd useflag on policykit being
>>> hardmasked (it also pulled in consolekit because of that).
>>> Since the errors are very similar you may check your useflags on
>>> policykit and - if necessary remove the use-mask of systemd for
>>> policykit.
>
>
> As Hinnerk said, it could be a PolKit problem, but you said that you
> had unmasked the systemd USE flag from PolKit. The new information I
> see in this mail is that you have =gnome-base/gnome-session-9999
> installed; why do you have a live version? Did you use --autounmask to
> install GNOME?
>
> I think that's the problem: gnome-session has no live version in the
> tree; therefore you are installing it from the GNOME overlay. The live
> version of gnome-session in the GNOME overlay doesn't use a specific
> version, tag or branch to checkout, so depending on when you installed
> it, it's possible you are running gnome-session-3.7.x.
>
> I would keep gnome-session keyworded, but unmasked; that would force
> the install of the 3.6.2 version. Also, if you have more live
> versions, I would recommend downgrading them to the last 3.6.x
> version. GNOME 3.6 is not masked in Gentoo, just keyworded.

I have

[I] gnome-base/gnome-session
Available versions: 2.32.1-r3 (~)3.4.2.1 (~)3.6.2 (~)3.6.2-r1
**9999[1] {debug doc gconf ipv6 systemd ELIBC="FreeBSD"}
Installed versions: 3.6.2-r1(17:40:19 30.01.2013)(ipv6 systemd
-debug -doc -gconf ELIBC="-FreeBSD")

-> so 3.6.2-r1 here ....

> If you have -consolekit, why it's still installed? What is pulling it
> into your system? Can you do a "equery depends consolekit"?

sure, did that already .... looks strange, right?

* These packages depend on sys-auth/consolekit:
gnome-base/gdm-3.6.2 (consolekit ? sys-auth/consolekit[pam])
gnome-base/gnome-control-center-3.6.3-r1 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-session-3.6.2-r1 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-settings-daemon-3.6.4 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-shell-3.6.2-r1 (!systemd ? sys-auth/consolekit)
net-misc/networkmanager-0.9.6.4-r1 (consolekit ? sys-auth/consolekit)
net-wireless/bluez-4.101-r5 (consolekit ? sys-auth/consolekit)
sys-apps/accountsservice-0.6.30 (!systemd ? sys-auth/consolekit)
sys-auth/pambase-20120417-r1 (consolekit ?
>=sys-auth/consolekit-0.4.5_p2012[pam])
sys-auth/polkit-0.110 (!systemd ?
>=sys-auth/consolekit-0.4.5_p2012[policykit])
x11-apps/xdm-1.1.11-r1 (consolekit ? sys-auth/consolekit)
Canek Peláez Valdés
2013-01-30 18:04:52 UTC
Permalink
On Wed, Jan 30, 2013 at 11:56 AM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 18:52, schrieb Canek Peláez Valdés:
>> On Wed, Jan 30, 2013 at 11:48 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>>> Am 30.01.2013 18:36, schrieb Hinnerk van Bruinehsen:
>>>
>>>> I've just installed systemd on one of my systems to give it a test and I
>>>> had similar problems due to the systemd useflag on policykit being
>>>> hardmasked (it also pulled in consolekit because of that).
>>>> Since the errors are very similar you may check your useflags on
>>>> policykit and - if necessary remove the use-mask of systemd for
>>>> policykit.
>>
>>
>> As Hinnerk said, it could be a PolKit problem, but you said that you
>> had unmasked the systemd USE flag from PolKit. The new information I
>> see in this mail is that you have =gnome-base/gnome-session-9999
>> installed; why do you have a live version? Did you use --autounmask to
>> install GNOME?
>>
>> I think that's the problem: gnome-session has no live version in the
>> tree; therefore you are installing it from the GNOME overlay. The live
>> version of gnome-session in the GNOME overlay doesn't use a specific
>> version, tag or branch to checkout, so depending on when you installed
>> it, it's possible you are running gnome-session-3.7.x.
>>
>> I would keep gnome-session keyworded, but unmasked; that would force
>> the install of the 3.6.2 version. Also, if you have more live
>> versions, I would recommend downgrading them to the last 3.6.x
>> version. GNOME 3.6 is not masked in Gentoo, just keyworded.
>
> I have
>
> [I] gnome-base/gnome-session
> Available versions: 2.32.1-r3 (~)3.4.2.1 (~)3.6.2 (~)3.6.2-r1
> **9999[1] {debug doc gconf ipv6 systemd ELIBC="FreeBSD"}
> Installed versions: 3.6.2-r1(17:40:19 30.01.2013)(ipv6 systemd
> -debug -doc -gconf ELIBC="-FreeBSD")
>
> -> so 3.6.2-r1 here ....

My mistake; I saw the asterisks and I thought that was the one installed.

>> If you have -consolekit, why it's still installed? What is pulling it
>> into your system? Can you do a "equery depends consolekit"?
>
> sure, did that already .... looks strange, right?
>
> * These packages depend on sys-auth/consolekit:
> gnome-base/gdm-3.6.2 (consolekit ? sys-auth/consolekit[pam])
> gnome-base/gnome-control-center-3.6.3-r1 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-session-3.6.2-r1 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-settings-daemon-3.6.4 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-shell-3.6.2-r1 (!systemd ? sys-auth/consolekit)
> net-misc/networkmanager-0.9.6.4-r1 (consolekit ? sys-auth/consolekit)
> net-wireless/bluez-4.101-r5 (consolekit ? sys-auth/consolekit)
> sys-apps/accountsservice-0.6.30 (!systemd ? sys-auth/consolekit)
> sys-auth/pambase-20120417-r1 (consolekit ?
>>=sys-auth/consolekit-0.4.5_p2012[pam])
> sys-auth/polkit-0.110 (!systemd ?
>>=sys-auth/consolekit-0.4.5_p2012[policykit])
> x11-apps/xdm-1.1.11-r1 (consolekit ? sys-auth/consolekit)

OK, did you run "emerge --depclean"? All of those packages use
consolekit or systemd, but not both; if after --depclean'ing
consolekit keeps lurking, could you uninstall it and mask it, followed
by a "emerge -uDNvp world"? That would tell us which package is
pulling it.

Also, remeber that you can set USE flags not only in
/etc/portage/make.conf, but also in /etc/portage/package.use (and I
think the latter takes precedence).

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-30 18:13:35 UTC
Permalink
Am 30.01.2013 19:04, schrieb Canek Peláez Valdés:

> OK, did you run "emerge --depclean"? All of those packages use
> consolekit or systemd, but not both; if after --depclean'ing
> consolekit keeps lurking, could you uninstall it and mask it, followed
> by a "emerge -uDNvp world"? That would tell us which package is
> pulling it.
>
> Also, remeber that you can set USE flags not only in
> /etc/portage/make.conf, but also in /etc/portage/package.use (and I
> think the latter takes precedence).

Good suggestion ... had a line in /etc/portage/package.use for pambase

... this one pulled in consolekit!

--depclean done now, I revdep-rebuild now and will then check back with
gdm.service.

just a minute ...

Thanks so far!

Stefan
Stefan G. Weichinger
2013-01-30 18:32:48 UTC
Permalink
Am 30.01.2013 19:13, schrieb Stefan G. Weichinger:

> --depclean done now, I revdep-rebuild now and will then check back with
> gdm.service.
>
> just a minute ...

Still no luck.

consolekit is gone now ....

# equery d consolekit

* These packages depend on consolekit:
gnome-base/gdm-3.6.2 (consolekit ? sys-auth/consolekit[pam])
gnome-base/gnome-control-center-3.6.3-r1 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-session-3.6.2-r1 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-settings-daemon-3.6.4 (!systemd ? sys-auth/consolekit)
gnome-base/gnome-shell-3.6.2-r1 (!systemd ? sys-auth/consolekit)
net-misc/networkmanager-0.9.6.4-r1 (consolekit ? sys-auth/consolekit)
net-wireless/bluez-4.101-r5 (consolekit ? sys-auth/consolekit)
sys-apps/accountsservice-0.6.30 (!systemd ? sys-auth/consolekit)
sys-auth/pambase-20120417-r1 (consolekit ?
>=sys-auth/consolekit-0.4.5_p2012[pam])
sys-auth/polkit-0.110 (!systemd ?
>=sys-auth/consolekit-0.4.5_p2012[policykit])
x11-apps/xdm-1.1.11-r1 (consolekit ? sys-auth/consolekit)


I rebuilt them ALL right now ....

As you easily can see I am confused ->

# grep systemd /etc/portage/package.use
#gnome-base/gdm -systemd
#gnome-base/gnome-session -systemd
#gnome-base/gnome-shell -systemd
#sys-auth/polkit -systemd

# grep systemd /etc/portage/profile/package.use.mask
media-sound/pulseaudio -systemd
net-misc/networkmanager -systemd
sys-auth/polkit -systemd
sys-fs/udisks -systemd
sys-power/upower -systemd

What to keep, what to set, please?

Thanks, Stefan


ah, btw, logs:


Jan 30 19:22:53 hiro polkitd[5014]: Loading rules from directory
/etc/polkit-1/rules.d
Jan 30 19:22:53 hiro polkitd[5014]: Loading rules from directory
/usr/share/polkit-1/rules.d
Jan 30 19:22:53 hiro polkitd[5014]: Finished loading, compiling and
executing 3 rules
Jan 30 19:23:13 hiro gdm-simple-slave[4991]: GLib-GObject-CRITICAL:
g_object_ref: assertion `object->ref_count > 0' failed
Jan 30 19:23:13 hiro gnome-session[5027]: Gdk-WARNING: gnome-session:
Fatal IO error 104 (Die Verbindung wurde vom Kommunikationspartner
zurückgesetzt) on X server :0.
Jan 30 19:23:13 hiro gdm-simple-slave[4991]: GLib-GObject-CRITICAL:
g_object_unref: assertion `object->ref_count > 0' failed
Jan 30 19:23:13 hiro polkitd[5014]: Unregistered Authentication Agent
for unix-session:c4 (system bus name :1.26, object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8')
(disconnected from bus)
Jan 30 19:23:13 hiro dbus[4616]: [system] Rejected send message, 1
matched rules; type="method_call", sender=":1.35" (uid=0 pid=27427
comm="/usr/bin/gdm --nodaemon ")
interface="org.freedesktop.DBus.Properties" member="GetAll" error
name="(unset)" requested_reply="0" destination=":1.36" (uid=0 pid=27469
comm="/usr/libexec/gdm-simple-slave --display-id /org/gn")
Jan 30 19:23:13 hiro acpid: client 4994[0:0] has disconnected
Jan 30 19:23:13 hiro acpid: client connected from 27484[0:0]
Jan 30 19:23:13 hiro acpid: 1 client rule loaded
Jan 30 19:23:14 hiro acpid: client connected from 27484[0:0]
Jan 30 19:23:14 hiro acpid: 1 client rule loaded
Jan 30 19:23:14 hiro gdm-simple-slave[27469]: WARNING: Failed to give
slave programs access to the display. Trying to proceed.
Jan 30 19:23:14 hiro gdm-launch-environment][27817]:
pam_unix(gdm-launch-environment:session): session opened for user gdm by
(unknown)(uid=0)
Jan 30 19:23:15 hiro polkitd[5014]: Registered Authentication Agent for
unix-session:c10 (system bus name :1.44 [gnome-shell --mode=gdm], object
path /org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8')
Jan 30 19:23:23 hiro gdm-password][28289]:
pam_unix(gdm-password:session): session opened for user sgw by
(unknown)(uid=0)


....
Canek Peláez Valdés
2013-01-30 18:47:14 UTC
Permalink
On Wed, Jan 30, 2013 at 12:32 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 19:13, schrieb Stefan G. Weichinger:
>
>> --depclean done now, I revdep-rebuild now and will then check back with
>> gdm.service.
>>
>> just a minute ...
>
> Still no luck.
>
> consolekit is gone now ....
>
> # equery d consolekit
>
> * These packages depend on consolekit:
> gnome-base/gdm-3.6.2 (consolekit ? sys-auth/consolekit[pam])
> gnome-base/gnome-control-center-3.6.3-r1 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-session-3.6.2-r1 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-settings-daemon-3.6.4 (!systemd ? sys-auth/consolekit)
> gnome-base/gnome-shell-3.6.2-r1 (!systemd ? sys-auth/consolekit)
> net-misc/networkmanager-0.9.6.4-r1 (consolekit ? sys-auth/consolekit)
> net-wireless/bluez-4.101-r5 (consolekit ? sys-auth/consolekit)
> sys-apps/accountsservice-0.6.30 (!systemd ? sys-auth/consolekit)
> sys-auth/pambase-20120417-r1 (consolekit ?
>>=sys-auth/consolekit-0.4.5_p2012[pam])
> sys-auth/polkit-0.110 (!systemd ?
>>=sys-auth/consolekit-0.4.5_p2012[policykit])
> x11-apps/xdm-1.1.11-r1 (consolekit ? sys-auth/consolekit)
>
>
> I rebuilt them ALL right now ....
>
> As you easily can see I am confused ->
>
> # grep systemd /etc/portage/package.use
> #gnome-base/gdm -systemd
> #gnome-base/gnome-session -systemd
> #gnome-base/gnome-shell -systemd
> #sys-auth/polkit -systemd
>
> # grep systemd /etc/portage/profile/package.use.mask
> media-sound/pulseaudio -systemd
> net-misc/networkmanager -systemd
> sys-auth/polkit -systemd
> sys-fs/udisks -systemd
> sys-power/upower -systemd
>
> What to keep, what to set, please?

Everything looks fine. However, you did rebuild pambase, did you
rebooted your computer? Also, do you have the following line in
/etc/pam.d/system-login?

-session optional pam_systemd.so

Also, before you reboot, could you edit /etc/gdm/custom.conf and add
the following to the [debug] section?

Enable=true

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-30 18:51:12 UTC
Permalink
Am 30.01.2013 19:47, schrieb Canek Peláez Valdés:

> Everything looks fine. However, you did rebuild pambase, did you
> rebooted your computer?

Yes, I think so, but I am not sure right now.
Will do now.

