Graphing time series for durations formatted as HH:MM:SS

Grafana
1

I’m using Grafana to access data in an influxdb database. The graphs for data in the form of floats and ints is working very well. Now, I’m trying to graph a time series of daylight hours (duration) in the form of a string as HH:MM:SS. So far I have had no luck.

This is my query from the query inspector:
{
“xhrStatus”: “complete”,
“request”: {
“method”: “GET”,
“url”: “api/datasources/proxy/3/query”,
“params”: {
“db”: “hass”,
“q”: “SELECT "state" FROM "variable.daylight" WHERE time >= now() - 7d”,
“epoch”: “ms”
},
“data”: null,
“precision”: “ms”
},
“response”: {
“results”: [
{
“statement_id”: 0,
“series”: [
{
“name”: “variable.daylight”,
“columns”: [
“time”,
“state”
],
“values”: [
[
1576476000161,
“10:01:16”
],
[
1576562400082,
“10:01:00”
],
[
1576648800116,
“10:00:45”
],
[
1576685180034,
“10:00:45”
],
[
1576735200049,
“10:00:35”
]
]
}
]
}
]
}
}

or in screenshot form:
image

That results in no data displayed in the graph. If I take the query and paste it into influx, I get this:

    InfluxDB shell version: 1.7.6
    Enter an InfluxQL query
    > use hass
    Using database hass
    > SELECT "state" FROM "variable.daylight" WHERE time >= now() - 7d
    name: variable.daylight
    time                state
    ----                -----
    1576476000161782016 10:01:16
    1576562400082861824 10:01:00
    1576648800116229120 10:00:45
    1576685180034132992 10:00:45
    1576735200049216000 10:00:35

which tells me that the data is there, I just need to get the query right in Grafana. Can anyone see what I’m doing wrong?

2

Grafana graphs only int/float data, not strings. There is no way of converting strings to int/float in InfluxDB. The best option is to save duration as int [seconds] into InfluxDB directly.

3

That was my first thought too. However, in setting the units for the Y axis, I found, under Duration, a unit value of HH:MM:SS. What would that be used for?

4

=> that is only for formatting of int (number of seconds) to more human friendly format, e.g. 1=>00:00:01.

5

Makes perfect sense now, Thanks for the help.