Kapacitor - Missing from point cannot aggregate

Hello everyone,

I have implemented a simple stream task based on the following TICK script :

dbrp "CCM_test"."autogen"

stream
|from()
.measurement(‘StrandFFF’)
|window()
.period(5s)
.every(5s)
|mean(‘test_field’)
.as(‘mean_test_field’)
|influxDBOut()
.database(‘CCM_test’)
.retentionPolicy(‘autogen’)
.measurement(‘StrandFFF’)
.flushInterval(20s)

This task runs successfully- when I check on the influxDB database the mean_test_filed is being pushed every 20 seconds and populated on a 5 seconds basis with no missing values.

However in the Kapacitor console I can see that a batch of 4 errors is being constantly raised at a 20 second rate :

ts=2020-04-15T09:30:23.634+02:00 lvl=error msg="failed to aggregate point in batch" service=kapacitor task_master=main task=stream_mean node=mean3 err="field test_field missing from point cannot aggregate"
ts=2020-04-15T09:30:23.636+02:00 lvl=error msg="failed to aggregate point in batch" service=kapacitor task_master=main task=stream_mean node=mean3 err="field test_field missing from point cannot aggregate"
ts=2020-04-15T09:30:23.637+02:00 lvl=error msg="failed to aggregate point in batch" service=kapacitor task_master=main task=stream_mean node=mean3 err="field test_field missing from point cannot aggregate"
ts=2020-04-15T09:30:23.637+02:00 lvl=error msg="failed to aggregate point in batch" service=kapacitor task_master=main task=stream_mean node=mean3 err="field test_field missing from point cannot aggregate"

For info the test_field field is being pushed in InfluxDB at a 10 Hz frequency.

Do you know what could cause this ?

Many thanks.