Google BigQuery Sink V2 Connector for Confluent Cloud

You can use the Kafka Connect Google BigQuery Sink V2 connector for Confluent Cloud to export Avro, JSON Schema, Protobuf, or JSON (schemaless) data from Apache Kafka® topics to BigQuery. The BigQuery table schema is based upon information in the Apache Kafka® schema for the topic.

Confluent Cloud is available through Google Cloud Marketplace or directly from Confluent.

With version 2 of the Google BigQuery Sink connector, you can perform upsert and delete actions on ingested data. Additionally, this version allows you to use the Storage Write API to ingest data directly into BigQuery tables. For more details, see the following section.

Features

The connector supports the following functionalities for data ingestion:

Upsert functionality: With the upsert functionality, you can insert new data or update existing matching key data. For more details, see Stream table updates with change data capture.
Upsert and delete functionality: With the upsert and delete functionality, you can insert new data, update existing matching key data, or delete matching key data for tombstone records. For more details, see Stream table updates with change data capture.

Google Cloud BigQuery Storage Write API. The BigQuery Storage Write API combines streaming ingestion and batch loading into a single high-performance API. For more information, see Batch load and stream data with BigQuery Storage Write API. Note that using the Storage Write API may provide a cost-benefit for your BigQuery project. Also, note that BigQuery API quotas apply.

Review the following table for details about the differences between using BATCH LOADING versus STREAMING mode with the BigQuery API. For more information, see Introduction to loading data.

BATCH LOADING	STREAMING
Records are available after the commit interval has expired and may not be provided in real time	Records are available immediately after an append call (minimal latency) in near real time
Requires creating specific application streams to handle the records	Uses the default stream
May cost less	May cost more
Multiple API quota limits with more restrictions (for example, maximum number of streams and buffering)	Lower quota with fewer restrictions