Also, do you have the following line in
> /etc/pam.d/system-login?
>
> -session optional pam_systemd.so

More than this:

# grep systemd /etc/pam.d/*
/etc/pam.d/system-auth:-session optional pam_systemd.so
/etc/pam.d/system-login:-session optional pam_systemd.so
/etc/pam.d/system-services:-session optional pam_systemd.so

> Also, before you reboot, could you edit /etc/gdm/custom.conf and add
> the following to the [debug] section?
>
> Enable=true

Yep. Done.

Rebooting as soon as some other emerge-work has finished ...

Stefan
Canek Peláez Valdés
2013-01-30 18:58:54 UTC
Permalink
On Wed, Jan 30, 2013 at 12:51 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 19:47, schrieb Canek Peláez Valdés:
>
>> Everything looks fine. However, you did rebuild pambase, did you
>> rebooted your computer?
>
> Yes, I think so, but I am not sure right now.
> Will do now.
>
> Also, do you have the following line in
>> /etc/pam.d/system-login?
>>
>> -session optional pam_systemd.so
>
> More than this:
>
> # grep systemd /etc/pam.d/*
> /etc/pam.d/system-auth:-session optional pam_systemd.so
> /etc/pam.d/system-login:-session optional pam_systemd.so
> /etc/pam.d/system-services:-session optional pam_systemd.so

I don't think it matters, but in my laptop and desktop (both running
GNOME 3.6), I only have the line in system-login.

Let's see what the logs have to say.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-30 19:22:30 UTC
Permalink
Am 30.01.2013 19:58, schrieb Canek Peláez Valdés:

> Let's see what the logs have to say.

Failed again.

Put up the failed session of gdm here to keep the list uncluttered:

https://dl.dropbox.com/u/24516209/gdm_problems.txt

I added the line "sgw FAILED ..." in there to mark the failed effort to
login.

Stefan
Stefan G. Weichinger
2013-01-30 20:22:59 UTC
Permalink
Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:

> Put up the failed session of gdm here to keep the list uncluttered:
>
> https://dl.dropbox.com/u/24516209/gdm_problems.txt

I am away from this system for now ... more tomorrow, thanks so far.

Stefan
Alecks Gates
2013-01-30 21:14:42 UTC
Permalink
On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:
>
>> Put up the failed session of gdm here to keep the list uncluttered:
>>
>> https://dl.dropbox.com/u/24516209/gdm_problems.txt
>
> I am away from this system for now ... more tomorrow, thanks so far.
>
> Stefan
>
>
>

I switched to systemd not too long ago and I have the same issue as
well, at least it sounds the same -- Basically, I get a hanging GDM
after typing my password and logging in. I'm certainly no expert, but
I've enjoyed the rest of systemd so I've stuck with it and just use
startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.
Canek Peláez Valdés
2013-01-31 05:35:42 UTC
Permalink
On Wed, Jan 30, 2013 at 3:14 PM, Alecks Gates <***@gmail.com> wrote:
> On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
>> Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:
>>
>>> Put up the failed session of gdm here to keep the list uncluttered:
>>>
>>> https://dl.dropbox.com/u/24516209/gdm_problems.txt
>>
>> I am away from this system for now ... more tomorrow, thanks so far.
>>
>> Stefan
>>
>>
>>
>
> I switched to systemd not too long ago and I have the same issue as
> well, at least it sounds the same -- Basically, I get a hanging GDM
> after typing my password and logging in. I'm certainly no expert, but
> I've enjoyed the rest of systemd so I've stuck with it and just use
> startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.

Alecks, your error is different, and one similar to one I had before:

https://bugs.gentoo.org/show_bug.cgi?id=363061

What does systemctl status accounts-daemon.service says? Actually,
could tell me what services are in red when you run "systemctl --full
--all"?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Canek Peláez Valdés
2013-01-31 05:48:29 UTC
Permalink
On Wed, Jan 30, 2013 at 11:35 PM, Canek Peláez Valdés <***@gmail.com> wrote:
> On Wed, Jan 30, 2013 at 3:14 PM, Alecks Gates <***@gmail.com> wrote:
>> On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
>>> Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:
>>>
>>>> Put up the failed session of gdm here to keep the list uncluttered:
>>>>
>>>> https://dl.dropbox.com/u/24516209/gdm_problems.txt
>>>
>>> I am away from this system for now ... more tomorrow, thanks so far.
>>>
>>> Stefan
>>>
>>>
>>>
>>
>> I switched to systemd not too long ago and I have the same issue as
>> well, at least it sounds the same -- Basically, I get a hanging GDM
>> after typing my password and logging in. I'm certainly no expert, but
>> I've enjoyed the rest of systemd so I've stuck with it and just use
>> startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.
>
> Alecks, your error is different, and one similar to one I had before:
>
> https://bugs.gentoo.org/show_bug.cgi?id=363061
>
> What does systemctl status accounts-daemon.service says? Actually,
> could tell me what services are in red when you run "systemctl --full
> --all"?

By the way, Alecks, maybe you could try to delete /var/lib/gdm (the
gdm user $HOME), and emerge again gdm? Also could you check that all
the files under /var/lib/gdm/ have gdm:gdm ownership?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Alecks Gates
2013-01-31 23:17:13 UTC
Permalink
On Wed, Jan 30, 2013 at 11:48 PM, Canek Peláez Valdés <***@gmail.com> wrote:
> On Wed, Jan 30, 2013 at 11:35 PM, Canek Peláez Valdés <***@gmail.com> wrote:
>> On Wed, Jan 30, 2013 at 3:14 PM, Alecks Gates <***@gmail.com> wrote:
[snip]
>>>
>>> I switched to systemd not too long ago and I have the same issue as
>>> well, at least it sounds the same -- Basically, I get a hanging GDM
>>> after typing my password and logging in. I'm certainly no expert, but
>>> I've enjoyed the rest of systemd so I've stuck with it and just use
>>> startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.
>>
>> Alecks, your error is different, and one similar to one I had before:
>>
>> https://bugs.gentoo.org/show_bug.cgi?id=363061
>>
>> What does systemctl status accounts-daemon.service says? Actually,
>> could tell me what services are in red when you run "systemctl --full
>> --all"?
>
> By the way, Alecks, maybe you could try to delete /var/lib/gdm (the
> gdm user $HOME), and emerge again gdm? Also could you check that all
> the files under /var/lib/gdm/ have gdm:gdm ownership?
>
> Regards.
> --
> Canek Peláez Valdés
> Posgrado en Ciencia e Ingeniería de la Computación
> Universidad Nacional Autónoma de México
>

Deleted and remerged, no dice.

$ ls -al /var/lib/gdm
total 4
drwxr-xr-x 1 gdm gdm 112 Jan 31 17:02 .
drwxr-xr-x 1 root root 546 Jan 31 16:57 ..
drwxr-xr-x 1 gdm gdm 56 Jan 31 17:00 .config
drwx------ 1 gdm gdm 22 Jan 31 16:59 .dbus
-rw------- 1 gdm gdm 628 Jan 31 17:02 .ICEauthority
-rw-r--r-- 1 root root 0 Jan 31 16:57 .keep_gnome-base_gdm-0
drwxr-xr-x 1 gdm gdm 10 Jan 31 16:57 .local
drwx------ 1 gdm gdm 14 Jan 31 16:59 .nv
Alecks Gates
2013-01-31 23:27:18 UTC
Permalink
On Wed, Jan 30, 2013 at 11:35 PM, Canek Peláez Valdés <***@gmail.com> wrote:
> On Wed, Jan 30, 2013 at 3:14 PM, Alecks Gates <***@gmail.com> wrote:
>> On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
[snip]
>>
>> I switched to systemd not too long ago and I have the same issue as
>> well, at least it sounds the same -- Basically, I get a hanging GDM
>> after typing my password and logging in. I'm certainly no expert, but
>> I've enjoyed the rest of systemd so I've stuck with it and just use
>> startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.
>
> Alecks, your error is different, and one similar to one I had before:
>
> https://bugs.gentoo.org/show_bug.cgi?id=363061
>
> What does systemctl status accounts-daemon.service says? Actually,
> could tell me what services are in red when you run "systemctl --full
> --all"?
>
> Regards.
> --
> Canek Peláez Valdés
> Posgrado en Ciencia e Ingeniería de la Computación
> Universidad Nacional Autónoma de México
>

$ systemctl status accounts-daemon.service
accounts-daemon.service - Accounts Service
Loaded: loaded (/usr/lib64/systemd/system/accounts-daemon.service; disabled)
Active: active (running) since Thu 2013-01-31 17:02:33 CST; 16min ago
Main PID: 3326 (accounts-daemon)
CGroup: name=systemd:/system/accounts-daemon.service
└─3326 /usr/libexec/accounts-daemon

$ systemctl --full --all | grep error
auditd.service error inactive dead auditd.service
plymouth-quit-wait.service error inactive dead
plymouth-quit-wait.service
plymouth-start.service error inactive dead plymouth-start.service
syslog.service error inactive dead syslog.service


A couple days ago (after reading your email from another topic) I
noticed plymouth services do not exist on my machine, and checked
where it's supposed to come from:
$ e-file plymouth-start.service
[I] sys-apps/systemd
Available Versions: 44-r1 44
Last Installed Ver: 197-r1(Mon 28 Jan 2013 03:58:41 PM CST)
Homepage: http://www.freedesktop.org/wiki/Software/systemd
Description: System and service manager for Linux
Matched Files: /usr/lib/systemd/system/sysinit.target.wants/plymouth-start.service;
/usr/lib/systemd/system/plymouth-start.service;
$ e-file plymouth-quit-wait.service
[I] sys-apps/systemd
Available Versions: 44 44-r1
Last Installed Ver: 197-r1(Mon 28 Jan 2013 03:58:41 PM CST)
Homepage: http://www.freedesktop.org/wiki/Software/systemd
Description: System and service manager for Linux
Matched Files: /usr/lib/systemd/system/multi-user.target.wants/plymouth-quit-wait.service;
/usr/lib/systemd/system/plymouth-quit-wait.service;

And auditd.service isn't found in e-file at all. Normally I'm not
surprised by a lack of .service files, as it's not a huge issue[1],
but if this one's so important, where is it?

Canek, I'm getting the feeling your systemd install has matured over
the years, at least with regard to unit files.


[1] Unit files are surprisingly easy for me to create -- I always
found a barrier to entry with init scripts.

Alecks
Canek Peláez Valdés
2013-02-01 04:27:42 UTC
Permalink
On Thu, Jan 31, 2013 at 5:27 PM, Alecks Gates <***@gmail.com> wrote:
> On Wed, Jan 30, 2013 at 11:35 PM, Canek Peláez Valdés <***@gmail.com> wrote:
>> On Wed, Jan 30, 2013 at 3:14 PM, Alecks Gates <***@gmail.com> wrote:
>>> On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> [snip]
>>>
>>> I switched to systemd not too long ago and I have the same issue as
>>> well, at least it sounds the same -- Basically, I get a hanging GDM
>>> after typing my password and logging in. I'm certainly no expert, but
>>> I've enjoyed the rest of systemd so I've stuck with it and just use
>>> startx to boot into Gnome3. I'll attatch some logs from /var/log/gdm.
>>
>> Alecks, your error is different, and one similar to one I had before:
>>
>> https://bugs.gentoo.org/show_bug.cgi?id=363061
>>
>> What does systemctl status accounts-daemon.service says? Actually,
>> could tell me what services are in red when you run "systemctl --full
>> --all"?
>>
>> Regards.
>> --
>> Canek Peláez Valdés
>> Posgrado en Ciencia e Ingeniería de la Computación
>> Universidad Nacional Autónoma de México
>>
>
> $ systemctl status accounts-daemon.service
> accounts-daemon.service - Accounts Service
> Loaded: loaded (/usr/lib64/systemd/system/accounts-daemon.service; disabled)
> Active: active (running) since Thu 2013-01-31 17:02:33 CST; 16min ago
> Main PID: 3326 (accounts-daemon)
> CGroup: name=systemd:/system/accounts-daemon.service
> └─3326 /usr/libexec/accounts-daemon
>
> $ systemctl --full --all | grep error
> auditd.service error inactive dead auditd.service
> plymouth-quit-wait.service error inactive dead
> plymouth-quit-wait.service
> plymouth-start.service error inactive dead plymouth-start.service
> syslog.service error inactive dead syslog.service
>
>
> A couple days ago (after reading your email from another topic) I
> noticed plymouth services do not exist on my machine, and checked
> where it's supposed to come from:
> $ e-file plymouth-start.service
> [I] sys-apps/systemd
> Available Versions: 44-r1 44
> Last Installed Ver: 197-r1(Mon 28 Jan 2013 03:58:41 PM CST)
> Homepage: http://www.freedesktop.org/wiki/Software/systemd
> Description: System and service manager for Linux
> Matched Files: /usr/lib/systemd/system/sysinit.target.wants/plymouth-start.service;
> /usr/lib/systemd/system/plymouth-start.service;
> $ e-file plymouth-quit-wait.service
> [I] sys-apps/systemd
> Available Versions: 44 44-r1
> Last Installed Ver: 197-r1(Mon 28 Jan 2013 03:58:41 PM CST)
> Homepage: http://www.freedesktop.org/wiki/Software/systemd
> Description: System and service manager for Linux
> Matched Files: /usr/lib/systemd/system/multi-user.target.wants/plymouth-quit-wait.service;
> /usr/lib/systemd/system/plymouth-quit-wait.service;
>
> And auditd.service isn't found in e-file at all. Normally I'm not
> surprised by a lack of .service files, as it's not a huge issue[1],
> but if this one's so important, where is it?
>
> Canek, I'm getting the feeling your systemd install has matured over
> the years, at least with regard to unit files.

Not really; at some point yes, but I believe I run a pretty much
out-of-the-box systemd. As I explained to Stefan a couple of mails
before, you can mask the unit files you don't have, by linking them in
/etc/systemd/system. That's all I do, and only with syslog and
rc-local:

# ls -l /etc/systemd/system/|grep /dev/null
lrwxrwxrwx 1 root root 9 Aug 16 13:51 rc-local.service -> /dev/null
lrwxrwxrwx 1 root root 9 Aug 16 13:49 syslog.service -> /dev/null

Besides that I have a vixie-cron.service, and that's all the
customization I do. Everything else is as defined by the respective
upstream developers. In my media center I have a couple more of custom
unit files.

