Data queried not enough under InfluxDB 1.2.0

[hongquan]

In InfluxDB 1.2.0, the data returned is chunked by 10,000 rows, including "partial":true to indicate this.
influxdb-python currently doesnot support this feature and hence, leaves missing data.

[hongquan]

To let you have idea where the "partial":true tag is, I include a screenshot here:

[nicolajkirchhof]

This is related to influxdata/influxdb#7879.
If you seek for a quick fix, try setting 'max-row-limit = 0' in your influxdb.conf to get the full data range.

[bradm-pa]

So, #418 which was merged allows you to work around this issue.

Arguably, having both chunked and chunk_size parameters feels a bit redundant, as they appear to be interdependent. Doesn't chunk_size != 0 imply chunked = True?

And there is still the issue that if you do not specify chunking and the server applies the max-row-limit, you will silently get fewer results than you expect.

Having defaults of chunked=True and chunk_size=10,000 would less surprising to the user, eliminate the issue of server overload, and have no impact on queries returning < 10,000 results.

Hmm. After further consideration, note that as of InfluxDB 1.2.2 the default has returned to max-row-limit = 0, so perhaps this becomes moot over time.