Provider integration support: The connector supports Google Cloud’s native identity authorization using Confluent Provider Integration. For more information about provider integration setup, see the connector authentication.
The connector supports OAuth 2.0 for connecting to BigQuery. Note that OAuth is only available when creating a connector using the Cloud Console. OAuth 2.0 support is available in two variations:
- Shared app: Use the application provided by Confluent, shared with all users, for an easy and quick start without the hassle of application management.
- Bring your own app: Register and manage your own connected application on the OAuth platform, maintaining control over your authentication setup.
The connector supports streaming from a list of topics into corresponding tables in BigQuery.
Even though the connector streams records by default (as opposed to running in batch mode), the connector is scalable because it contains an internal thread pool that allows it to stream records in parallel. The internal thread pool defaults to 10 threads. Note that this is only applicable for BATCH LOADING and STREAMING mode and not UPSERT and UPSERT_DELETE mode.
The connector supports several time-based table partitioning strategies.
The connector supports routing invalid records to the DLQ. This includes any records that have gRPC status code INVALID_ARGUMENT from the BigQuery Storage Write API.
Note
DLQ routing does not work if Auto update schemas (auto.update.schemas) is enabled and the connector detects that the failure is due to schema mismatch.
The connector supports default_missing_value_interpretation. For more information about default value interpretation for missing value, see Google BigQuery default value settings. Contact Confluent Support to enable default value settings in your connector.
The connector supports Avro, JSON Schema, Protobuf, or JSON (schemaless) input data formats. Schema Registry must be enabled to use a Schema Registry-based format (for example, Avro, JSON_SR (JSON Schema), or Protobuf). See Schema Registry Enabled Environments for additional information.
For Avro, JSON_SR, and Protobuf, the connector provides the following configuration properties that support automated table creation and schema updates. You can select these properties in the UI or add them to the connector configuration, if using the Confluent CLI.
- auto.create.tables: Automatically create BigQuery tables if they don’t already exist. The connector expects that the BigQuery table name is the same as the topic name. If you create the BigQuery tables manually, make sure the table name matches the topic name. This property adds/updates Kafka record keys as primary keys if ingestion.mode is set to UPSERT or UPSERT_DELETE. Note that you must adhere to the BiqQuery’s primary key constraints.
- auto.update.schemas: Automatically update BigQuery schemas. Note that new fields are added as NULLABLE in the BigQuery schema. This property adds/updates Kafka record keys as primary keys if ingestion.mode is set to UPSERT or UPSERT_DELETE. Note that you must adhere to the BiqQuery’s primary key constraints.
- sanitize.topics: Automatically sanitize topic names before using them as BigQuery table names. If not enabled, topic names are used as table names. If enabled, the table names created may be different from the topic names.
- sanitize.field.names Automatically sanitize field names before using them as column names in BigQuery.
- sanitize.field.names.in.array: Automatically sanitize field names inside array-type objects before using them as column names in BigQuery.
Note
New tables and schema updates may take a few minutes to be detected by the Google Client Library. For more information see the Google Cloud BigQuery API guide.
Clustering Support: Clustering organizes data within a table based on one or more specified columns. The order of the clustered columns determines the sort order of the data, thus enhancing the query performance. Clustering is beneficial for queries that include filter clauses or aggregate large datasets, as it reduces the amount of data scanned.
The topic2ClusteringFieldsMap configuration allow users to define clustering columns for each topic. Clustering is only applicable when auto.create.table is enabled and applies only to tables created through the connector. For more information, check the following resources:
- For limits on clustering, see BigQuery Clustering Limitations
- For supported data types, see BigQuery Clustering Data Types
_CHANGE_SEQUENCE_NUMBER in CDC mode: The connector supports the _CHANGE_SEQUENCE_NUMBER pseudo-column to maintain correct record ordering in CDC upsert mode. For identical primary keys, the record with the higher sequence number takes precedence. If both primary key and sequence number are identical, the later ingested record is retained. Note that _CHANGE_SEQUENCE_NUMBER is a pseudo-column and is not included in the BigQuery table schema. When enabled, this is an optional column and the value for this needs to be added in the data. Contact Confluent Support to enable the feature in your connector.
Note
- To enable or disable this feature in your connector, contact Confluent Support.
- Disabling the feature will not remove _CHANGE_SEQUENCE_NUMBER values from existing records. To remove these values, contact GCP Support.

For more information and examples to use with the Confluent Cloud API for Connect, see the Confluent Cloud API for Connect Usage Examples section.

Limitations

Be sure to review the following information.

For connector limitations, see Google BigQuery Sink V2 Connector limitations.
If you plan to use OAuth 2.0 for connecting to BigQuery, see OAuth Limitations.
If you plan to use one or more Single Message Transforms (SMTs), see SMT Limitations.
If you plan to use Confluent Cloud Schema Registry, see Schema Registry Enabled Environments.
If you plan to consume schemaless JSON, contact Confluent Support.

Supported data types

The following sections describe supported BigQuery data types and the associated connector mapping.

General mapping

The following is a general mapping of supported data types for the Google BigQuery Sink V2 connector. Note that this mapping applies even if a table is created or updated outside of the connector lifecycle. For automatic table creation and schema update mapping, see the following section.

