[gpm] debugging pm-utils/g-p-m from utopia updates?

["Paul Johnson" <pauljohn32 gmail com>]

Maybe this is just a simple "do you see this problem too" message.
Or, "how can we get the older RPMs from the Utopia server?

I run FC5 and have been testing the RPMs offered the utopia server for
about 9 days.  Suspend to RAM was "just working" as we discussed in
this list earlier this week, although I still don't know exactly why.
(Apparently, while we weren't looking, some big improvements happened
in FC5 kernel, because USB modules suspend and resume without any
special maneuvering.)  And then yum update installed:

May 21 07:03:40 Updated: PolicyKit.i386 0.1-cvs20060509rhughes
May 21 07:03:45 Updated: hal.i386 0.5.8-cvs20060515rhughes
May 21 07:04:03 Updated: gnome-power-manager.i386 2.15.3-cvs20060515rhughes
May 21 07:04:04 Updated: hal-devel.i386 0.5.8-cvs20060515rhughes
May 21 07:04:06 Updated: hal-gnome.i386 0.5.8-cvs20060515rhughes
May 21 07:04:08 Updated: pm-utils.i386 0.18.0-cvs20060501rhughes

After these install, suspend to RAM does not work anymore.  If I run
pm-suspend from terminal, a little window pops up saying  "unknown
error".  If I try the suspend in the g-p-m pulldown menu, I see a
balloon popup saying "Suspend Problem" and then the screensaver login
window appears.

The end of /var/log/messages has:

May 21 16:56:27 pols125 gnome-power-manager: Susspending computer
because user clicked suspend from tray menu
May 21 16:56:27 pols125 gnome-power-manager: Resuming computer
May 21 16:56:27 pols125 gnome-power-manager: Screen dim because idle
mode started
May 21 16:56:34 pols125 gnome-power-manager: Screen resume because
idle mode ended

The end of /var/log/acpid has:
[Sun May 21 16:46:19 2006] starting up
[Sun May 21 16:46:19 2006] 3 rules loaded
[Sun May 21 16:46:24 2006] client connected from 2194[68:68]
[Sun May 21 16:46:24 2006] 1 client rule loaded

Note there is nothing added to acpid when the suspend is attempted.

If you give me some hints about how to debug this, I can make a full report.

Maybe I have a hal problem?  I don't think so.  I killed hald and
re-started it like this:

/usr/sbin/hald --daemon=yes --verbose=yes --use-syslog

After hald restarts, there are many messages in /var/log/messages,
ending like this:

May 21 17:16:10 pols125 hald[3207]: 17:16:10.925 [I] hald.c:616:
Device probing completed
May 21 17:16:10 pols125 hald[3207]: 17:16:10.925 [I] hald_dbus.c:3253: entering
May 21 17:16:10 pols125 hald[3207]: 17:16:10.935 [I] hald_dbus.c:3197:
3207: Got a connection
May 21 17:16:10 pols125 hald[3207]: 17:16:10.935 [I] hald_dbus.c:3198:
dbus_connection_get_is_connected = 1
May 21 17:16:11 pols125 hald[3207]: 17:16:11.014 [W] hald_dbus.c:89:
No property battery.remaining_time on device with id
/org/freedesktop/Hal/devices/acpi_BAT0
May 21 17:16:11 pols125 hald[3207]: 17:16:11.025 [W] hald_dbus.c:89:
No property info.vendor on device with id
/org/freedesktop/Hal/devices/acpi_BAT0

Then when I try pm-suspend, it still fails with "unknown error", but
no new messages are added in /var/log/messages.

Doesn't that mean hal is working well?

--
Paul E. Johnson
Professor, Political Science
1541 Lilac Lane, Room 504
University of Kansas