insane fan speed readings from lm_sensors

[Lonni J Friedman]

I'm using collectd's sensor plugin, which is calling out to
libsensors.so.4 (which is part of lm_sensors).

I'm using a UPS, which definitely is doing its job, as I have
semi-frequent power issues, and the UPS has shielded everything from
them.

I'd be inclined to agree that there's a hardware problem, however the
fact that this weird problem consistently happens during the same 6
hour span every time makes me think something else is causing this
behavior.

On Thu, Sep 12, 2019 at 5:44 PM Leon Goldstein <metapsych at earthlink.net> wrote:
>
> Lonni:  what app are you using to report the fan RPM's?
>
> Is the fan actually running faster when the app reports that it is
> speeding up?  I also presume you are using a good UPS with voltage
> stabilization?  I have a voltage monitor plugged into a wall socket, and
> observe that the voltage can fluctuate a bit during the day,
> particularly during the hottest time of the day when air conditioners
> are in heavy use.  That, and the fact that hot power transmission lines
> have more resistance.
>
> Assuming your case fan is tapping off of a socket on the mobo, there may
> be a voltage excursion.  It might be a good time to replace your power
> supply.
>
> On 9/12/19 8:00 PM, Lonni J Friedman wrote:
> > This isn't the CPU fan, its the chassis case fan. I'm already
> > capturing CPU activity, and its basically idle the entire time.  Keep
> > in mind, this starts happens in the early morning hours (4AM), when
> > the system isn't being actively used.  The exhaust fan on this system
> > is whisper quiet, even when running at nearly 1400RPM.
> >
> > On Thu, Sep 12, 2019 at 9:21 AM Leon Goldstein via Linux-users
> > <linux-users at linux-sxs.org> wrote:
> >> Hi Lonni:
> >>
> >> All I can suggest is run System Monitor to see the CPU % activity when
> >> the fan speed rises.
> >>
> >> Can you hear the difference in fan speed?  I have a laptop that gets
> >> very conspicuously noisier when CPU capacity is peaking, as e.g. during
> >> an update kernel installation.  (I'm using Mint 18.3 Mate.)
> >>
> >> On 9/11/19 8:39 PM, Lonni J Friedman via Linux-users wrote:
> >>> Since its so quiet here (when is it not?), I'll drop a puzzle on you
> >>> fine folks to noodle on.
> >>>
> >>> I recently started monitoring my system exhaust fan speed (RPMs),
> >>> mostly out of curiosity.  What I discovered has exposed a mystery that
> >>> I'm struggling to explain.  Every day between 04:00 & 10:00, the fan
> >>> speed readings start to rapidly oscillate between sane numbers (under
> >>> 3000 RPM) and insane numbers (over 10,000 RPM).  For the rest of the
> >>> day, the readings are perfectly normal & reasonable.  Its always
> >>> during that same time window.  There's no way this exhaust fan is even
> >>> capable of spinning at the rate that is being reported.  This has to
> >>> be an erroneous reading.
> >>>
> >>> Here's a graph of what I'm talking about:
> >>> https://imgur.com/erXGz4J
> >>>
> >>> There's nothing in dmesg or /var/log/messages to explain this
> >>> behavior.  I've looked at every cronjob running on the system for some
> >>> bizarre correlation, and can't find anything.
> >>>
> >>> I'd like to think that if this was a flaky sensor, it wouldn't be so
> >>> reliable in when it happens.  So something else must explain the
> >>> weirdness.  Anyone have ideas/suggestions?
> >>> _______________________________________________
> >>> Linux-users mailing list
> >>> Linux-users at linux-sxs.org
> >>> http://mailman.celestial.com/mailman/listinfo/linux-users
> >>>
> >> --
> >> Leon A. Goldstein
> >> System L3
> >> Linux Mint 17.3
> >>
> >> _______________________________________________
> >> Linux-users mailing list
> >> Linux-users at linux-sxs.org
> >> http://mailman.celestial.com/mailman/listinfo/linux-users
>
> --
> Leon A. Goldstein
>
> Compulab System
> Linux Mint 18.3
>