> [1] Unit files are surprisingly easy for me to create -- I always
> found a barrier to entry with init scripts.

I didn't have any problem writing init scripts, but I agree unit files
are really easy.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-31 09:13:40 UTC
Permalink
Am 2013-01-30 21:22, schrieb Stefan G. Weichinger:
> Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:
>
>> Put up the failed session of gdm here to keep the list uncluttered:
>>
>> https://dl.dropbox.com/u/24516209/gdm_problems.txt
>
> I am away from this system for now ... more tomorrow, thanks so far.


Took that as a opportunity to put up stuff on my github:

https://github.com/stefangweichinger/systemd-services

Stefan
Canek Peláez Valdés
2013-01-31 18:06:50 UTC
Permalink
On Wed, Jan 30, 2013 at 2:22 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 30.01.2013 20:22, schrieb Stefan G. Weichinger:
>
>> Put up the failed session of gdm here to keep the list uncluttered:
>>
>> https://dl.dropbox.com/u/24516209/gdm_problems.txt
>
> I am away from this system for now ... more tomorrow, thanks so far.

I booted my laptop with [debug]Enabled=true to compare the logs, and
your problem seems to be permissions related. In my logs, if I grep
"Found x" I get:

Jan 30 18:09:46 acero gdm-password][543]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session '1': :0
Jan 30 18:09:46 acero gdm-password][543]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session '2': :0

In your case, is this:

Jan 30 20:14:05 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:05 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:12 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c12': :0
Jan 30 20:14:12 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c13': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c14': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c12': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c13': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c14': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c12': :0
Jan 30 20:14:16 hiro gdm-password][6188]: AccountsService-DEBUG(+):
ActUserManager: Found x11 display of session 'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:16 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0
Jan 30 20:14:18 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c12': :0
Jan 30 20:14:18 hiro gdm-launch-environment][6116]:
AccountsService-DEBUG(+): ActUserManager: Found x11 display of session
'c11': :0

I tries in the consoles c11, c12, and then on the c13 and 14. I don't
know why, but it would seem that gdm-password fails to authenticate
you. Do you have anything in /etc/polkit-1/rules.d/? I only have
50-default.rules, which Gentoo adds. Also check
/usr/share/polkit-1/rules.d.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-31 18:17:20 UTC
Permalink
Am 31.01.2013 19:06, schrieb Canek Peláez Valdés:

> I tries in the consoles c11, c12, and then on the c13 and 14.

Yes, I noticed that trying-around as well ...

> I don't
> know why, but it would seem that gdm-password fails to authenticate
> you.

Just changed my password to something simple without german umlauts or
special characters to rule out charset-topics. No change.

> Do you have anything in /etc/polkit-1/rules.d/? I only have
> 50-default.rules, which Gentoo adds.

Same here.

> Also check
> /usr/share/polkit-1/rules.d.

These 2 files:

# cat 01-org.freedesktop.NetworkManager.settings.modify.system.rules
// Let users in plugdev group modify NetworkManager
polkit.addRule(function(action, subject) {
if (action.id ==
"org.freedesktop.NetworkManager.settings.modify.system" &&
subject.isInGroup("plugdev") && subject.active) {
return "yes";
}
});

# cat gnome-control-center.rules
polkit.addRule(function(action, subject) {
if (action.id == "org.freedesktop.hostname1.set-static-hostname" &&
subject.local &&
subject.active &&
subject.isInGroup ("wheel")) {
return polkit.Result.YES;
}
});

-

My user "sgw" is members of both mentioned groups, btw:

# getent group wheel
wheel:x:10:root,sgw

# getent group plugdev
plugdev:x:443:haldaemon,sgw

S
Canek Peláez Valdés
2013-01-31 18:26:49 UTC
Permalink
On Thu, Jan 31, 2013 at 12:17 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 31.01.2013 19:06, schrieb Canek Peláez Valdés:
>
>> I tries in the consoles c11, c12, and then on the c13 and 14.
>
> Yes, I noticed that trying-around as well ...
>
>> I don't
>> know why, but it would seem that gdm-password fails to authenticate
>> you.
>
> Just changed my password to something simple without german umlauts or
> special characters to rule out charset-topics. No change.
>
>> Do you have anything in /etc/polkit-1/rules.d/? I only have
>> 50-default.rules, which Gentoo adds.
>
> Same here.
>
>> Also check
>> /usr/share/polkit-1/rules.d.
>
> These 2 files:
>
> # cat 01-org.freedesktop.NetworkManager.settings.modify.system.rules
> // Let users in plugdev group modify NetworkManager
> polkit.addRule(function(action, subject) {
> if (action.id ==
> "org.freedesktop.NetworkManager.settings.modify.system" &&
> subject.isInGroup("plugdev") && subject.active) {
> return "yes";
> }
> });
>
> # cat gnome-control-center.rules
> polkit.addRule(function(action, subject) {
> if (action.id == "org.freedesktop.hostname1.set-static-hostname" &&
> subject.local &&
> subject.active &&
> subject.isInGroup ("wheel")) {
> return polkit.Result.YES;
> }
> });

Same here.

> My user "sgw" is members of both mentioned groups, btw:
>
> # getent group wheel
> wheel:x:10:root,sgw
>
> # getent group plugdev
> plugdev:x:443:haldaemon,sgw

And I suppose both sgw and gdm are in the video group (the later is
done by the ebuild, if I'm not mistaken). What is the uid and gid of
gdm? Also, did GDM (the same version) worked with OpenRC, or did you
installed systemd and upgraded gdm at the same time?

What does "systemctl --all --full" says, which units are in red? And
lastly, how did you set gdm as your display manager? Do you have:

# ls -l /etc/systemd/system/display-manager.service
lrwxrwxrwx 1 root root 37 Dec 6 00:40
/etc/systemd/system/display-manager.service ->
/usr/lib64/systemd/system/gdm.service

?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-01-31 18:35:49 UTC
Permalink
Am 31.01.2013 19:26, schrieb Canek Peláez Valdés:

> And I suppose both sgw and gdm are in the video group (the later is
> done by the ebuild, if I'm not mistaken).

Yes, they are:

# getent group video
video:x:27:root,mythtv,sgw,gdm

> What is the uid and gid of
> gdm?

# getent passwd gdm
gdm:x:104:446:added by portage for gdm:/var/lib/gdm:/sbin/nologin

> Also, did GDM (the same version) worked with OpenRC, or did you
> installed systemd and upgraded gdm at the same time?

hmm. No upgrade of gdm, but a re-build as it changed USE-flags
(-consolekit systemd).

> What does "systemctl --all --full" says, which units are in red?


# systemctl --all --full | grep erro
auditd.service
error inactive dead
auditd.service

plymouth-quit-wait.service
error inactive dead
plymouth-quit-wait.service

plymouth-start.service
error inactive dead
plymouth-start.service

syslog.service
error inactive dead
syslog.service


# systemctl --all --full | grep fail
gdm.service
loaded failed failed Gnome
Display Manager

I dont't have plymouth or sys-process/audit ... nothing pulled that in.

> And lastly, how did you set gdm as your display manager? Do you have:
>
> # ls -l /etc/systemd/system/display-manager.service
> lrwxrwxrwx 1 root root 37 Dec 6 00:40
> /etc/systemd/system/display-manager.service ->
> /usr/lib64/systemd/system/gdm.service
>
> ?

Right now it links to xdm, but I had it the way you posted and tested that.

When I test, I switch to a text console:

systemctl stop xdm
systemctl start gdm

...

Thanks! Stefan
Canek Peláez Valdés
2013-01-31 18:54:37 UTC
Permalink
On Thu, Jan 31, 2013 at 12:35 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 31.01.2013 19:26, schrieb Canek Peláez Valdés:
>
>> And I suppose both sgw and gdm are in the video group (the later is
>> done by the ebuild, if I'm not mistaken).
>
> Yes, they are:
>
> # getent group video
> video:x:27:root,mythtv,sgw,gdm
>
>> What is the uid and gid of
>> gdm?
>
> # getent passwd gdm
> gdm:x:104:446:added by portage for gdm:/var/lib/gdm:/sbin/nologin
>
>> Also, did GDM (the same version) worked with OpenRC, or did you
>> installed systemd and upgraded gdm at the same time?
>
> hmm. No upgrade of gdm, but a re-build as it changed USE-flags
> (-consolekit systemd).
>
>> What does "systemctl --all --full" says, which units are in red?
>
>
> # systemctl --all --full | grep erro
> auditd.service
> error inactive dead
> auditd.service
>
> plymouth-quit-wait.service
> error inactive dead
> plymouth-quit-wait.service
>
> plymouth-start.service
> error inactive dead
> plymouth-start.service
>
> syslog.service
> error inactive dead
> syslog.service
>
>
> # systemctl --all --full | grep fail
> gdm.service
> loaded failed failed Gnome
> Display Manager
>
> I dont't have plymouth or sys-process/audit ... nothing pulled that in.

sshd.service, ***@.service, systemd-update-utmp-runlevel.service, and
systemd-update-utmp-shutdown.service have auditd.service in their
After= field; several others have plymouth services. After= is just
for ordering of units, is not a requirement; systemd detects that
auditd.service doesn't exists, and it starts the units that have it in
ther After= field anyway. To make a unit depend on another, you need
Require=.

You can mask the services you don't have by creating a soft link to /dev/null:

# ll /etc/systemd/system/rc-local.service
lrwxrwxrwx 1 root root 9 Aug 16 13:51
/etc/systemd/system/rc-local.service -> /dev/null

It cleans up the output of systemctl --full --all.

>> And lastly, how did you set gdm as your display manager? Do you have:
>>
>> # ls -l /etc/systemd/system/display-manager.service
>> lrwxrwxrwx 1 root root 37 Dec 6 00:40
>> /etc/systemd/system/display-manager.service ->
>> /usr/lib64/systemd/system/gdm.service
>>
>> ?
>
> Right now it links to xdm, but I had it the way you posted and tested that.
>
> When I test, I switch to a text console:
>
> systemctl stop xdm
> systemctl start gdm

Well, I have no idea why your gdm is not letting you log in; obviously
it's related to polkit (since it started when you changed from
consolekit to polkit), but nothing in your config seems to differ from
mine. It is not impossible that somehow the configuration files of the
gdm user got messed up when the change happened. I don't know how this
could happen, but as a hail Mary you could delete /var/lib/gdm, and
reemerge it so it gets a clean install.

Also, you have USE=pam for polkit, right? And could you post the
output from "journalctl -b /usr/lib/polkit-1/polkitd"?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-01 18:37:08 UTC
Permalink
Am 2013-01-31 19:54, schrieb Canek Peláez Valdés:

> sshd.service, ***@.service, systemd-update-utmp-runlevel.service, and
> systemd-update-utmp-shutdown.service have auditd.service in their
> After= field; several others have plymouth services. After= is just
> for ordering of units, is not a requirement; systemd detects that
> auditd.service doesn't exists, and it starts the units that have it in
> ther After= field anyway. To make a unit depend on another, you need
> Require=.
>
> You can mask the services you don't have by creating a soft link to /dev/null:
>
> # ll /etc/systemd/system/rc-local.service
> lrwxrwxrwx 1 root root 9 Aug 16 13:51
> /etc/systemd/system/rc-local.service -> /dev/null
>
> It cleans up the output of systemctl --full --all.

Ok, so I don't need auditd or plymouth, right?

> Well, I have no idea why your gdm is not letting you log in; obviously
> it's related to polkit (since it started when you changed from
> consolekit to polkit), but nothing in your config seems to differ from
> mine. It is not impossible that somehow the configuration files of the
> gdm user got messed up when the change happened. I don't know how this
> could happen, but as a hail Mary you could delete /var/lib/gdm, and
> reemerge it so it gets a clean install.

Tried that as well, same problems after .... :-(

> Also, you have USE=pam for polkit, right?

Yes.

[I] sys-auth/polkit
Available versions: 0.107-r1 0.110 {examples gtk +introspection
kde nls pam selinux systemd}
Installed versions: 0.110(19:19:55 30.01.2013)(gtk introspection
nls pam systemd -examples -kde -selinux)

> And could you post the
> output from "journalctl -b /usr/lib/polkit-1/polkitd"?

That is empty!

S
Canek Peláez Valdés
2013-02-01 19:39:53 UTC
Permalink
On Fri, Feb 1, 2013 at 12:37 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 2013-01-31 19:54, schrieb Canek Peláez Valdés:
>
>> sshd.service, ***@.service, systemd-update-utmp-runlevel.service, and
>> systemd-update-utmp-shutdown.service have auditd.service in their
>> After= field; several others have plymouth services. After= is just
>> for ordering of units, is not a requirement; systemd detects that
>> auditd.service doesn't exists, and it starts the units that have it in
>> ther After= field anyway. To make a unit depend on another, you need
>> Require=.
>>
>> You can mask the services you don't have by creating a soft link to /dev/null:
>>
>> # ll /etc/systemd/system/rc-local.service
>> lrwxrwxrwx 1 root root 9 Aug 16 13:51
>> /etc/systemd/system/rc-local.service -> /dev/null
>>
>> It cleans up the output of systemctl --full --all.
>
> Ok, so I don't need auditd or plymouth, right?

No, you don't.

>> Well, I have no idea why your gdm is not letting you log in; obviously
>> it's related to polkit (since it started when you changed from
>> consolekit to polkit), but nothing in your config seems to differ from
>> mine. It is not impossible that somehow the configuration files of the
>> gdm user got messed up when the change happened. I don't know how this
>> could happen, but as a hail Mary you could delete /var/lib/gdm, and
>> reemerge it so it gets a clean install.
>
> Tried that as well, same problems after .... :-(
>
>> Also, you have USE=pam for polkit, right?
>
> Yes.
>
> [I] sys-auth/polkit
> Available versions: 0.107-r1 0.110 {examples gtk +introspection
> kde nls pam selinux systemd}
> Installed versions: 0.110(19:19:55 30.01.2013)(gtk introspection
> nls pam systemd -examples -kde -selinux)
>
>> And could you post the
>> output from "journalctl -b /usr/lib/polkit-1/polkitd"?
>
> That is empty!

