Just after I sent this, I happened to notice these lines in
/var/log/messages. These came from the tests with the "low=1:high=2"
attributes set in sensorsd.conf per the Undeadly example.

Nov 23 10:58:08 microserver sensorsd[6250]: upd0.indicator2: exceeds
limits: On is below On
Nov 23 10:59:54 microserver sensorsd[12047]: upd0.indicator2: exceeds
limits: On is below On
Nov 23 11:07:00 microserver sensorsd[27413]: upd0.indicator0: exceeds
limits: On is below On

On Sun, November 23, 2014 11:15 am, Joe Gidi wrote:
> Hi Marcus,
>
> Thanks for the reply. Unfortunately, the "low=1:high=2" doesn't seem to
> work for indicator2. When I start sensorsd I see an initial event logged
> as the status goes from undefined to OK, but no further events as I
> unplug/plug the UPS. I tried monitoring indicator0 as in the Undeadly
> example, and I see exactly the same behavior.
>
> It appears to me that the driver should be changing the status (%s token)
> of the indicators to something other than "OK" when the UPS loses mains
> power, but it simply doesn't.
>
> BTW, I've tested with various check interval values for sensorsd, from the
> default 20 seconds down to as low as 1 second, with no change in results.
>
> Is anyone successfully using sensorsd with upd?
>
> Thanks,
>
> Joe
>
> On Sun, November 23, 2014 4:13 am, Marcus MERIGHI wrote:
>> j...@entropicblur.com (Joe Gidi), 2014.11.23 (Sun) 01:22 (CET):
>>> I'm running OpenBSD 5.6/amd64 on my fileserver. It has an APC UPS that
>>> was
>>> previously managed with apcupsd. Since I upgraded to 5.6, the UPS now
>>> attaches as a upd device:
>>>
>>> $ dmesg | grep uhidev3
>>> uhidev3 at uhub3 port 5 configuration 1 interface 0 "APC Back-UPS ES
>>> 450
>>> FW:844.K2 .D USB FW:K2" rev 1.10/1.06 addr 2
>>> uhidev3: iclass 3/0, 123 report ids
>>> upd0 at uhidev3
>>>
>>> And it reports sensible values in hw.sensors:
>>> $ sysctl hw.sensors.upd0
>>> hw.sensors.upd0.indicator0=On (Charging), OK
>>> hw.sensors.upd0.indicator1=Off (Discharging), OK
>>> hw.sensors.upd0.indicator2=On (ACPresent), OK
>>> hw.sensors.upd0.indicator3=On (BatteryPresent), OK
>>> hw.sensors.upd0.indicator4=Off (ShutdownImminent), OK
>>> hw.sensors.upd0.percent0=79.00% (RemainingCapacity), OK
>>> hw.sensors.upd0.percent1=100.00% (FullChargeCapacity), OK
>>>
>>> So far, so good. Now, I'd like to configure sensorsd to monitor the
>>> device
>>> and invoke a script when the power goes out. I have this line in
>>> sensorsd.conf:
>>>
>>> hw.sensors.upd0.indicator2:command=/etc/sensorsd/ups.sh %s %2
>>>
>>> The ups.sh script currently just echoes the token values that it's
>>> passed
>>> to a log file.
>>>
>>> The issue I'm running into is this: the status of the sensors seems to
>>> always be "OK", even when their state changes. I can unplug the UPS
>>> from
>>> the wall and then I see this:
>>>
>>> hw.sensors.upd0.indicator0=Off (Charging), OK
>>> hw.sensors.upd0.indicator1=On (Discharging), OK
>>> hw.sensors.upd0.indicator2=Off (ACPresent), OK
>>> hw.sensors.upd0.indicator3=On (BatteryPresent), OK
>>> hw.sensors.upd0.indicator4=Off (ShutdownImminent), OK
>>> hw.sensors.upd0.percent0=76.00% (RemainingCapacity), OK
>>> hw.sensors.upd0.percent1=100.00% (FullChargeCapacity), OK
>>>
>>> We're not charging, we're discharging, AC power is not present, but
>>> none
>>> of the status indicators (the %s token) ever leaves the "OK" state. As
>>> I
>>> understand it, that lack of state change results in sensorsd doing
>>> nothing, even though the sensor's value (the %2 token, On/Off) changes.
>>>
>>> Can anyone clue me in? I feel like I must be missing something silly
>>> and
>>> obvious here.
>>
>> see here: http://undeadly.org/cgi?action=article&sid=20140320093943
>>
>> ``hw.sensors.upd0.indicator0:low=1:high=2:command=echo "who turned %2 \
>>   the lights?" | mail -s "power sensors" root''
>>
>> the trick seems to be to specify "low=1:high=2". I suppose that works
>> for indicator2, too.
>>
>> Bye, Marcus
>>
>>> !DSPAM:54712928273131330177583!
>>
>
>
> --
> Joe Gidi
> j...@entropicblur.com
>
> "You cannot buy skill." -- Ross Seyfried
>


-- 
Joe Gidi
j...@entropicblur.com

"You cannot buy skill." -- Ross Seyfried

Reply via email to