Getting 10 Minute Average From Persistence

Setup, Configuration and Use Scripts & Rules
1
  • Platform information:
    • Hardware: OpenHABian on Pi 3B
    • OS: Current
    • Java Runtime Environment:
    • openHAB version: 2.4

I am trying to get a 10 minute average of an item persisted by influxDB.

I have a rule that looks like this:

rule "Basement Dryer Update"
when
    Item SonoffDryer_AmpUsage changed
then
    logInfo("Averaging", "Dryer")
    N_Dryer10minAverage.postUpdate(SonoffDryer_AmpUsage.averageSince(now.minusMinutes(20)))
end

The value is coming from a SonoffPOW2, which I’m not sure is relevant, but anyway…

Both the updated value and the 10 minute average show the same value on the sitemap. I changed the calculation to a 20 minute average, but that too shows the same value (eliminating how the Sonoff calculates its reported value).

What am I missing? Am I wrong about how persistence should work?

2

Show your persistence config. Are you sure there’s more than one value persisted in a 10 or 20 minutes timeframe ? Enable persistence debugging to find out.

3

Thanks Marcus: here’s the config. I’ll try to figure out how to enable logging a bit later, but I do get fancy graphs in Grafana for these items:

 SonoffWashingMachine_Power : strategy = everyChange, everyMinute
 SonoffWashingMachine_Reachable : strategy = everyChange, everyMinute
 SonoffWashingMachine_WiFiSignalStrength : strategy = everyChange, everyMinute
 SonoffWashingMachine_AmpUsage : strategy = everyChange, everyMinute
   
 SonoffDryer_Power : strategy = everyChange, everyMinute
 SonoffDryer_Reachable : strategy = everyChange, everyMinute
 SonoffDryer_WiFiSignalStrength : strategy = everyChange, everyMinute 
 SonoffDryer_AmpUsage : strategy = everyChange, everyMinute
4

… and did the rule trigger ?

5

Yes, the rule is posting its logInfo…

I just enabled logging, hopefully correctly, from the Karaf console. Now I need to know where to look for stuff…

openhab> log:set DEBUG org.openhab.persistence.influxdb

Thanks again for the reply.

6

DEBUG logging seems to be working:

2019-04-01 17:19:29.582 [DEBUG] [.internal.InfluxDBPersistenceService] - got DecimalType value 0.16
2019-04-01 17:19:29.638 [DEBUG] [.internal.InfluxDBPersistenceService] - got DecimalType value 0.16

These are the original item, and the one that is intended to display the calculated average…

7

Partially solved, and another wasted evening on something that seems like it should be simple.

I had mapdb persisting all items for restore on startup as well as influxes, and I think it was averaging the single data point from there.

However, now I get a little more data - the average is just half of the actual value.

image
image.png2184×576 55 KB

Any ideas?

8

Had to change my default persistence service, and it seems to now be working.