That's weird. *At least* it should tell you that it started and
compiled the available rules. OK, let's try to see the problem outside
systemd. First, stop polkit with:

systemctl stop polkit.service

and immediately after start it from the command line directly:

/usr/lib/polkit-1/polkitd --replace

the "immediately" thing is because polkit is dbus started, so it can
be started by systemd if anything asks for its service. Then in
another terminal stop gdm:

systemctl stop gdm.service

and start it from the command line:

/usr/bin/gdm --no-daemon &> gdm.log

Since you still have the Enable=true in the [debug] section of its
config, it will spew quite a lot of info, hence the redirection to a
log file. I'm more interested in polkit's output when you try to log
in, could you send that? By default polkit doesn't log almost
anything, and I'm pretty sure the problem is with polkit refusing gdm
to log you (or create a console for you, or something like that).
Having an empty log is also weird; mine says:

Jan 30 01:19:20 centurion polkitd[1614]: Started polkitd version 0.110
Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
/etc/polkit-1/rules.d
Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
/usr/share/polkit-1/rules.d
Jan 30 01:19:22 centurion polkitd[1614]: Finished loading, compiling
and executing 3 rules
Jan 30 01:19:22 centurion polkitd[1614]: Acquired the name
org.freedesktop.PolicyKit1 on the system bus
Jan 30 01:19:30 centurion polkitd[1614]: Registered Authentication
Agent for unix-session:1 (system bus name :1.30 [gnome-shell
--mode=gdm], object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
Jan 30 01:19:39 centurion polkitd[1614]: Unregistered Authentication
Agent for unix-session:1 (system bus name :1.30, object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
(disconnected from bus)
Jan 30 01:19:55 centurion polkitd[1614]: Registered Authentication
Agent for unix-session:2 (system bus name :1.58
[/usr/bin/gnome-shell], object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-02 10:10:22 UTC
Permalink
Am 2013-02-01 20:39, schrieb Canek Peláez Valdés:

> Having an empty log is also weird; mine says:
>
> Jan 30 01:19:20 centurion polkitd[1614]: Started polkitd version 0.110
> Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
> /etc/polkit-1/rules.d
> Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
> /usr/share/polkit-1/rules.d
> Jan 30 01:19:22 centurion polkitd[1614]: Finished loading, compiling
> and executing 3 rules
> Jan 30 01:19:22 centurion polkitd[1614]: Acquired the name
> org.freedesktop.PolicyKit1 on the system bus
> Jan 30 01:19:30 centurion polkitd[1614]: Registered Authentication
> Agent for unix-session:1 (system bus name :1.30 [gnome-shell
> --mode=gdm], object path
> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
> Jan 30 01:19:39 centurion polkitd[1614]: Unregistered Authentication
> Agent for unix-session:1 (system bus name :1.30, object path
> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
> (disconnected from bus)
> Jan 30 01:19:55 centurion polkitd[1614]: Registered Authentication
> Agent for unix-session:2 (system bus name :1.58
> [/usr/bin/gnome-shell], object path
> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)

I removed and re-installed polkit, then followed your suggestions.

My polkit only logs this (yes, I tried to log in and got rejected):

# /usr/lib/polkit-1/polkitd --replace
Successfully changed to user polkitd
10:58:53.739: Loading rules from directory /etc/polkit-1/rules.d
10:58:53.739: Loading rules from directory /usr/share/polkit-1/rules.d
10:58:53.739: Finished loading, compiling and executing 3 rules
Entering main event loop
Connected to the system bus
10:58:53.741: Acquired the name org.freedesktop.PolicyKit1 on the system bus
10:59:25.755: Registered Authentication Agent for unix-session:c6
(system bus name :1.28 [gnome-shell --mode=gdm], object path
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8)

Nothing more.

The gdm.log is empty.

---- thanks for you patience with this issue, btw!

S
Canek Peláez Valdés
2013-02-06 07:19:08 UTC
Permalink
On Sat, Feb 2, 2013 at 4:10 AM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 2013-02-01 20:39, schrieb Canek Peláez Valdés:
>
>> Having an empty log is also weird; mine says:
>>
>> Jan 30 01:19:20 centurion polkitd[1614]: Started polkitd version 0.110
>> Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
>> /etc/polkit-1/rules.d
>> Jan 30 01:19:22 centurion polkitd[1614]: Loading rules from directory
>> /usr/share/polkit-1/rules.d
>> Jan 30 01:19:22 centurion polkitd[1614]: Finished loading, compiling
>> and executing 3 rules
>> Jan 30 01:19:22 centurion polkitd[1614]: Acquired the name
>> org.freedesktop.PolicyKit1 on the system bus
>> Jan 30 01:19:30 centurion polkitd[1614]: Registered Authentication
>> Agent for unix-session:1 (system bus name :1.30 [gnome-shell
>> --mode=gdm], object path
>> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
>> Jan 30 01:19:39 centurion polkitd[1614]: Unregistered Authentication
>> Agent for unix-session:1 (system bus name :1.30, object path
>> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
>> (disconnected from bus)
>> Jan 30 01:19:55 centurion polkitd[1614]: Registered Authentication
>> Agent for unix-session:2 (system bus name :1.58
>> [/usr/bin/gnome-shell], object path
>> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf8)

Sorry for answering so late; long weekend in here.

> I removed and re-installed polkit, then followed your suggestions.
>
> My polkit only logs this (yes, I tried to log in and got rejected):
>
> # /usr/lib/polkit-1/polkitd --replace
> Successfully changed to user polkitd
> 10:58:53.739: Loading rules from directory /etc/polkit-1/rules.d
> 10:58:53.739: Loading rules from directory /usr/share/polkit-1/rules.d
> 10:58:53.739: Finished loading, compiling and executing 3 rules
> Entering main event loop
> Connected to the system bus
> 10:58:53.741: Acquired the name org.freedesktop.PolicyKit1 on the system bus
> 10:59:25.755: Registered Authentication Agent for unix-session:c6
> (system bus name :1.28 [gnome-shell --mode=gdm], object path
> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8)
>
> Nothing more.

I was expecting more from that.

> The gdm.log is empty.

That is weird; if you ran gdm outside of systemd, with the --nodaemon
option, the program should print everything to stdout and/or stderr,
and then &> should capture it.

I'm running out of ideas. What does the file /etc/pam.d/gdm-password contains?

> ---- thanks for you patience with this issue, btw!

That's what the list is for,

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-06 11:58:12 UTC
Permalink
Am 06.02.2013 08:19, schrieb Canek Peláez Valdés:

> Sorry for answering so late; long weekend in here.

No problem ... I had other issues here and so far it is just OK to use
xdm.service instead.

btw. even with xdm there is a pretty slow startup of gnome (time between
hitting enter after the password and a working gnome-shell) ...

Over the last days I pretty much rebuilt most of that machine ...

>> 10:59:25.755: Registered Authentication Agent for unix-session:c6
>> (system bus name :1.28 [gnome-shell --mode=gdm], object path
>> /org/freedesktop/PolicyKit1/AuthenticationAgent, locale de_DE.UTF-8)
>>
>> Nothing more.
>
> I was expecting more from that.
>
>> The gdm.log is empty.
>
> That is weird; if you ran gdm outside of systemd, with the --nodaemon
> option, the program should print everything to stdout and/or stderr,
> and then &> should capture it.

I will retry this asap ... but not right now ... maybe later this day.

> I'm running out of ideas. What does the file /etc/pam.d/gdm-password contains?

# cat gdm-password
#%PAM-1.0
auth optional pam_env.so
auth include system-local-login
account include system-local-login
password include system-local-login
session include system-local-login


Stefan
Stefan G. Weichinger
2013-02-06 15:56:00 UTC
Permalink
Am 06.02.2013 12:58, schrieb Stefan G. Weichinger:

> I will retry this asap ... but not right now ... maybe later this day.

did not work, no real different output.

In /var/log/messages I have:

Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUser: session not found: c15
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed: c14
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed for user stef
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUser: session not found: c14
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed: c13
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed for user sgw
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUser: session not found: c13
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed: c12
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUserManager: Session removed for user sgw
Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
AccountsService-DEBUG(+): ActUser: session not found: c12


which lead me to this issue:

https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=861440

Looks very similar, but I don't know yet how to solve it. But it might
be the right direction as I use "su" within my terminator-sessions very
often.

digging further.
Canek Peláez Valdés
2013-02-06 17:18:10 UTC
Permalink
On Wed, Feb 6, 2013 at 9:56 AM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 06.02.2013 12:58, schrieb Stefan G. Weichinger:
>
>> I will retry this asap ... but not right now ... maybe later this day.
>
> did not work, no real different output.
>
> In /var/log/messages I have:
>
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUser: session not found: c15
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed: c14
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed for user stef
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUser: session not found: c14
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed: c13
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed for user sgw
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUser: session not found: c13
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed: c12
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUserManager: Session removed for user sgw
> Feb 6 16:34:15 hiro gdm-launch-environment][24620]:
> AccountsService-DEBUG(+): ActUser: session not found: c12
>
>
> which lead me to this issue:
>
> https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=861440
>
> Looks very similar, but I don't know yet how to solve it. But it might
> be the right direction as I use "su" within my terminator-sessions very
> often.
>
> digging further.

What's the difference between the users stef and sgw? What do you mean
by "I use "su" within my terminator-sessions very often"?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-06 17:55:10 UTC
Permalink
Am 06.02.2013 18:18, schrieb Canek Peláez Valdés:

> What's the difference between the users stef and sgw?

sgw is my everyday-user with dozens of dot-dirs cumulated over years. It
is the user I work with every day.

stef is a new and "empty" user I created lately to check things with
this gdm-topic. stef has no old stuff in its home-dir but fails to login
as well as sgw.

> What do you mean
> by "I use "su" within my terminator-sessions very often"?

For my work I prefer x11-terms/terminator over gnome-terminal, it allows
me to run multiple gnome-terminals in one window.

I login to gnome as user sgw ... open up a terminator ... and if I need
root privileges for doing stuff I run "su" within the terminator session.

ok?

Greets, Stefan
Stefan G. Weichinger
2013-02-09 18:56:52 UTC
Permalink
Next episode:

I also migrated my gentoo thinkpad to systemd today.

Generally very similar to my desktop ... ~amd64 with Gnome 3.6.

Things went pretty well, I have to say.

I can login to gdm here! ;-)

An issue I haven't solved yet: encrypted swap.

I always get timeouts as systemd waits for the decrypted mapper-device
to come up. Swap doesn't get enabled but when it finally continues to
boot I see the valid mapper-device there and can "swapon" it manually.

# cat /etc/crypttab

swap /dev/disk/by-id/ata-INTEL_SSDSA2M080G2GC_CVPO015404LR080JGN-part5
/dev/urandom swap,cipher=aes-cbc-essiv:sha256,size=256

# grep swap /etc/fstab
/dev/mapper/swap none swap defaults 0 0

AFAI understand these 2 lines should be enough to let systemd generate
its relevant unit-files etc.

Right?

Best regards, have a nice weekend, Stefan
Stefan G. Weichinger
2013-02-09 19:44:29 UTC
Permalink
Am 2013-02-09 19:56, schrieb Stefan G. Weichinger:

> AFAI understand these 2 lines should be enough to let systemd generate
> its relevant unit-files etc.
>
> Right?

Additional thoughts:

Is pam_mount obsolete with systemd?

It is possible to mount my /home via systemd-unit as well ... the
difference seems to be that systemd would (try to) mount it at boot-time
while with pam_mount it would be mounted at login.

Thoughts? Experiences?

Stefan
Canek Peláez Valdés
2013-02-09 23:30:32 UTC
Permalink
On Sat, Feb 9, 2013 at 1:44 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 2013-02-09 19:56, schrieb Stefan G. Weichinger:
>
>> AFAI understand these 2 lines should be enough to let systemd generate
>> its relevant unit-files etc.
>>
>> Right?
>
> Additional thoughts:
>
> Is pam_mount obsolete with systemd?

I don't know if "obsolete" is the correct definition, but it is not
installed in any of my systems.

> It is possible to mount my /home via systemd-unit as well ... the
> difference seems to be that systemd would (try to) mount it at boot-time
> while with pam_mount it would be mounted at login.

You can mount almost all partitions with system units; there was a
discussion some days ago about getting rid of /etc/fstab for the
embedded case and stuff like that. Also, you can set the .mount unit
for your $HOME, and make the gdm service depend on it (it would be
mounted at gdm startup, not at session startup, though).

> Thoughts? Experiences?

I have never used pam_mount; what's the upside? Just delaying the
mounting (and perhaps fsck'ing) of the partition until session login?

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Canek Peláez Valdés
2013-02-09 23:26:08 UTC
Permalink
On Sat, Feb 9, 2013 at 12:56 PM, Stefan G. Weichinger <***@xunil.at> wrote:
>
> Next episode:
>
> I also migrated my gentoo thinkpad to systemd today.

Cool.

> Generally very similar to my desktop ... ~amd64 with Gnome 3.6.
>
> Things went pretty well, I have to say.
>
> I can login to gdm here! ;-)

Try to list the differences between your laptop and your desktop
(world files, USE flags, partition schemes, etc.) That was my approach
when you described your problem to me; try to see what it
differentiates from mine, but we never got too far with tat.

> An issue I haven't solved yet: encrypted swap.
>
> I always get timeouts as systemd waits for the decrypted mapper-device
> to come up. Swap doesn't get enabled but when it finally continues to
> boot I see the valid mapper-device there and can "swapon" it manually.
>
> # cat /etc/crypttab
>
> swap /dev/disk/by-id/ata-INTEL_SSDSA2M080G2GC_CVPO015404LR080JGN-part5
> /dev/urandom swap,cipher=aes-cbc-essiv:sha256,size=256
>
> # grep swap /etc/fstab
> /dev/mapper/swap none swap defaults 0 0
>
> AFAI understand these 2 lines should be enough to let systemd generate
> its relevant unit-files etc.
>
> Right?

I haven't used an encrypted swap (nor partition), but I believe that's
all you need. A workaround perhaps is to put the nofail option, which
at least will skip the partition when booting.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-10 10:09:56 UTC
Permalink
Am 2013-02-10 00:26, schrieb Canek Peláez Valdés:
> On Sat, Feb 9, 2013 at 12:56 PM, Stefan G. Weichinger
> <***@xunil.at> wrote:
>>
>> Next episode:
>>
>> I also migrated my gentoo thinkpad to systemd today.
>
> Cool.

