InfluxDB
RctMon uses the official InfluxDB Python module, which supports the "new" API introduced with InfluxDB 2.x only. Luckily, that API was backported to version 1.8 to ease transition.
Warning
As of version 0.0.1, RctMon blocks if it can't reach the InfluxDB. This will be addressed in a later release.
Configuration
In the configuration, the influxdb section needs to be configured to enable RctMon to push to an InfluxDB.
Setting |
V1.8 |
V2.x |
|---|---|---|
|
Whether to enable pushing to InfluxDB |
|
|
URL to connect to, e.g. |
|
|
|
Access token |
|
ignored |
Organization name, e.g. |
|
|
Bucket name, e.g. |
InfluxDB 1.8 setup
InfluxDB 2.x setup
Setting this up is rather complicated compared to earlier versions. Assuming you have access to an almighty admin user, let's create the full set of structures:
An organization:
influx org create -n rct-orgCreate a user in that organization:
influx user create -n rctmon -o rct-orgA bucket:
influx bucket create -n rct-inverter -o rct-orgGet the bucket id:
influx bucket list -o rct-inverterCreate an access token for the user, allowing it to write to the bucket:
influx auth create -o rct-org --user rctmon --write-bucket <bucket id from previous command>
Measurements
All measurements have a field inverter that bears the name of the inverter, the same name as displayed in the app.
rct_raw
All of the data that is received is written to this measurement immediately. Fields are used to record the different data types:
value_boolvalue_floatvalue_intvalue_string
The raw name of the frame (such as dc_conv.dc_conv_struct[1].p_dc, which is the power in watt of the second solar
generator)) is in the field name, and the corresponding OID (such as 0xAA9AA253) is in the oid field.
Some fields are seldomly or never queried appart from the start of the application, especially the string and boolean fields that are used to query the inventory of the device.