BigQuery Data Type	Connector Mapping	Conditions
JSON	STRING
GEOGRAPHY	STRING
INTEGER	STRING
INTEGER	INT32
INTEGER	INT64
FLOAT	INT8
FLOAT	INT16
FLOAT	INT32
FLOAT	INT64
FLOAT	STRING
FLOAT	FLOAT32
FLOAT	FLOAT64
BOOL	BOOLEAN
BOOL	STRING	“true” or “false”
BYTES	BYTES
STRING	STRING
BIGNUMERIC	INT16
BIGNUMERIC	INT32
BIGNUMERIC	INT64
BIGNUMERIC	STRING
BIGNUMERIC	FLOAT32
BIGNUMERIC	FLOAT64
NUMERIC	INT16
NUMERIC	INT32
NUMERIC	INT64
NUMERIC	STRING
NUMERIC	FLOAT32
NUMERIC	FLOAT64
DATE	STRING	YYYY-MM-DD
DATE	INT32	Number of days since epoch. The valid range is -719162 (0001-01-01) to 2932896 (9999-12-31).
DATE	INT64	Number of days since epoch. The valid range is -719162 (0001-01-01) to 2932896 (9999-12-31).
DATETIME	STRING	YYYY-MM-DD[t\|T]HH:mm:ss[.F]
TIMESTAMP	STRING	YYYY-MM-DD HH:mm:SS[.F]
TIMESTAMP	INT64	microseconds since epoch
TIME	STRING	HH:mm:SS[.F]
TIMESTAMP	Logical TIMESTAMP
TIME	Logical TIME
DATE	Logical DATE
DATE	Debezium Date
TIME	Debezium MicroTime
TIME	Debezium Time
TIMESTAMP	Debezium MicroTimestamp
TIMESTAMP	Debezium TIMESTAMP
TIMESTAMP	Debezium ZonedTimestamp

Table creation and schema update mapping

The following mapping applies if the connector creates a table automatically or updates schemas (that is, if either auto.create.tables or auto.update.schemas is not disabled).

BigQuery Data Type	Connector Mapping
STRING	String
FLOAT	INT8
FLOAT	INT16
INTEGER	INT32
INTEGER	INT64
FLOAT	FLOAT32
FLOAT	FLOAT64
BOOLEAN	Boolean
BYTES	Bytes
TIMESTAMP	Logical TIMESTAMP
TIME	Logical TIME
DATE	Logical DATE
FLOAT	Logical Decimal
DATE	Debezium Date
TIME	Debezium MicroTime
TIME	Debezium Time
TIMESTAMP	Debezium MicroTimestamp
TIMESTAMP	Debezium TIMESTAMP
TIMESTAMP	Debezium ZonedTimestamp

Note

Enable use.integer.for.int8.int16 configuration property to store INT8 and INT16 as INTEGER in BigQuery.

Quick Start

Use this quick start to get up and running with the Confluent Cloud Google BigQuery Storage API Sink connector. The quick start provides the basics of selecting the connector and configuring it to stream events to a BigQuery data warehouse.

Prerequisites

An active Google Cloud account with authorization to create resources.
A BigQuery project is required. The project can be created using the Google Cloud Console. A BigQuery dataset is required in the project.

To use the Storage Write API, the connector user account authenticating with BigQuery must have bigquery.tables.updateData permissions. The minimum permissions are:

bigquery.datasets.get
bigquery.tables.create
bigquery.tables.get
bigquery.tables.getData
bigquery.tables.list
bigquery.tables.update
bigquery.tables.updateData

You must create a BigQuery table before using the connector, if you leave Auto create tables (auto.create.tables) disabled (the default).
You may need to create a schema in BigQuery, depending on how you set the Auto update schemas property (auto.update.schemas).
- Auto update schemas set to ADD NEW FIELDS: You do not have to create a schema.
- Auto update schemas set to DISABLED (the default): You must create a schema in BigQuery (as shown below). The connector does not automatically update the table.

Kafka cluster credentials. The following lists the different ways you can provide credentials.
- Enter an existing service account resource ID.
- Create a Confluent Cloud service account for the connector. Make sure to review the ACL entries required in the service account documentation. Some connectors have specific ACL requirements.
- Create a Confluent Cloud API key and secret. To create a key and secret, you can use confluent api-key create or you can autogenerate the API key and secret directly in the Cloud Console when setting up the connector.

The Confluent CLI installed and configured for the cluster.
Schema Registry must be enabled to use a Schema Registry-based format (for example, Avro, JSON_SR (JSON Schema), or Protobuf). For more information, see Schema Registry Enabled Environments.

Using the Confluent Cloud Console

Step 1: Launch your Confluent Cloud cluster

To create and launch a Kafka cluster in Confluent Cloud, see Create a kafka cluster in Confluent Cloud.

Step 2: Add a connector

In the left navigation menu, click Connectors. If you already have connectors in your cluster, click + Add connector.

Step 3: Select your connector

Click the Google BigQuery Sink V2 connector card.

Step 4: Enter the connector details

Note

Be sure you have all the prerequisites completed.
An asterisk ( * ) in the Cloud Console designates a required entry.

At the Add Google BigQuery Sink V2 screen, complete the following:

Topic selection

If you’ve already added Kafka topics, select the topics you want to connect from the Topics list.

To create a new topic, click +Add new topic.

Kafka credentials

Select the way you want to provide Kafka Cluster credentials. You can choose one of the following options:
- My account: This setting allows your connector to globally access everything that you have access to. With a user account, the connector uses an API key and secret to access the Kafka cluster. This option is not recommended for production.
- Service account: This setting limits the access for your connector by using a service account. This option is recommended for production.
- Use an existing API key: This setting allows you to specify an API key and a secret pair. You can use an existing pair or create a new one. This method is not recommended for production environments.
Note
Freight clusters support only service accounts for Kafka authentication.

Click Continue.

Authentication

Select the Authentication method.
- OAuth 2.0 (Shared app): (Default) Select this option to use shared OAuth 2.0 credentials. Click Continue with Google Cloud to select the account credentials to use.
- OAuth 2.0 (Bring your own app): Select this option to use client credentials you configure for OAuth 2.0. Click Continue with Google Cloud to select the account credentials to use. To configure Google OAuth 2.0 to bring your own app, see Set up Google OAuth (Bring your own app).
  - Client ID: Enter the OAuth client ID returned when setting up OAuth 2.0.
  - Client secret: Enter the client secret for the client ID.
- Google Cloud service account: Upload your GCP credentials file. This is a Google Cloud service account JSON file with write permissions for BigQuery. For additional details, see Create a Service Account.
- Google service account impersonation: Choose an existing integration name under Provider integration name dropdown that has access to your resource. For more information, see Manage a Google Cloud Provider Integration.
In the Project ID field, enter the ID for the Google Cloud project where BigQuery is located.
Enter the name for the BigQuery dataset the connector writes to in the Dataset field.
If using a Service Account, click Continue. If using OAuth 2.0, select your Google account, and choose whether to use your account information for the connector to authenticate with BigQuery.

Configuration

Select the Ingestion Mode: STREAMING, BATCH LOADING, UPSERT, or UPSERT_DELETE. Defaults to STREAMING.

Review the following table for details about the differences between using BATCH LOADING versus STREAMING mode with the BigQuery API. For more information, see Introduction to loading data.

BATCH LOADING	STREAMING
Records are available after a stream commit and may not be real time	Records are available for reading immediately after append call (minimal latency)
Requires creation of application streams	Default stream is used
May cost less	May cost more
More API quota limits (max streams, buffering, etc.)	Less quota limits

Note that BigQuery API quotas apply.

Select the Input Kafka record value format (data coming from the Kafka topic): AVRO, JSON_SR (JSON Schema), or PROTOBUF. A valid schema must be available in Schema Registry to use a schema-based message format (for example, Avro, JSON_SR (JSON Schema), or Protobuf). See Schema Registry Enabled Environments for additional information.
Select the Input Kafka record key format: Sets the data format for incoming record keys. Valid entries are: Avro, Bytes, JSON, JSON Schema, Protobuf, or String. A valid schema must be available in Schema Registry to use a schema-based message format.
Note
If the ingestion.mode is set to UPSERT or UPSERT_DELETE, you must set the record keys to one of the following formats: Avro, JSON, JSON Schema, or Protobuf.
(Optional) Enable Client-Side Field Level Encryption for data encryption. Specify a Service Account to access the Schema Registry and associated encryption rules or keys with that schema. For more information on CSFLE setup, see Manage CSFLE for connectors.