;-) ... next try, I had systemd on both of my work-systems already a
year ago or so ... but there were some showstoppers back then. This time
I already solved most of them ....

>> I can login to gdm here! ;-)
>
> Try to list the differences between your laptop and your desktop
> (world files, USE flags, partition schemes, etc.) That was my
> approach when you described your problem to me; try to see what it
> differentiates from mine, but we never got too far with tat.

Your help was great ... and we already excluded several issues.

> I haven't used an encrypted swap (nor partition), but I believe
> that's all you need. A workaround perhaps is to put the nofail
> option, which at least will skip the partition when booting.

Into fstab, right? I will try to play with automount-options as well.

S
Stefan G. Weichinger
2013-02-10 10:35:42 UTC
Permalink
Am 2013-02-10 11:09, schrieb Stefan G. Weichinger:

> Into fstab, right? I will try to play with automount-options as well.

"nofail" gives me a straight bootup as the system does not (try to)
enable swap. The cryptsetup-unit gets set up correctly:

# cryptsetup status swap
/dev/mapper/swap is active.
type: PLAIN
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sda5
offset: 0 sectors
size: 2093056 sectors
mode: read/write

I can "swapon" manually, but it didn't get enabled by systemd so far.

Maybe it *would* have been enabled if swap was accessed?

I can live with that so far ... but it would be interesting to get that
right, just to learn things.

Maybe I should try the other approach and create a unit-file for the
encrypted swap by myself.

Stefan
Stefan G. Weichinger
2013-02-10 14:41:10 UTC
Permalink
Am 2013-02-10 11:35, schrieb Stefan G. Weichinger:

> I can live with that so far ... but it would be interesting to get that
> right, just to learn things.
>
> Maybe I should try the other approach and create a unit-file for the
> encrypted swap by myself.

Next steps done ... no real success so far.
What I find in the journal:

Feb 10 15:34:46 enzo systemd-udevd[3848]: conflicting device node
'/dev/mapper/swap' found, link to '/dev/dm-0' will not be created
Feb 10 15:34:46 enzo mkswap[5420]: [58B blob data]
Feb 10 15:34:46 enzo mkswap[5420]: kein Label,
UUID=fe8f8768-bffb-4c05-8ebd-763a50cd6ae4
Feb 10 15:34:46 enzo systemd[1]: Started Cryptography Setup for swap.
Feb 10 15:34:46 enzo systemd[1]: Starting Encrypted Volumes.
Feb 10 15:34:46 enzo systemd[1]: Reached target Encrypted Volumes.

...

Feb 10 15:36:16 enzo systemd[1]: Job dev-mapper-swap.device/start timed out.
Feb 10 15:36:16 enzo systemd[1]: Timed out waiting for device
dev-mapper-swap.device.
Feb 10 15:36:16 enzo systemd[1]: Dependency failed for /dev/mapper/swap.

hmmm ....

gotta google that ...

Stefan
Stefan G. Weichinger
2013-02-10 19:32:08 UTC
Permalink
Am 2013-02-10 15:41, schrieb Stefan G. Weichinger:
> Am 2013-02-10 11:35, schrieb Stefan G. Weichinger:
>
>> I can live with that so far ... but it would be interesting to get that
>> right, just to learn things.

next learnings (I maybe should write some wiki-entry somewhere to
collect all that for others ...):

systemd and acpid and Gnome all try to handle suspending my thinkpad to
RAM ... it seems.

So I get the behavior that it suspends fine when I close the lid but
when I open it again I get an immediate suspend *again* ... I then tried
to disable acpid completely, same behavior.

I also cleaned up /etc/acpi and re-installed acpid to get default behavior.

"systemctl suspend" from the shell (without that lid-event) works fine
and resumes correctly.

Interesting ;-)

I already noticed that there are parameters to let systemd ignore the
various acpi-events (haven't yet tested that) but I would prefer to let
systemd handle suspend and resume.

-

Canek, do you run systemd on a laptop as well?

Anyone else?

Best regards, Stefan
Canek Peláez Valdés
2013-02-10 19:47:16 UTC
Permalink
On Sun, Feb 10, 2013 at 1:32 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 2013-02-10 15:41, schrieb Stefan G. Weichinger:
>> Am 2013-02-10 11:35, schrieb Stefan G. Weichinger:
>>
>>> I can live with that so far ... but it would be interesting to get that
>>> right, just to learn things.
>
> next learnings (I maybe should write some wiki-entry somewhere to
> collect all that for others ...):
>
> systemd and acpid and Gnome all try to handle suspending my thinkpad to
> RAM ... it seems.
>
> So I get the behavior that it suspends fine when I close the lid but
> when I open it again I get an immediate suspend *again* ... I then tried
> to disable acpid completely, same behavior.
>
> I also cleaned up /etc/acpi and re-installed acpid to get default behavior.
>
> "systemctl suspend" from the shell (without that lid-event) works fine
> and resumes correctly.
>
> Interesting ;-)
>
> I already noticed that there are parameters to let systemd ignore the
> various acpi-events (haven't yet tested that) but I would prefer to let
> systemd handle suspend and resume.
>
> -
>
> Canek, do you run systemd on a laptop as well?

Yep, had the same problem, solved with:

LidSwitchIgnoreInhibited=no

in /etc/systemd/logind.conf. Since then it has happened again maybe a
couple of times (I have no idea why), but most of the time (and I'm
talking above 99%), it works as intended.

These options are pretty new, I think they went live after GNOME 3.6,
so I hope that with GNOME 3.8 we will be able to comment that line
again and everything will work automagically.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-11 09:51:23 UTC
Permalink
Am 10.02.2013 20:47, schrieb Canek Peláez Valdés:

> Yep, had the same problem, solved with:
>
> LidSwitchIgnoreInhibited=no
>
> in /etc/systemd/logind.conf. Since then it has happened again maybe a
> couple of times (I have no idea why), but most of the time (and I'm
> talking above 99%), it works as intended.
>
> These options are pretty new, I think they went live after GNOME 3.6,
> so I hope that with GNOME 3.8 we will be able to comment that line
> again and everything will work automagically.

hopefully ...

Unfortunately that parameter didn't help so far.

Do you have acpid installed/enabled? Anything aside the default
acpi-scripts?

I removed hibernate-script from my system now just to check things.

Stefan
Canek Peláez Valdés
2013-02-11 21:03:00 UTC
Permalink
On Mon, Feb 11, 2013 at 3:51 AM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 10.02.2013 20:47, schrieb Canek Peláez Valdés:
>
>> Yep, had the same problem, solved with:
>>
>> LidSwitchIgnoreInhibited=no
>>
>> in /etc/systemd/logind.conf. Since then it has happened again maybe a
>> couple of times (I have no idea why), but most of the time (and I'm
>> talking above 99%), it works as intended.
>>
>> These options are pretty new, I think they went live after GNOME 3.6,
>> so I hope that with GNOME 3.8 we will be able to comment that line
>> again and everything will work automagically.
>
> hopefully ...
>
> Unfortunately that parameter didn't help so far.
>
> Do you have acpid installed/enabled? Anything aside the default
> acpi-scripts?

The last time I installed acpid was in November of 2010, and I
uninstalled for the last time in April 2011. My machines are all acpid
free since then; systemd + UPower takes cares of everything AFAIK.

> I removed hibernate-script from my system now just to check things.

I haven't used scripts to suspend or hibernate in ages; again UPower
does everything, or perhaps some other part of the GNOME stack.
sys-power/pm-utils is still being pulled in by upower-0.9.19, but it
only calls pm-is-supported (src/linux/up-backend.c:363-390) to
determine if the machine can suspend/hibernate. Which is kinda stupid,
since pm-is-supported is only a set of scripts which test files in the
/sys directory. UPower should test for those files directly (is in the
linux backend anyway), and remove the pm-utils dependency.

For the kernel I use vanilla-sources unstable; I haven't used
gentoo-sources in ages (long before systemd), and I never used
tuxonice-sources.

Suspend/hibernate works perfectly in all my machines; I haven't had a
failed resume in (literally) years.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-11 21:19:18 UTC
Permalink
Am 11.02.2013 22:03, schrieb Canek Peláez Valdés:
> On Mon, Feb 11, 2013 at 3:51 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>> Do you have acpid installed/enabled? Anything aside the default
>> acpi-scripts?
>
> The last time I installed acpid was in November of 2010, and I
> uninstalled for the last time in April 2011. My machines are all acpid
> free since then; systemd + UPower takes cares of everything AFAIK.

I see. I take this as an impulse to cleanup my system even more ...
removing acpi means getting rid of those app-laptop/laptop-mode-tools as
well afai understand ... they relied on ACPI to switch stuff ...

Both removed now ...

> I haven't used scripts to suspend or hibernate in ages; again UPower
> does everything, or perhaps some other part of the GNOME stack.
> sys-power/pm-utils is still being pulled in by upower-0.9.19, but it
> only calls pm-is-supported (src/linux/up-backend.c:363-390) to
> determine if the machine can suspend/hibernate. Which is kinda stupid,
> since pm-is-supported is only a set of scripts which test files in the
> /sys directory. UPower should test for those files directly (is in the
> linux backend anyway), and remove the pm-utils dependency.

Yep, another issue (bug-report ;-) ).

> For the kernel I use vanilla-sources unstable; I haven't used
> gentoo-sources in ages (long before systemd), and I never used
> tuxonice-sources.

I see. gentoo-sources here, 3.7.6 at the moment, from time to time I
"git pull" some kernel from linux-git or linux-stable (Linus or Greg ...).

> Suspend/hibernate works perfectly in all my machines; I haven't had a
> failed resume in (literally) years.

Good to hear.

I see upower.service as active but disabled ... ? hmm..

Stefan
Canek Peláez Valdés
2013-02-11 21:30:16 UTC
Permalink
On Mon, Feb 11, 2013 at 3:19 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 11.02.2013 22:03, schrieb Canek Peláez Valdés:
>> On Mon, Feb 11, 2013 at 3:51 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>>> Do you have acpid installed/enabled? Anything aside the default
>>> acpi-scripts?
>>
>> The last time I installed acpid was in November of 2010, and I
>> uninstalled for the last time in April 2011. My machines are all acpid
>> free since then; systemd + UPower takes cares of everything AFAIK.
>
> I see. I take this as an impulse to cleanup my system even more ...
> removing acpi means getting rid of those app-laptop/laptop-mode-tools as
> well afai understand ... they relied on ACPI to switch stuff ...
>
> Both removed now ...
>
>> I haven't used scripts to suspend or hibernate in ages; again UPower
>> does everything, or perhaps some other part of the GNOME stack.
>> sys-power/pm-utils is still being pulled in by upower-0.9.19, but it
>> only calls pm-is-supported (src/linux/up-backend.c:363-390) to
>> determine if the machine can suspend/hibernate. Which is kinda stupid,
>> since pm-is-supported is only a set of scripts which test files in the
>> /sys directory. UPower should test for those files directly (is in the
>> linux backend anyway), and remove the pm-utils dependency.
>
> Yep, another issue (bug-report ;-) ).
>
>> For the kernel I use vanilla-sources unstable; I haven't used
>> gentoo-sources in ages (long before systemd), and I never used
>> tuxonice-sources.
>
> I see. gentoo-sources here, 3.7.6 at the moment, from time to time I
> "git pull" some kernel from linux-git or linux-stable (Linus or Greg ...).
>
>> Suspend/hibernate works perfectly in all my machines; I haven't had a
>> failed resume in (literally) years.
>
> Good to hear.
>
> I see upower.service as active but disabled ... ? hmm..

It's OK; disabled means that it's not enabled, i.e., there is no link
to it from /etc/systemd/system/*.wants. It's Dbus activable, so the
first time someone calls a method from org.freedesktop.UPower via
dbus, the service is activated automatically. There is no need to
enable the service (which will mean that it starts even if no other
process calls a method from org.freedesktop.UPower).

Enabled/Disabled is orthogonal to Active/Inactive; the first means
"the service will start when reaching its target no matter what", and
the latter means "the service is running".

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-11 21:51:43 UTC
Permalink
Am 11.02.2013 22:30, schrieb Canek Peláez Valdés:
> On Mon, Feb 11, 2013 at 3:19 PM, Stefan G. Weichinger <***@xunil.at> wrote:
>> I see upower.service as active but disabled ... ? hmm..
>
> It's OK; disabled means that it's not enabled, i.e., there is no link
> to it from /etc/systemd/system/*.wants. It's Dbus activable, so the
> first time someone calls a method from org.freedesktop.UPower via
> dbus, the service is activated automatically. There is no need to
> enable the service (which will mean that it starts even if no other
> process calls a method from org.freedesktop.UPower).

I assumed something like that, yes. Thanks for explaining.

Still seeing that immediate suspend after resume ... it looks as if it
actually gets the signal to suspend twice somehow. Maybe some
ACPI-related issue on the thinkpad, I read about something like that
back with HAL etc. (the lid-open-event was interpreted as lid-close or
similar ...).

Anyway, enough for today. Late here.

Thanks, greets, Stefan
Stefan G. Weichinger
2013-02-13 20:32:59 UTC
Permalink
Am 11.02.2013 22:03, schrieb Canek Peláez Valdés:
> On Mon, Feb 11, 2013 at 3:51 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>> Do you have acpid installed/enabled? Anything aside the default
>> acpi-scripts?
>
> The last time I installed acpid was in November of 2010, and I
> uninstalled for the last time in April 2011. My machines are all acpid
> free since then; systemd + UPower takes cares of everything AFAIK.

getting OT here ... ->

Would you mind showing the results of

zgrep ACPI /proc/config.gz

?

USE= " .... -acpi ....." as well?

