Hello,
I’m currently trying out to upload large CSV files on my Influx instance. However, one of my major problems is the fact that my InfluxDB instance doesn’t seem to return the correct values for the data point count. My schema is something like this _time,_measurement,_field,_value,location,sensor whereas location and sensor are tags which depend on each other. For instance if I have the city Berlin, I will always have the sensor LM00000. Still, with a CSV file of the size of 100.000 rows I only get exactly 658 rows. Do I have to wait for the writes to finish or did I do something wrong during uploading?
Thanks for advance for any help
Hello @influxdb_user,
It could depend on your schema.
like if you have the same timestamp for the same series you could be overwriting your data.
Ex:
my_meas, tagkey=tagkey1 fieldkey=fieldvalue1 timestamp1
my_meas, tagkey=tagkey1 fieldkey=fieldvalue2 timestamp1
Only the second line will be retained because it will overwrite the line with fieldvalue1 since the series it the same.
Could that be a cause?
How are you determining how many rows you’ve written?
I would suggest using the telegraf file plugin for this work that will allow you to more closely monitor what’s going on with your writes. You can also elect to write all the csv data in batches and verify that you’ve written a certain amount of records.
