[lm-sensors] Handling of fault files in generic print code

[lm-sensors] Handling of fault files in generic print code

[Hans de Goede]

Hi All,

I noticed that currently the temp code prints a faults in the place where 
normally an alarm gets printed, and ignores alarms. I mimicked this for the 
fault support I added to the fan code.

But is this wise, since the reading are bogus when a FAULT is raised, shouldn't 
the reading be omitted, just like were omitting the alarm?

Regards,

Hans

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Handling of fault files in generic print code

[Jean Delvare]

Hi Hans,

On Wed, 04 Jul 2007 06:56:23 +0200, Hans de Goede wrote:
> I noticed that currently the temp code prints a faults in the place where 
> normally an alarm gets printed, and ignores alarms. I mimicked this for the 
> fault support I added to the fan code.
> 
> But is this wise, since the reading are bogus when a FAULT is raised, shouldn't 
> the reading be omitted, just like were omitting the alarm?

I agree that the current implementation doesn't make much sense and
can be confusing to the users. The "FAULT" statement should replace the
(invalid) measured value rather than the ALARM flag. Can you please
propose a patch doing this?

Thanks,
-- 
Jean Delvare

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Handling of fault files in generic print code

[Hans de Goede]

Jean Delvare wrote:
> Hi Hans,
> 
> On Wed, 04 Jul 2007 06:56:23 +0200, Hans de Goede wrote:
>> I noticed that currently the temp code prints a faults in the place where 
>> normally an alarm gets printed, and ignores alarms. I mimicked this for the 
>> fault support I added to the fan code.
>>
>> But is this wise, since the reading are bogus when a FAULT is raised, shouldn't 
>> the reading be omitted, just like were omitting the alarm?
> 
> I agree that the current implementation doesn't make much sense and
> can be confusing to the users. The "FAULT" statement should replace the
> (invalid) measured value rather than the ALARM flag.

Done, see 3.0.0 svn

Regards,

Hans

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors