Acer Aspire needs reboot for fan to work

[luiznetto]

I have Debian 10 installed on my Acer Aspire 5315. I boot, the fan doesn't work; I reboot, it works perfectly. So I always have to boot twice for the fan to work. In my attempts to solve this problem, I have installed tlp, acpi, thermald, following recommendations from this site:

https://itsfoss.com/reduce-overheating-laptops-linux/

but it doesn't work.

While the bootloader is on, the fan is working all right; it is at the exact moment when the bootloader hands over control to the kernel that the fan stops. On the first boot, the kernel fails to turn on the fan; after reboot, then the kernel turns on the fan when it takes over from the bootloader. This is a mystery that I cannot figure out how to solve.

Can anybody please help me? I would be immensely grateful.

[pylkko]

From what you describe it is possible that the fan does not go on in the first boot because the temperature is low enough that it is not required. If you then reboot and the processor runs through all the booting stuff and the temperature is higher. Are you monitoring the temperature while this is going on? Use psensors (for example), and if the fan never turns on despite you doing cpu intensive stuff, then you have a problem. You might see some kind of messages of error in the logs. Did you check?

[luiznetto]

Yes, I am checking the temperature. I have three ways of checking the temperature. One is

Code: Select all

$ acpi -t

the other is

Code: Select all

$ cat /sys/class/thermal/thermal_zone*/temp

and the third is

Code: Select all

# tlp-stat

whose output is

Code: Select all

+++ Temperatures
CPU temp               =    77 [°C]
Fan speed              = (not available)


The last one seems to be the only reliable, because the first two give the same result as the LXDE panel temperature monitor, which remains unchanged at 40 degrees Celsius, no matter how hot the CPU gets, if the fan didn't start at the beginning. When the temperature gets to 77 degrees as above, I reboot because I fear damage to my hardware.
But experience shows me that, if the fan does not start work a few seconds after the kernel takes over from the bootloader, it never works, and I have to reboot or the system will overheat and shut down. If it does start to work a few seconds after the kernel takes over, then it works normally the whole session, and I don't have to worry about it, which usually happens following the second boot. Strange behavior, isn't it?

[pylkko]

So, did you look at the logs?

[luiznetto]

So, did you look at the logs?

What log are you talking about?

[luiznetto]

I have not been able to solve this problem yet. I still have to boot twice for the fan to work. What have I done so far to try to solve this problem:

Code: Select all

# apt-get install lm-sensors
# apt-get install hddtemp

# /sbin/sensors-detect
# sensors-detect revision $Revision$
# System: Acer Aspire 5315 [V1.27]
# Board: Acer Acadia
# Kernel: 4.19.0-5-686 i686
# Processor: Intel(R) Celeron(R) CPU 550 @ 2.00GHz (6/22/1)

This program will help you determine which kernel modules you need
to load to use lm_sensors most effectively. It is generally safe
and recommended to accept the default answers to all questions,
unless you know what you're doing.

Some south bridges, CPUs or memory controllers contain embedded sensors.
Do you want to scan for them? This is totally safe. (YES/no): y
Module cpuid loaded successfully.
Silicon Integrated Systems SIS5595...                       No
VIA VT82C686 Integrated Sensors...                          No
VIA VT8231 Integrated Sensors...                            No
AMD K8 thermal sensors...                                   No
AMD Family 10h thermal sensors...                           No
AMD Family 11h thermal sensors...                           No
AMD Family 12h and 14h thermal sensors...                   No
AMD Family 15h thermal sensors...                           No
AMD Family 16h thermal sensors...                           No
AMD Family 17h thermal sensors...                           No
AMD Family 15h power sensors...                             No
AMD Family 16h power sensors...                             No
Intel digital thermal sensor...                             Success!
    (driver `coretemp')
Intel AMB FB-DIMM thermal sensor...                         No
Intel 5500/5520/X58 thermal sensor...                       No
VIA C7 thermal sensor...                                    No
VIA Nano thermal sensor...                                  No

Some Super I/O chips contain embedded sensors. We have to write to
standard I/O ports to probe them. This is usually safe.
Do you want to scan for Super I/O sensors? (YES/no): y
Probing for Super-I/O at 0x2e/0x2f
Trying family `National Semiconductor/ITE'...               No
Trying family `SMSC'...                                     No
Trying family `VIA/Winbond/Nuvoton/Fintek'...               No
Trying family `ITE'...                                      No
Probing for Super-I/O at 0x4e/0x4f
Trying family `National Semiconductor/ITE'...               No
Trying family `SMSC'...                                     No
Trying family `VIA/Winbond/Nuvoton/Fintek'...               No
Trying family `ITE'...                                      No

