For AI agents: a documentation index is available at https://www.mongodb.com/docs/llms.txt — markdown versions of all pages are available by appending .md to any URL path.
Use the following configuration settings to configure startup of the MongoDB Kafka source connector to convert MongoDB collections into Change Stream events.
Tip
For an example using the copy existing feature, see the Copy Existing Data Usage Example.
Description: Specifies how the connector should start up when there is no
source offset available. Resuming a change stream requires a
resume token, which the connector gets from the source offset.
If no source offset is available, the connector may either
ignore all or some of the existing source data, or may at first
copy all existing source data and then continue with processing
new data.
If startup.mode=latest, the connector ignores all existing
source data.
If startup.mode=timestamp, the connector
actuates startup.mode.timestamp.* properties. If no
properties are configured, timestamp is equivalent to
latest.
If startup.mode=copy_existing, the connector
copies all existing source data to Change Stream events. This
setting is equivalent to the deprecated setting copy.existing=true.
If any system changes the data in the database while the source connector converts existing data from it, MongoDB may produce duplicate change stream events to reflect the latest changes. Since the change stream events on which the data copy relies are idempotent, the copied data is eventually consistent, conformant to an "at-least-once" delivery guarantee.
An integer number of seconds since the Epoch in decimal format (for example, 30)
An instant in the ISO-8601 format with one second precision (for example, 1970-01-01T00:00:30Z)
A BSON Timestamp in the canonical extended JSON (v2) format (for example, {"$timestamp": {"t": 30, "i": 0}})
startup.mode.copy.existing.namespace.regex
Type: string
Description: Regular expression the connector uses to match namespaces from
which to copy data. A namespace describes the MongoDB database name
and collection separated by a period (for example, databaseName.collectionName).
For example, the following regular-expression setting matches
collections that start with "page" in the stats database:
The \ character in the example above escapes the . character
that follows it in the regular expression. For more information on
how to build regular expressions, see
Patterns
in the Java API documentation.
Default: "" Accepted Values: A valid regular expression
startup.mode.copy.existing.pipeline
Type: string
Description: An inline array of pipeline operations
the connector runs when copying existing data. You can use this
setting to filter the source collection and improve the use of
indexes in the copying process.
For example, the following setting uses the $match
aggregation operator to instruct the connector to copy only
documents that contain a closed field with a value of false.