About Time precision on influxdb v2.0rc

I used CLI comman to write csv into influxDB v2.0, and my csv file is as below.

擷取3

擷取

But all the time displayed on influxDB is 8:00, which leads to data is overwrited and only one data is displayed. The problem seems that the time data is unconditionally rounded to the hour unit, but I have specified the time precision = ns. How can I solve this problem?


擷取2

The influx write part is OK, ns precision is always expected when you have CSV data with timestamps on input. The query that is run by web UI by default aggregates the results in time windows. Click the Script Editor button in the UI to take a look of what is exactly being executed. It is probably this:

from(bucket: "my-bucket")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "test")
  |> aggregateWindow(every: v.windowPeriod, fn: mean, createEmpty: false)
  |> yield(name: "mean")

Remove the aggregation part to see the measurement points that you have written

from(bucket: "my-bucket")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "test")
1 Like