Some systems (mainly servers) implement IPMI, a set of common interfaces
through which system health data may be retrieved, amongst other things.
We first try to get the information from SMBIOS. If we don't find it
there, we have to read from arbitrary I/O ports to probe for such
interfaces. This is normally safe. Do you want to scan for IPMI
interfaces? (YES/no): y
Probing for `IPMI BMC KCS' at 0xca0...                      No
Probing for `IPMI BMC SMIC' at 0xca8...                     No

Some hardware monitoring chips are accessible through the ISA I/O ports.
We have to write to arbitrary I/O ports to probe them. This is usually
safe though. Yes, you do have ISA I/O ports even if you do not have any
ISA slots! Do you want to scan the ISA I/O ports? (YES/no): y
Probing for `National Semiconductor LM78' at 0x290...       No
Probing for `National Semiconductor LM79' at 0x290...       No
Probing for `Winbond W83781D' at 0x290...                   No
Probing for `Winbond W83782D' at 0x290...                   No

Lastly, we can probe the I2C/SMBus adapters for connected hardware
monitoring devices. This is the most risky part, and while it works
reasonably well on most systems, it has been reported to cause trouble
on some systems.
Do you want to probe the I2C/SMBus adapters now? (YES/no): n

Now follows a summary of the probes I have just done.
Just press ENTER to continue:

Driver `coretemp':
  * Chip `Intel digital thermal sensor' (confidence: 9)

To load everything that is needed, add this to /etc/modules:
#----cut here----
# Chip drivers
coretemp
#----cut here----
If you have some drivers built into your kernel, the list above will
contain too many modules. Skip the appropriate ones!

Do you want to add these lines automatically to /etc/modules? (yes/NO)n

Unloading cpuid... OK

# apt-get install psensor
# apt-get install laptop-mode-tools
# apt-get install thermald


I thought thermald would solve the problem, but it hasn't. Apparently, I have to write a file
/etc/thermald/thermal-conf.xml
but I don't know how to write it. Does anyone have any idea?

There is some data here that can be of interest:

Code: Select all

root@debian:/home/luiz# thermald --no-daemon --loglevel=debug
bash: thermald: command not found
root@debian:/home/luiz# /sbin/thermald --no-daemon --loglevel=debug
NO RAPL sysfs present
10 CPUID levels; family:model:stepping 0x6:16:1 (6:22:1)
 Need Linux PowerCap sysfs
failed to open /dev/acpi_thermal_rel
failed to open /dev/acpi_thermal_rel
TRT/ART read failed
I/O warning : failed to load external entity "/etc/thermald/thermal-conf.xml"
error: could not parse file /etc/thermald/thermal-conf.xml
Unsupported cpu model, use thermal-conf.xml file or run with --ignore-cpuid-check
THD engine start failed


I don't know why so many commands in this new Debian (10) have to be preceded by /sbin or whatever for them to run, when previously the system seemed to know the path to the command. Is it a security measure?

Another thing that I did: I created a file /etc/modprobe.d/acerhdf.conf with the following content:

Code: Select all

options acerhdf kernelmode=1 interval=10 fanon=40000 fanoff=35000


[stevepusser]

You must now become root with "su -" to have /sbin in root's path. That's in the Buster release notes.

Have you tried any other kernel versions to see if it's just a quirk in the particular kernel you're running?

[luiznetto]

How the problem was finally solved:

Right after boot, at the login prompt, all I have to do is hit "suspend", wait a few seconds and re-awake the system by pushing the power button. After this, the fan will work normally. Here is where I got the hint from:

https://bugs.launchpad.net/ubuntu/+sour ... ug/1471530

As the guy says

What I would like is to have the system in state 4 (fan turning and adjusting to temperature) without requiring human intervention, either by avoiding the fan stopping in the first place (preferred) or by re-starting it automated sometime in the boot process.

But I am not so ambitious. As long as I found a successful workaround for this problem that has been bothering me for such a long time, I consider myself satisfied. There was no way I could have guessed that the solution was so easy; normally I never use either suspend or hibernate, I didn't even know what those things are good for.

I should also note that thermal control is nearly perfect now, more sensitive to temperature changes. When I had to reboot, after the reboot the temperature indicator on the LXDE panel would show a single value, and remain at that value throughout the sesssion. Now the temperature remains usually between 45-55 degrees Celsius, causing less stress to my hardware, at the same time as the fan makes less noise.

Thanks a lot for your help.

[luiznetto]

I suspect thermald is not doing anything useful and can be safely uninstalled, even though I feel little inclination to do so. When things are working, I'd rather not touch.

[oswaldkelso]

I to have a 5315 but rarely reboot it. Usually it's a power outage that does that. I just checked and the uptime is nearly 26 days and the fan is whirring away gently. I suspect because of how its used. The kernel version is newer than the ubuntu bug report but older than Buster. Anyway thanks for the heads up it's people documenting these things that make everyone's life just a wee bit easier.