Multiple ACPI related issues after upgrade Dapper > Edgy

errors apt-get ( linux image and acpid )

The obvious reason why acpi isn't working here is because of an upgrade problem in the support packages.

The "errors" you pulled from dmesg are meaningless, except for the acpi_cpufreq, but I think the issue there is either your system (you've done something wrong), or the acpi packages mucking things up.

Please repost the apt-get errors with LANG=C

Ben,

I don't know how to use "LANG=C", so I loged-in using english langage and repost apt-get errors and few other files. Hope it is what you asked for ....

I was able to remove the apt-get errors by doing :
sudo /etc/init.d/acpid stop
dpkg --configure -a

But I still have the errors in log files ( dmesg attached ).

And at exit, most of the time ( not every time ) the system froze with the fans going full speed instead of shut down.

Message from the kernel "ACPI: Unable to turn cooling device [dffdfdec] 'on'" still appears every 6 sec..... a lot of lines in the log files after few hours !

Looking to many forums, it seams to be quite frequent with edgy.

Dear Ubuntu developers, do you consider this as a normal issue linked to some hardwares, or can we expect a fix in a future kernel release ?

some news .

I lurned that the message "ACPI: Unable to turn cooling device [dffdfdec] 'on'" is linked to the module "thermal". I checked it disappears after "sudo modprobe -r thermal".

Something is quite strange. After a reboot , the removed thermal module is not there ( not in the list provided by lsmod ) but the error message reappears ... and disappears again after a "sudo modprobe -r thermal".

Serge.

Hello,
This is my first post here - guidance/clarification on Launchpad protocol/process welcomed :-)

Previously posted to ubuntuforums (http://ubuntuforums.org/showthread.php?p=2113284#post2113284) but received no useful responses... investigated Launchpad and this appears to be the right place(?).

Various manifestations of this issue appear to have been logged here, this thread appears the most like my experience.

I would like to understand "how is the fan control *intended* to work in edgy?"

=====
Quick Summary.
I have a problem regarding overheating/acpi control of the CPU fan.
Essentially: when the temp gets hot it seems the CPU fan cannot be activated, resulting in the temp warning siren going off and the temp rising and rising as reported in /proc files.

This (or something like it) has been reported elsewhere but I believe not properly addressed/fixed - correct me if I am wrong

Below is a full report - I hope it helps.

1) Intro:
=========
I have witnessed this a few times and have a reproducible case.
I have done some initial research (googling/forum searching) and this has been reported/discussed - however it seems people have treated it as a problem with the message, rather than a an actual problem with the acpi/fan control.

I think this is potentially a dangerous problem as it could cause overheating (and consequent hardware damage?)

Some system info:
-----------------
Running Edgy
uname -a
Linux shaft 2.6.17-10-generic #2 SMP Tue Dec 5 22:28:26 UTC 2006 i686 GNU/Linux

from: cpuinfo
------
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Pentium(R) 4 CPU 3.00GHz
stepping : 9
cpu MHz : 3007.288
cache size : 512 KB
-------

from: /proc/acpi/thermal_zone/THRM
----------------------
% more *
::::::::::::::
cooling_mode
::::::::::::::
cooling mode: active
::::::::::::::
polling_frequency
::::::::::::::
<polling disabled>
::::::::::::::
state
::::::::::::::
state: ok
::::::::::::::
temperature
::::::::::::::
temperature: 49 C
::::::::::::::
trip_points
::::::::::::::
critical (S5): 100 C
active[0]: 85 C: devices=0xdffe3bd0
---------

2) Problem description
======================
When I run something CPU intensive (i.e. wav => mp3 encoding with lame) for a period of time (a few minutes plus) I get the following symptoms:
a) CPU load at 50% (is 100% of a virtual CPU) as expected for Hyper Threading P4
b) Temp reported in /proc/acpi/thermal_zone/THRM/temperature climbs/increments steadily as job continues
c) when temp reaches 85C (the threshold in trip_points file) warnings are generated in /var/log/messages:
------------
Jan 17 18:51:59 shaft kernel: [18129602.424000] ACPI: Transitioning device [FAN] to D0
Jan 17 18:51:59 shaft kernel: [18129602.424000] ACPI: Transitioning device [FAN] to D0
Jan 17 18:51:59 shaft kernel: [18129602.424000] ACPI: Unable to turn cooling device [dffe3bd0] 'on'
Jan 17 18:52:00 shaft kernel: [18129602.732000] ACPI: Transitioning device [FAN] to D0
Jan 17 18:52:00 shaft kernel: [18129602.732000] ACPI: Transitioning device [FAN] to D0
Jan 17 18:52:00 shaft kernel: [18129602.732000] ACPI: Unable to turn cooling device [dffe3bd0] 'on'
------------

d) A minute or two later there is a...

On the same hardware I have Ubuntu 6.06LTS ( fresh install) without the error ACPI: Unable to turn cooling device and Ubuntu 6.10 ( dist-upgrade 5.10>6.06>6.10) with the error.
I'm going to attached the dmesg from both ( only the parts with different lines, up tu the error).
Hope it can help.

Hello again,

There is no sign that this log has been looked at recently.
Please can somebody post here to confirm:
  i) This has been read and considered
 ii) Assign a status to it
iii) Set expectations as to how and when it will be properly investigated.

Personal Note: :-)
This is the first bug I have logged here. Whilst using FLOSS for several years, I am aware I have contributed little to the community.
I am an ubuntu convert, and I see the fantastic effort by Canonical/Ubuntu in creating the community resources and processes as a great way to foster interest and involvement (by someone like me).

I feel that by noticing this bug, testing it, finding and demonstrating a reproducible case, carefully and sanely documenting it i am helping and in some way contributing.
I could do more... I would like to do more.
However if i submit a report and say:
 "I think this is potentially a dangerous problem as it could cause overheating (and consequent hardware damage)"
and then diligently check back to see how this is handled to find no sign of interest whatsoever, this does not encourage me to bother next time, and also undermines all the good work and effort you have put in to create a great resource (Launchpad).

If the inability to properly deal with such logs is due to lack of resource I will understand and sympathise. however you still need to properly set expectations (with people like me) so that we understand how this is intended to work.

To be clear: This is not about ME.
This is not about "My problem".
This is about the launchpad service/system, whether it works and how to get useful and valuable contributions from people (like me) who are not "on the inside" but have lots to contribute.

Please (please please) update this log in some way to show it is not ignored. And kindly set my expectations as to when & how it will be progressed.

Many thanks - Steve. :-)

Please see bug #63550 for further developments .....

Edgy is no longer a supported release, so this bug is being closed.