Stefan
Canek Peláez Valdés
2013-02-13 21:08:19 UTC
Permalink
On Wed, Feb 13, 2013 at 2:32 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 11.02.2013 22:03, schrieb Canek Peláez Valdés:
>> On Mon, Feb 11, 2013 at 3:51 AM, Stefan G. Weichinger <***@xunil.at> wrote:
>>> Do you have acpid installed/enabled? Anything aside the default
>>> acpi-scripts?
>>
>> The last time I installed acpid was in November of 2010, and I
>> uninstalled for the last time in April 2011. My machines are all acpid
>> free since then; systemd + UPower takes cares of everything AFAIK.
>
> getting OT here ... ->
>
> Would you mind showing the results of
>
> zgrep ACPI /proc/config.gz

I haven't had /proc/config.gz in years, but I always keep my current config:

# grep ACPI config-3.7.6
# Power management and ACPI options
CONFIG_ACPI=y
CONFIG_ACPI_SLEEP=y
# CONFIG_ACPI_PROCFS is not set
# CONFIG_ACPI_PROCFS_POWER is not set
# CONFIG_ACPI_EC_DEBUGFS is not set
# CONFIG_ACPI_PROC_EVENT is not set
CONFIG_ACPI_AC=m
CONFIG_ACPI_BATTERY=m
CONFIG_ACPI_BUTTON=m
CONFIG_ACPI_VIDEO=m
CONFIG_ACPI_FAN=m
CONFIG_ACPI_DOCK=y
CONFIG_ACPI_PROCESSOR=m
CONFIG_ACPI_HOTPLUG_CPU=y
CONFIG_ACPI_PROCESSOR_AGGREGATOR=m
CONFIG_ACPI_THERMAL=m
# CONFIG_ACPI_CUSTOM_DSDT is not set
CONFIG_ACPI_BLACKLIST_YEAR=0
CONFIG_ACPI_DEBUG=y
# CONFIG_ACPI_DEBUG_FUNC_TRACE is not set
CONFIG_ACPI_PCI_SLOT=m
CONFIG_ACPI_CONTAINER=m
CONFIG_ACPI_SBS=m
# CONFIG_ACPI_HED is not set
CONFIG_ACPI_CUSTOM_METHOD=m
# CONFIG_ACPI_APEI is not set
CONFIG_X86_ACPI_CPUFREQ=m
CONFIG_X86_ACPI_CPUFREQ_CPB=y
CONFIG_PNPACPI=y
CONFIG_ATA_ACPI=y
# CONFIG_PATA_ACPI is not set
# ACPI drivers
# CONFIG_THINKPAD_ACPI is not set
CONFIG_ACPI_WMI=m
CONFIG_ACPI_TOSHIBA=m
# CONFIG_ACPI_CMPC is not set

This is my laptop, my desktop is obviously a little different.

> USE= " .... -acpi ....." as well?

There are still relevant packages that have "acpi"? AFAIK, the kernel,
udev and the userspace *kits take are of everything. In my laptop I
don't have the acpi USE flag, but the interesting fact is that I can
do an "emerge -uDNvp world", with USE=acpi OR USE=-acpi, and nothing
gets rebuilt.

In other words, in my laptop I don't have a single package that uses
the acpi USE flag. I just did this:

# grep "IUSE=\".*acpi.*\"" $(find /usr/portage -name "*.ebuild") | cut
-d / -f 4-5 | sort | uniq
app-emulation/bochs
app-laptop/laptop-mode-tools
app-laptop/pbbuttonsd
sys-freebsd/freebsd-usbin
sys-power/cpufreqd
x11-drivers/nvidia-drivers
x11-misc/keytouch
x11-plugins/wmacpi
x11-wm/matchbox-panel
xfce-extra/xfce4-sensors-plugin

Those are ALL the packages that have the acpi USE flag in the tree. Of
those, I just use nvidia-drivers in my media center, and I must use
USE="-acpi", since I don't have acpid in any of my machines (the flag
just plugs in acpid).

I do use ACPI, but (as I said) the kernel, udev and the userspace
*kits take care of it. The acpid daemon is a relic from the time when
we didn't have standarized interfaces for ACPI events, like power
button pushes. Nowadays we have smarter software that deals with it.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-13 21:36:27 UTC
Permalink
Am 13.02.2013 22:08, schrieb Canek Peláez Valdés:


> # grep ACPI config-3.7.6
> # Power management and ACPI options
> CONFIG_ACPI=y
> CONFIG_ACPI_SLEEP=y

[..]

> This is my laptop, my desktop is obviously a little different.

I'll compare with my thinkpad asap, just to get an impression ...

> I do use ACPI, but (as I said) the kernel, udev and the userspace
> *kits take care of it. The acpid daemon is a relic from the time when
> we didn't have standarized interfaces for ACPI events, like power
> button pushes. Nowadays we have smarter software that deals with it.

... as I see with my resume-issue :-P ;-)

Thanks a lot for sharing, Stefan
Stefan G. Weichinger
2013-02-16 13:02:44 UTC
Permalink
Am 2013-02-10 20:32, schrieb Stefan G. Weichinger:

> systemd and acpid and Gnome all try to handle suspending my thinkpad to
> RAM ... it seems.
>
> So I get the behavior that it suspends fine when I close the lid but
> when I open it again I get an immediate suspend *again* ... I then tried
> to disable acpid completely, same behavior.

Got that. I called "gnome-tweak-tool" and disabled the Gnome-action for
close/open lid. Works now and I *think* it is managed by systemd alone now.

Stefan
Stefan G. Weichinger
2013-02-13 18:17:02 UTC
Permalink
Am 2013-02-10 15:41, schrieb Stefan G. Weichinger:
> conflicting device node
> '/dev/mapper/swap' found

googled that and found this similar issue:

https://bugs.archlinux.org/task/32403

The command there returns on my thinkpad:


# journalctl -b --no-pager _SYSTEMD_UNIT=systemd-udevd.service
-- Logs begin at Mit 2012-08-15 20:16:00 CEST, end at Mit 2013-02-13
19:03:42 CET. --
Feb 13 18:48:30 enzo systemd-udevd[4415]: failed to execute
'/lib/udev/lmt-udev' '/lib/udev/lmt-udev auto': No such file or directory
Feb 13 18:48:30 enzo systemd-udevd[4519]: failed to execute
'/lib/udev/lmt-udev' '/lib/udev/lmt-udev force modules=usb-autosuspend
devices=1-1.3:1...directory
Feb 13 18:48:30 enzo systemd-udevd[3857]: conflicting device node
'/dev/mapper/swap' found, link to '/dev/dm-0' will not be created
Feb 13 18:53:28 enzo systemd-udevd[5411]: conflicting device node
'/dev/mapper/_dev_sda4' found, link to '/dev/dm-1' will not be created


hmm.

Might be that I have some outdated udev-rules, right?

# equery f udev | grep rules

/lib/udev/rules.d/40-gentoo.rules
/lib/udev/rules.d/42-usb-hid-pm.rules
/lib/udev/rules.d/50-udev-default.rules
/lib/udev/rules.d/60-cdrom_id.rules
/lib/udev/rules.d/60-persistent-alsa.rules
/lib/udev/rules.d/60-persistent-input.rules
/lib/udev/rules.d/60-persistent-serial.rules
/lib/udev/rules.d/60-persistent-storage-tape.rules
/lib/udev/rules.d/60-persistent-storage.rules
/lib/udev/rules.d/60-persistent-v4l.rules
/lib/udev/rules.d/61-accelerometer.rules
/lib/udev/rules.d/64-btrfs.rules
/lib/udev/rules.d/75-net-description.rules
/lib/udev/rules.d/75-probe_mtd.rules
/lib/udev/rules.d/75-tty-description.rules
/lib/udev/rules.d/78-sound-card.rules
/lib/udev/rules.d/80-drivers.rules
/lib/udev/rules.d/80-net-name-slot.rules
/lib/udev/rules.d/95-keyboard-force-release.rules
/lib/udev/rules.d/95-keymap.rules
/lib/udev/rules.d/95-udev-late.rules

while I have stuff like this in /lib/udev/rules.d:

# ls -1 /lib/udev/rules.d/
10-dm.rules
11-dm-lvm.rules
13-dm-disk.rules
40-gentoo.rules
40-usb-media-players.rules
40-usb_modeswitch.rules
42-usb-hid-pm.rules
50-udev-default.rules
60-cdrom_id.rules
60-fprint-autosuspend.rules
60-persistent-alsa.rules
60-persistent-input.rules
60-persistent-serial.rules
60-persistent-storage.rules
60-persistent-storage-tape.rules
60-persistent-v4l.rules
60-vmware.rules
60-zvol.rules
61-accelerometer.rules
61-gnome-bluetooth.rules
64-btrfs.rules
64-md-raid.rules
65-kvm.rules
69-cd-sensors.rules
69-dm-lvm-metad.rules
69-vdev.rules
70-libgphoto2.rules
70-power-switch.rules
70-printers.rules
70-uaccess.rules
71-seat.rules
73-seat-late.rules
75-net-description.rules
75-probe_mtd.rules
75-tty-description.rules
77-mm-ericsson-mbm.rules
77-mm-longcheer-port-types.rules
77-mm-nokia-port-types.rules
77-mm-pcmcia-device-blacklist.rules
77-mm-platform-serial-whitelist.rules
77-mm-simtech-port-types.rules
77-mm-usb-device-blacklist.rules
77-mm-x22x-port-types.rules
77-mm-zte-port-types.rules
77-nm-olpc-mesh.rules
78-sound-card.rules
80-drivers.rules
80-mm-candidate.rules
80-net-name-slot.rules
80-udisks2.rules
80-udisks.rules
85-regulatory.rules
90-alsa-restore.rules
90-network.rules
90-pulseaudio.rules
90-zfs.rules
95-cd-devices.rules
95-dm-notify.rules
95-keyboard-force-release.rules
95-keymap.rules
95-osinfo.rules
95-udev-late.rules
95-upower-battery-recall-dell.rules
95-upower-battery-recall-fujitsu.rules
95-upower-battery-recall-gateway.rules
95-upower-battery-recall-ibm.rules
95-upower-battery-recall-lenovo.rules
95-upower-battery-recall-toshiba.rules
95-upower-csr.rules
95-upower-hid.rules
95-upower-wup.rules
97-bluetooth-hid2hci.rules
99-fuse.rules
99-ntfs3g.rules
99-systemd.rules

*sigh*

Before I completely ruin that setup:

Would someone pls compare that directory's content with mine and tell me
what to get rid of?

-

I assume the problematic files (in the issue with encrypted swap) come
from lvm2:

# equery b 10-dm.rules
* Searching for 10-dm.rules ...
sys-fs/lvm2-2.02.98 (/lib/udev/rules.d/10-dm.rules)

# equery b 11-dm-lvm.rules
* Searching for 11-dm-lvm.rules ...
sys-fs/lvm2-2.02.98 (/lib/udev/rules.d/11-dm-lvm.rules)

....

I will unmerge lvm2 for now as I think I don't need it on the laptop (I
don't use PVs/VGs/LVs here and that device-mapper-stuff happens
elsewhere, correct?)

S
Canek Peláez Valdés
2013-02-13 20:01:19 UTC
Permalink
On Wed, Feb 13, 2013 at 12:17 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 2013-02-10 15:41, schrieb Stefan G. Weichinger:
>> conflicting device node
>> '/dev/mapper/swap' found
>
> googled that and found this similar issue:
>
> https://bugs.archlinux.org/task/32403
>
> The command there returns on my thinkpad:
>
>
> # journalctl -b --no-pager _SYSTEMD_UNIT=systemd-udevd.service
> -- Logs begin at Mit 2012-08-15 20:16:00 CEST, end at Mit 2013-02-13
> 19:03:42 CET. --
> Feb 13 18:48:30 enzo systemd-udevd[4415]: failed to execute
> '/lib/udev/lmt-udev' '/lib/udev/lmt-udev auto': No such file or directory
> Feb 13 18:48:30 enzo systemd-udevd[4519]: failed to execute
> '/lib/udev/lmt-udev' '/lib/udev/lmt-udev force modules=usb-autosuspend
> devices=1-1.3:1...directory
> Feb 13 18:48:30 enzo systemd-udevd[3857]: conflicting device node
> '/dev/mapper/swap' found, link to '/dev/dm-0' will not be created
> Feb 13 18:53:28 enzo systemd-udevd[5411]: conflicting device node
> '/dev/mapper/_dev_sda4' found, link to '/dev/dm-1' will not be created
>
>
> hmm.
>
> Might be that I have some outdated udev-rules, right?
>
> # equery f udev | grep rules
>
> /lib/udev/rules.d/40-gentoo.rules
> /lib/udev/rules.d/42-usb-hid-pm.rules
> /lib/udev/rules.d/50-udev-default.rules
> /lib/udev/rules.d/60-cdrom_id.rules
> /lib/udev/rules.d/60-persistent-alsa.rules
> /lib/udev/rules.d/60-persistent-input.rules
> /lib/udev/rules.d/60-persistent-serial.rules
> /lib/udev/rules.d/60-persistent-storage-tape.rules
> /lib/udev/rules.d/60-persistent-storage.rules
> /lib/udev/rules.d/60-persistent-v4l.rules
> /lib/udev/rules.d/61-accelerometer.rules
> /lib/udev/rules.d/64-btrfs.rules
> /lib/udev/rules.d/75-net-description.rules
> /lib/udev/rules.d/75-probe_mtd.rules
> /lib/udev/rules.d/75-tty-description.rules
> /lib/udev/rules.d/78-sound-card.rules
> /lib/udev/rules.d/80-drivers.rules
> /lib/udev/rules.d/80-net-name-slot.rules
> /lib/udev/rules.d/95-keyboard-force-release.rules
> /lib/udev/rules.d/95-keymap.rules
> /lib/udev/rules.d/95-udev-late.rules
>
> while I have stuff like this in /lib/udev/rules.d:
>
> # ls -1 /lib/udev/rules.d/
> 10-dm.rules
> 11-dm-lvm.rules
> 13-dm-disk.rules
> 40-gentoo.rules
> 40-usb-media-players.rules
> 40-usb_modeswitch.rules
> 42-usb-hid-pm.rules
> 50-udev-default.rules
> 60-cdrom_id.rules
> 60-fprint-autosuspend.rules
> 60-persistent-alsa.rules
> 60-persistent-input.rules
> 60-persistent-serial.rules
> 60-persistent-storage.rules
> 60-persistent-storage-tape.rules
> 60-persistent-v4l.rules
> 60-vmware.rules
> 60-zvol.rules
> 61-accelerometer.rules
> 61-gnome-bluetooth.rules
> 64-btrfs.rules
> 64-md-raid.rules
> 65-kvm.rules
> 69-cd-sensors.rules
> 69-dm-lvm-metad.rules
> 69-vdev.rules
> 70-libgphoto2.rules
> 70-power-switch.rules
> 70-printers.rules
> 70-uaccess.rules
> 71-seat.rules
> 73-seat-late.rules
> 75-net-description.rules
> 75-probe_mtd.rules
> 75-tty-description.rules
> 77-mm-ericsson-mbm.rules
> 77-mm-longcheer-port-types.rules
> 77-mm-nokia-port-types.rules
> 77-mm-pcmcia-device-blacklist.rules
> 77-mm-platform-serial-whitelist.rules
> 77-mm-simtech-port-types.rules
> 77-mm-usb-device-blacklist.rules
> 77-mm-x22x-port-types.rules
> 77-mm-zte-port-types.rules
> 77-nm-olpc-mesh.rules
> 78-sound-card.rules
> 80-drivers.rules
> 80-mm-candidate.rules
> 80-net-name-slot.rules
> 80-udisks2.rules
> 80-udisks.rules
> 85-regulatory.rules
> 90-alsa-restore.rules
> 90-network.rules
> 90-pulseaudio.rules
> 90-zfs.rules
> 95-cd-devices.rules
> 95-dm-notify.rules
> 95-keyboard-force-release.rules
> 95-keymap.rules
> 95-osinfo.rules
> 95-udev-late.rules
> 95-upower-battery-recall-dell.rules
> 95-upower-battery-recall-fujitsu.rules
> 95-upower-battery-recall-gateway.rules
> 95-upower-battery-recall-ibm.rules
> 95-upower-battery-recall-lenovo.rules
> 95-upower-battery-recall-toshiba.rules
> 95-upower-csr.rules
> 95-upower-hid.rules
> 95-upower-wup.rules
> 97-bluetooth-hid2hci.rules
> 99-fuse.rules
> 99-ntfs3g.rules
> 99-systemd.rules
>
> *sigh*
>
> Before I completely ruin that setup:
>
> Would someone pls compare that directory's content with mine and tell me
> what to get rid of?

