Specify writetime timestamp column
Optionally specify the column to use for the writetime timestamp when inserting records from Kafka into supported database tables.
In the Kafka topic mapping, you can optionally specify which column should be used to note
the timestamp when the Kafka-produced record was inserted into any of the following:
- DataStax Astra cloud databases
- DataStax Enterprise (DSE) 4.7 and later databases
- Open source Apache Cassandra® 2.1 and later databases
The mapping setting provides a special property,
__timestamp
. The format
is:topic.my_topic.my_ks.my_table.mapping=col1=key.f1, col2=value.f1, __timestamp=value.f2
By default, the database internally tracks the
writetime
timestamp of
records inserted from Kafka. However, this __timestamp
feature in the mapping
supports the scenario where the Kafka records have an explicit timestamp field that you want
to use as a writetime for the database record produced by the connector.Attention: The column value that you use must be a number.
When you map the
__timestamp
value, you may also want to specify the
timeUnit
of the provided timestamp. The default is
MICROSECONDS
. For example, if your column used as the timestamp is in a
different format (for example SECONDS
) set it to SECONDS
.
The connector will automatically transform it to the database's proper numeric timestamp
format; that is, to MICROSECONDS
. Example:
topic.my_topic.my_ks.my_table.timestampTimeUnit=SECONDS