You should not be deleting rules arbitrarily from your system;
perharps you have orphan rules, and you can detect that by equery'ing
to what package each rule belongs.

> I assume the problematic files (in the issue with encrypted swap) come
> from lvm2:
>
> # equery b 10-dm.rules
> * Searching for 10-dm.rules ...
> sys-fs/lvm2-2.02.98 (/lib/udev/rules.d/10-dm.rules)
>
> # equery b 11-dm-lvm.rules
> * Searching for 11-dm-lvm.rules ...
> sys-fs/lvm2-2.02.98 (/lib/udev/rules.d/11-dm-lvm.rules)
>
> ....
>
> I will unmerge lvm2 for now as I think I don't need it on the laptop (I
> don't use PVs/VGs/LVs here and that device-mapper-stuff happens
> elsewhere, correct?)

LVM2 will be pulled again by udisks2; it's a mandatory dependency. I
don't think the problem is related to LVM, but I don't really know.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-02-13 20:05:58 UTC
Permalink
Am 2013-02-13 21:01, schrieb Canek Peláez Valdés:

> LVM2 will be pulled again by udisks2; it's a mandatory dependency. I
> don't think the problem is related to LVM, but I don't really know.

I already found out, yes. Maybe it's just some systemd race condition.
It's not that important to me right now ... I will check back with later
releases and google bug-reports now and then.

Stefan
Stefan G. Weichinger
2013-03-04 19:42:26 UTC
Permalink
Am 13.02.2013 21:05, schrieb Stefan G. Weichinger:
> Am 2013-02-13 21:01, schrieb Canek Peláez Valdés:
>
>> LVM2 will be pulled again by udisks2; it's a mandatory dependency. I
>> don't think the problem is related to LVM, but I don't really know.
>
> I already found out, yes. Maybe it's just some systemd race condition.
> It's not that important to me right now ... I will check back with later
> releases and google bug-reports now and then.

Another update:

posted to systemd-devel and got pointed to LVM/DM-people ... I don't know:

http://lists.freedesktop.org/archives/systemd-devel/2013-March/009323.html

-

Next systemd-issue (yes, I know ... openrc is there as well ...):

I get warnings like "superblock date is in the future --- REPAIRED" for
filesystems at boot-time.

Canek, (how) do you handle time and hwclock with systemd?

I *had* a hwclock.service and removed it now ... no change.

Greets, thanks, Stefan
Stefan G. Weichinger
2013-03-04 20:07:36 UTC
Permalink
Am 04.03.2013 20:42, schrieb Stefan G. Weichinger:
> Next systemd-issue (yes, I know ... openrc is there as well ...):
>
> I get warnings like "superblock date is in the future --- REPAIRED" for
> filesystems at boot-time.
>
> Canek, (how) do you handle time and hwclock with systemd?
>
> I *had* a hwclock.service and removed it now ... no change.

Aside from being interested if to run hwclock.service:

solved that by entering BIOS and correcting time (was one hour behind,
why ever ...)

Stefan
Canek Peláez Valdés
2013-03-05 06:36:23 UTC
Permalink
On Mon, Mar 4, 2013 at 2:07 PM, Stefan G. Weichinger <***@xunil.at> wrote:
> Am 04.03.2013 20:42, schrieb Stefan G. Weichinger:
>> Next systemd-issue (yes, I know ... openrc is there as well ...):
>>
>> I get warnings like "superblock date is in the future --- REPAIRED" for
>> filesystems at boot-time.
>>
>> Canek, (how) do you handle time and hwclock with systemd?

I don't. AFAIK, systemd provides systemd-timedated(8) since systemd 30:

http://www.freedesktop.org/wiki/Software/systemd/timedated

In normal desktops/laptops/servers, it just works.

>> I *had* a hwclock.service and removed it now ... no change.
>
> Aside from being interested if to run hwclock.service:
>
> solved that by entering BIOS and correcting time (was one hour behind,
> why ever ...)

It helps if the hardware clock is set to the correct time, yes. The
only problem is if you dual boot Windows (or so I heard).

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Canek Peláez Valdés
2013-03-05 06:40:05 UTC
Permalink
On Tue, Mar 5, 2013 at 12:36 AM, Canek Peláez Valdés <***@gmail.com> wrote:
> On Mon, Mar 4, 2013 at 2:07 PM, Stefan G. Weichinger <***@xunil.at> wrote:
>> Am 04.03.2013 20:42, schrieb Stefan G. Weichinger:
>>> Next systemd-issue (yes, I know ... openrc is there as well ...):
>>>
>>> I get warnings like "superblock date is in the future --- REPAIRED" for
>>> filesystems at boot-time.
>>>
>>> Canek, (how) do you handle time and hwclock with systemd?
>
> I don't. AFAIK, systemd provides systemd-timedated(8) since systemd 30:
>
> http://www.freedesktop.org/wiki/Software/systemd/timedated
>
> In normal desktops/laptops/servers, it just works.
>
>>> I *had* a hwclock.service and removed it now ... no change.
>>
>> Aside from being interested if to run hwclock.service:
>>
>> solved that by entering BIOS and correcting time (was one hour behind,
>> why ever ...)
>
> It helps if the hardware clock is set to the correct time, yes. The
> only problem is if you dual boot Windows (or so I heard).

Sorry, long trip, just got home. I messed up NTP with hwclock. Anyway,
I don't handle hwclock either: it's basically included in systemd:

http://cgit.freedesktop.org/systemd/systemd/tree/src/shared/hwclock.c

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Stefan G. Weichinger
2013-03-05 07:49:03 UTC
Permalink
Am 05.03.2013 07:40, schrieb Canek Peláez Valdés:

> Sorry, long trip, just got home. I messed up NTP with hwclock. Anyway,
> I don't handle hwclock either: it's basically included in systemd:
>
> http://cgit.freedesktop.org/systemd/systemd/tree/src/shared/hwclock.c

wow, and I thought there has to be _one_ thing which is not included in
systemd ;-)

I also run ntp-client.service ... nice to have. Although I read that
timedatectl also somehow includes or controls ntp synchronization.

http://www.freedesktop.org/software/systemd/man/timedatectl.html
mentions chronyd as ntp-service.

Gotta read on.

I assume it won't make any real difference for me as a user.

Stefan
Stefan G. Weichinger
2013-03-05 07:44:16 UTC
Permalink
Am 05.03.2013 07:36, schrieb Canek Peláez Valdés:

> I don't. AFAIK, systemd provides systemd-timedated(8) since systemd
> 30:
>
> http://www.freedesktop.org/wiki/Software/systemd/timedated

Yes, found that as well yesterday.

> In normal desktops/laptops/servers, it just works.
>
>>> I *had* a hwclock.service and removed it now ... no change.
>>
>> Aside from being interested if to run hwclock.service:
>>
>> solved that by entering BIOS and correcting time (was one hour
>> behind, why ever ...)
>
> It helps if the hardware clock is set to the correct time, yes. The
> only problem is if you dual boot Windows (or so I heard).

I read that it should be preferred to registry-fix the behavior in
Windows. I will have a look sometimes ... I very rarely boot that win7
on my workstation.

So the following service-file is unnecessary and at best redundant?

# cat /etc/systemd/system/hwclock.service
[Unit]
Description=hwclock

[Service]
Type=oneshot
ExecStart=/sbin/hwclock --hctosys --localtime
ExecStop=/sbin/hwclock --systohc --localtime

[Install]
WantedBy=multi-user.target

pulled that one in from arch linux or so ...

Greets, Stefan
Stefan G. Weichinger
2013-03-09 16:23:34 UTC
Permalink
Upgrading to systemd-198 and udev-198 magically enabled me to login via
gdm again.
Nice ...

Stefan
Randy Barlow
2013-02-12 20:26:42 UTC
Permalink
On Wed, 30 Jan 2013, Canek Peláez Valdés wrote:
> GNOME 3.6 is not masked in Gentoo, just keyworded.

Apologies for the slight thread hijack, but I've been curious if anyone
knows the current state of Gnome 3 in Gentoo? I'm currently on Gnome 2,
and I'm one of those weirdos who kind of likes Gnome 3 (it's not without
its faults). I also prefer to stick with stable when I can, but I've
been curious on how things are going with stabilization of Gnome 3. It's
kind of difficult to find information about this using Google, so I
thought I'd ask if any of you knew. Thanks!

--
R
walt
2013-02-13 03:23:23 UTC
Permalink
On 02/12/2013 12:26 PM, Randy Barlow wrote:
> On Wed, 30 Jan 2013, Canek Peláez Valdés wrote:
>> GNOME 3.6 is not masked in Gentoo, just keyworded.
>
> Apologies for the slight thread hijack, but I've been curious if anyone
> knows the current state of Gnome 3 in Gentoo?

Heh. That's a complicated question, and I'm ready to give you a complicated
answer but I probably can't stay awake that long :)

Most important: are you ready to suffer long hours of learning/debugging/hair-
pulling/screaming to help guide the future of gnome3? Big question, and I'm
still trying to decide for myself after months of not deciding :)

I loved gnome2 and I wish for the good old days because every step the gnome3
devs have taken is a mystery to me. They probably have a grand vision for the
future of the gnome desktop, but it's been nothing but a PITA for me because
along the way they've removed some features from apps that I use everyday for
important stuff like, scheduling/keeping appointments on time, etc.

In my arrogant opinion, gnome3 is betaware (just-barely-beyond-alphaware) and
you should 'up'grade only if you're willing to live with that excitement.

One little tidbit of good news (IMO) is that the gnome3 devs have decided to
maintain a 'legacy' gnome desktop for those of us who don't like gnome-shell,
which I definitely don't (but Canek definitely does).

I suggest you try the latest gnome3 liveCD in a virtual machine so you don't
have to risk your current gnome2 install.

It's a big step, and one you may want to postpone if you're not ready for
hard work and big headaches.

IOW, try gnome3 on a virtual machine first :)
Canek Peláez Valdés
2013-02-13 04:13:08 UTC
Permalink
On Tue, Feb 12, 2013 at 9:23 PM, walt <***@gmail.com> wrote:
[...]
> IOW, try gnome3 on a virtual machine first :)

I think it would be easier if you tried a LiveCD:

http://www.gnome.org/getting-gnome/

For what is worth, I find myself more productive and much more at ease
with GNOME 3 than with GNOME 2 (or any other desktop, for that
matter). Every desktop/laptop computer I put my hands on, the first
thing I try to do is to press the windows key to get the overview;
it's really disappointing when it doesn't work and I realize I'm not
using GNOME 3.

I suggest to try the LiveCD; Walt seems to hate the shell and I love
it, but that's because is a matter of personal opinion. I hate KDE
(all four versions); that doesn't mean it's bad software, or that it
has the wrong design. It's just not for me.

Try it for a couple of weeks, and then decide if you want to try it;
we on the list can help you to get the keyworded version installed.

Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
Alan McKinnon
2013-02-13 10:39:38 UTC
Permalink
On 13/02/2013 06:13, Canek Peláez Valdés wrote:
> On Tue, Feb 12, 2013 at 9:23 PM, walt <***@gmail.com> wrote:
> [...]
>> IOW, try gnome3 on a virtual machine first :)
>
> I think it would be easier if you tried a LiveCD:
>
> http://www.gnome.org/getting-gnome/
>
> For what is worth, I find myself more productive and much more at ease
> with GNOME 3 than with GNOME 2 (or any other desktop, for that
> matter). Every desktop/laptop computer I put my hands on, the first
> thing I try to do is to press the windows key to get the overview;
> it's really disappointing when it doesn't work and I realize I'm not
> using GNOME 3.
>
> I suggest to try the LiveCD; Walt seems to hate the shell and I love
> it, but that's because is a matter of personal opinion. I hate KDE
> (all four versions); that doesn't mean it's bad software, or that it
> has the wrong design. It's just not for me.


Purely out of morbid curiosity, I've just spent an hour playing with the
Gnome 3 LiveCd in a VM.

What I'm seeing is a KDE4 ripoff, done badly, plus a few MacOS-isms and
some ideas from Unity:

- Highly generic launcher on the left, just like Unity.
- Click something, get big list of launchable apps in big "area" in
middle of screen, just like Unity.
- Go to settings, hey this looks *exactly* like KDE4 SystemSettings with
the useful stuff removed. Even the icon is the same!
- Contacts and IM settings integrated into the system just like KDE (but
actually that might be more Telepathy than KDE)
- Menu seems to magically morph into system taskbar, sort of like MacOS

I don't mean to rain on anyone's parade, but I really do not see the
point of Gnome3 at all. It has no identity of its own apart from the
recognizable icon set, that covers some of the look. The feel is as I
said very much copying KDE and some MacOS (just not enough MacOS to get
sued by Apple).

So what's the point of Gnome3?

If people like the Unity-ish bits, they should run Unity. Same with the
KDE and MacOS bits.

I'm happy to be shown to be wrong and to be shown where Gnome3 has merit
for being itself, where it can proudly stand on it's own. But I'm just
not seeing it yet


--
Alan McKinnon
***@gmail.com
Stroller
2013-02-13 11:36:00 UTC
Permalink
On 13 February 2013, at 10:39, Alan McKinnon wrote:
> …
> I don't mean to rain on anyone's parade, but I really do not see the
> point of Gnome3 at all. It has no identity of its own …
>
> So what's the point of Gnome3?
>
> If people like the Unity-ish bits, they should run Unity. Same with the
> KDE and MacOS bits.

Politics?

You know the history between Ubuntu / Unity and Gnome, don't you?

AIUI Ubuntu did a huge set of patches to Gnome to provide notifications, and they were rejected.

AIUI there was a big thing between Shuttleworth and some of the Gnome devs, with Shuttleworth saying that the patches were discussed with Gnome devs; Shuttleworth claimed they'd followed the Gnome devs' advisement, agreed the best way forward and notifications (and their API) had been implemented on the understanding they'd likely be accepted. However it was a couple of different Gnome devs that claimed responsibility for this area, and that they'd decided to do things differently, and that basically Ubuntu's work was an unwelcome code dump - thanks, but no thanks.

Thus Ubuntu made Unity, and Gnome carried on with doing it their way.

There were a couple of really long articles on Shuttleworth's blog about this at the time. They're actually really interesting reading, if you've got the time for an epic, an insight into the politics or "society" of OSS development. The impression I got was that there was some upset, but actually no-one had deceived anyone or stitched anyone else up, it was just a misunderstanding (or series of misunderstandings) due to the nature of the relationships / hierarchies involved in the two development groups. But I think Shuttleworth was a bit aggrieved and felt the only way to get what he wanted was to develop Unity in house, and Gnome wasn't going to stop what it was doing just because Ubuntu were doing something similar-but-different.

Stroller.
Alan McKinnon
2013-02-13 12:23:40 UTC
Permalink
On 13/02/2013 13:36, Stroller wrote:
>
> On 13 February 2013, at 10:39, Alan McKinnon wrote:
>> …
>> I don't mean to rain on anyone's parade, but I really do not see the
>> point of Gnome3 at all. It has no identity of its own …
>>
>> So what's the point of Gnome3?
>>
>> If people like the Unity-ish bits, they should run Unity. Same with the
>> KDE and MacOS bits.
>
> Politics?
>
> You know the history between Ubuntu / Unity and Gnome, don't you?
>
> AIUI Ubuntu did a huge set of patches to Gnome to provide notifications, and they were rejected.
>
> AIUI there was a big thing between Shuttleworth and some of the Gnome devs, with Shuttleworth saying that the patches were discussed with Gnome devs; Shuttleworth claimed they'd followed the Gnome devs' advisement, agreed the best way forward and notifications (and their API) had been implemented on the understanding they'd likely be accepted. However it was a couple of different Gnome devs that claimed responsibility for this area, and that they'd decided to do things differently, and that basically Ubuntu's work was an unwelcome code dump - thanks, but no thanks.
>
> Thus Ubuntu made Unity, and Gnome carried on with doing it their way.
>
> There were a couple of really long articles on Shuttleworth's blog about this at the time. They're actually really interesting reading, if you've got the time for an epic, an insight into the politics or "society" of OSS development. The impression I got was that there was some upset, but actually no-one had deceived anyone or stitched anyone else up, it was just a misunderstanding (or series of misunderstandings) due to the nature of the relationships / hierarchies involved in the two development groups. But I think Shuttleworth was a bit aggrieved and felt the only way to get what he wanted was to develop Unity in house, and Gnome wasn't going to stop what it was doing just because Ubuntu were doing something similar-but-different.


I do remember most of that, although I never read up on it in lots of
detail. It was something that was happening "over there" and I could
easily keep it out of my head space.

Unity has been shipping for ages now, what is it? 3 years at least?
And if my memory serves, Gnome 3 is *very* much more recent in a
shippable state. It certainly looks like Gnome is copy-catting, and
doing it badly.

I'm still struggling to see where Gnome figures it's going to fit in in
the world. What are the devs trying to build and what is their vision
for their project? Because I just don't see one at all. Saying things
like "we want to build a modern, functional, relevant desktop for todays
needs" is really just marketing crap, it tells you nothing. It's empty
vapid words devoid of meaning (if the sentence was a human it would be
the dumb blonde stereotype from sitcoms). Windows8 is modern,
functional, yadda yadda yadda...., for that matter so is Android Eclair.

All the progress I see from Gnome3 (and I get this only from blog posts
on the tubes) is that stuff is being ripped out and replaced with mostly
nothing. Take a file manager; I understand the concept of treating your
stuff on disk as meta-stuff and you just search for stuff, the desktop
tells you where your stuff is, even if it's in the cloud. But sometimes
the user really does want to view his stuff as actual files and folders.
So, err, where's the file manager?

Why is the system settings app a straight rip right out of KDE4? Even
the categories and names are recognizably the same. I would think
SystemSettings is the one major part of a desktop where Gnome would
*not* copy something else. If anything in a desktop needs to follow your
overall vision for the user, it would be that one.

So I dunno, I look at the project and what I'm seeing is a bunch of
folks with an aura of "we actually have no idea really what we are
doing..." I'm not saying that's the way it is, I'm saying that's the
conclusion I'm coming to based on what I see on the screen.

I'd still really like someone who groks what Gnome3 is all about to fill
in these blanks in my understanding with truthiness ;-)


--
Alan McKinnon
***@gmail.com
walt
2013-02-13 14:12:39 UTC
Permalink
On 02/13/2013 04:23 AM, Alan McKinnon wrote:
> All the progress I see from Gnome3 (and I get this only from blog posts
> on the tubes) is that stuff is being ripped out and replaced with mostly
> nothing.

That's exactly my problem with gnome3 in a sentence. I don't hate gnome-shell
as an interface, I hate it because some important (to me) stuff is missing and
if it's possible to get the missing stuff back, I don't know how to do it.

The gnome3 devs may intend to restore the missing stuff at some point, but I
don't know, and meanwhile I'm frustrated and my attitude is deteriorating.

Heh. I should just go back to gentoo stable and keep using gnome2 for a year
until I'm ready to be frustrated again -- as I surely will be soon enough.
Paul Hartman
2013-02-13 17:32:28 UTC
Permalink
On Wed, Feb 13, 2013 at 8:12 AM, walt <***@gmail.com> wrote:
> On 02/13/2013 04:23 AM, Alan McKinnon wrote:
>> All the progress I see from Gnome3 (and I get this only from blog posts
>> on the tubes) is that stuff is being ripped out and replaced with mostly
>> nothing.
>
> That's exactly my problem with gnome3 in a sentence. I don't hate gnome-shell
> as an interface, I hate it because some important (to me) stuff is missing and
> if it's possible to get the missing stuff back, I don't know how to do it.
>
> The gnome3 devs may intend to restore the missing stuff at some point, but I
> don't know, and meanwhile I'm frustrated and my attitude is deteriorating.

This is giving me flashbacks to the threads from when KDE4 was first
released, people complaining about missing features from KDE3. A good
time was had by all. At least KDE4 proved that, if you can wait a few
years, most of those beloved features will probably be reimplemented
at some point, maybe. :)
Kevin Chadwick
2013-02-13 17:56:51 UTC
Permalink
> I'd still really like someone who groks what Gnome3 is all about to fill
> in these blanks in my understanding with truthiness ;-)

Apparently the main drive is to have a brand, so a constant and so
simple look is recognised as a Gnome/? machine. A bit pointless if
no-one uses it or changes to something better (negative brand).

>> The gnome3 devs may intend to restore the missing stuff at some point, but I
>> don't know, and meanwhile I'm frustrated and my attitude is deteriorating.

Certainly not all unless they change the 'Brand' position.

--
_______________________________________________________________________

'Write programs that do one thing and do it well. Write programs to work
together. Write programs to handle text streams, because that is a
universal interface'

(Doug McIlroy)
_______________________________________________________________________
Alan McKinnon
2013-02-13 19:05:10 UTC
Permalink
On 13/02/2013 19:56, Kevin Chadwick wrote:
>> I'd still really like someone who groks what Gnome3 is all about to fill
>> in these blanks in my understanding with truthiness ;-)
>
> Apparently the main drive is to have a brand, so a constant and so
> simple look is recognised as a Gnome/? machine. A bit pointless if
> no-one uses it or changes to something better (negative brand).


Oh yes, I'd forgotten about that little jewel, the bit where a senior
Gnome dev is on record as saying that themes the user can control are
bad. And this person wants to remove (or at least tone down) themes so
Gnome can control Gnome's brand.

>
>>> The gnome3 devs may intend to restore the missing stuff at some point, but I
>>> don't know, and meanwhile I'm frustrated and my attitude is deteriorating.
>
> Certainly not all unless they change the 'Brand' position.

Who cares about a brand? Projects make software that works.

Distros do branding. Or does Gnome want every Gnome-using distro to look
the same?

Do Gnome devs know how to spell "fork"?


--
Alan McKinnon
***@gmail.com
Kevin Chadwick
2013-02-13 19:32:37 UTC
Permalink
> Do Gnome devs know how to spell "fork"?

I think not they have an accent and keep saying

'pass me the fork an knife'

Puzzled why they only got a knife they just get their heads down and
start cutting away due to the funny look from the passer.

--
_______________________________________________________________________

'Write programs that do one thing and do it well. Write programs to work
together. Write programs to handle text streams, because that is a
universal interface'

(Doug McIlroy)
_______________________________________________________________________
Kevin Chadwick
2013-02-13 17:50:44 UTC
Permalink
> I'm happy to be shown to be wrong and to be shown where Gnome3 has merit
> for being itself, where it can proudly stand on it's own. But I'm just
> not seeing it yet

I thought the following brilliant feature was obvious?

So your Gran has absolutely no chance of finding the "power off" button
so that you can spy on her bedroom TV's camera ;-)


p.s. In case your wondering, all my grans are long dead, you sick....

--
_______________________________________________________________________

'Write programs that do one thing and do it well. Write programs to work
together. Write programs to handle text streams, because that is a
universal interface'

(Doug McIlroy)
_______________________________________________________________________
Alecks Gates
2013-02-13 23:54:49 UTC
Permalink
On Wed, Feb 13, 2013 at 11:50 AM, Kevin Chadwick <***@yahoo.co.uk> wrote:
>> I'm happy to be shown to be wrong and to be shown where Gnome3 has merit
>> for being itself, where it can proudly stand on it's own. But I'm just
>> not seeing it yet
>
> I thought the following brilliant feature was obvious?
>
> So your Gran has absolutely no chance of finding the "power off" button
> so that you can spy on her bedroom TV's camera ;-)
>
>
> p.s. In case your wondering, all my grans are long dead, you sick....
>
> --
> _______________________________________________________________________
>
> 'Write programs that do one thing and do it well. Write programs to work
> together. Write programs to handle text streams, because that is a
> universal interface'
>
> (Doug McIlroy)
> _______________________________________________________________________
>

If you can't find the power off button in a modern GNOME installation
you have to be quite blind... of course, I don't even use it when I
have it, powering off from the console and all.
Kevin Chadwick
2013-02-14 00:11:59 UTC
Permalink
> If you can't find the power off button in a modern GNOME installation
> you have to be quite blind... of course, I don't even use it when I
> have it, powering off from the console and all.

I guess you haven't seen the mountains of users who didn't consider
holding ALT to change the suspend option to power off from the
desktop and why would they???

--
_______________________________________________________________________

'Write programs that do one thing and do it well. Write programs to work
together. Write programs to handle text streams, because that is a
universal interface'

(Doug McIlroy)
_______________________________________________________________________
Allan Gottlieb
2013-02-14 03:12:28 UTC
Permalink
On Wed, Feb 13 2013, Kevin Chadwick wrote:

>> If you can't find the power off button in a modern GNOME installation
>> you have to be quite blind... of course, I don't even use it when I
>> have it, powering off from the console and all.
>
> I guess you haven't seen the mountains of users who didn't consider
> holding ALT to change the suspend option to power off from the
> desktop and why would they???

That was indeed quite well hidden. But the two-step procedure, logout
then power-off, was easy to find.

allan
Yohan Pereira
2013-02-13 18:23:02 UTC
Permalink
On 13/02/13 at 12:39pm, Alan McKinnon wrote:
> Purely out of morbid curiosity, I've just spent an hour playing with the
> Gnome 3 LiveCd in a VM.
>
> What I'm seeing is a KDE4 ripoff, done badly, plus a few MacOS-isms and
> some ideas from Unity:
>
> - Highly generic launcher on the left, just like Unity.
> - Click something, get big list of launchable apps in big "area" in
> middle of screen, just like Unity.
> - Go to settings, hey this looks *exactly* like KDE4 SystemSettings with
> the useful stuff removed. Even the icon is the same!
> - Contacts and IM settings integrated into the system just like KDE (but
> actually that might be more Telepathy than KDE)

I think gnome had this before though, empathy predating kde telepathy by
quite some time. Although I have no clue on how things were < kde4.
The IM chat integreated into the shell thing was the only thing that
caught my attention in gnome3 the one time I tried it.

--

- Yohan Pereira

The difference between a Miracle and a Fact is exactly the difference
between a mermaid and a seal.
-- Mark Twain
Randy Barlow
2013-02-13 06:23:42 UTC
Permalink
Thanks for the reply! I agree that it is complicated, and that the direction of Gnome is mysterious. I've been using it at work, and I've enjoyed some things about it. Some other choices are puzzling. I'm interested to stick aro
Continue reading on narkive:
Loading...