Data encryption

(Optional) Enable Client-Side Field Level Encryption for data decryption. Specify a Service Account to access the Schema Registry and associated encryption rules or keys with that schema. Select the connector behavior (ERROR or NONE) on data decryption failure. If set to ERROR, the connector fails and writes the encrypted data in the DLQ. If set to NONE, the connector writes the encrypted data in the target system without decryption. For more information on CSFLE setup, see Manage CSFLE for connectors.
Show advanced configurations
- Schema context: Select a schema context to use for this connector, if using a schema-based data format. This property defaults to the Default context, which configures the connector to use the default schema set up for Schema Registry in your Confluent Cloud environment. A schema context allows you to use separate schemas (like schema sub-registries) tied to topics in different Kafka clusters that share the same Schema Registry environment. For example, if you select a non-default context, a Source connector uses only that schema context to register a schema and a Sink connector uses only that schema context to read from. For more information about setting up a schema context, see What are schema contexts and when should you use them?.
- Commit interval: Used only when BATCH LOADING (ingestion.mode) is enabled. Enter an interval, in seconds, in which the connector attempts to commit records. You can set this to a minimum of 60 seconds up to 14,400 seconds (four hours). BigQuery API quotas apply.
- Kafka Topic to BigQuery Table Map: Map of topics to tables (optional). The required format is comma-separated tuples. For example, <topic-1>:<table-1>,<topic-2>:<table-2>,... Note that a topic name must not be modified using a regex single message transform (SMT) if using this option. Note that if this property is used, sanitize.topics is ignored. Also, if the topic-to-table map doesn’t contain the topic for a record, the connector creates a table with the same name as the topic name.
- Sanitize topics: Specifies whether to automatically sanitize topic names before using them as table names in BigQuery. If not enabled, topic names are used as table names. If enabled, the table names created may be different from the topic names.
- Sanitize field names: Specifies whether to automatically sanitize field names before using them as field names in BigQuery.
- Auto update schemas: Defaults to DISABLED. Designates whether or not to automatically update BigQuery schemas. If ADDNEW FIELDS is selected, new fields are added with mode NULLABLE in the BigQuery schema. Note that this property is applicable for AVRO, JSON_SR, and PROTOBUF message formats only.
- Auto create tables: Designates whether to automatically create BigQuery tables. Defaults to DISABLED. Note that this property is applicable for AVRO, JSON_SR, and PROTOBUF message formats only. The other options available are listed below.
  - NON-PARTITIONED: The connector creates non-partitioned tables.
  - PARTITION by INGESTION TIME: The connector creates tables partitioned by ingestion time. Uses Partitioning type to set partitioning.
  - PARTITION by FIELD: The connector creates tables partitioned using a field in a Kafka record value. The field name is set using the property Timestamp partition field name (timestamp.partition.field.name). Uses Partitioning type to set partitioning.
  New tables and schema updates may take a few minutes to be detected by the Google Client Library. For more information see the Google Cloud BigQuery API guide.
- Partitioning type: The time partitioning type to use when creating new partitioned tables. Existing tables are not altered to use this partitioning type. Defaults to DAY.
- Time partitioning field name: The name of the field in the value that contains the timestamp to partition by in BigQuery. Used when PARTITION by FIELD is used.
- Value Converter Reference Subject Name Strategy: Applicable only for PROTOBUF format. Sets the subject reference name strategy for values. Valid entries are DefaultReferenceSubjectNameStrategy or QualifiedReferenceSubjectNameStrategy. Defaults to DefaultReferenceSubjectNameStrategy.
Auto-restart policy
- Enable Connector Auto-restart: Control the auto-restart behavior of the connector and its task in the event of user-actionable errors. Defaults to true, enabling the connector to automatically restart in case of user-actionable errors. Set this property to false to disable auto-restart for failed connectors. In such cases, you would need to manually restart the connector.
Consumer configuration
- Max poll interval(ms): Set the maximum delay between subsequent consume requests to Kafka. Use this property to improve connector performance in cases when the connector cannot send records to the sink system. The default is 300,000 milliseconds (5 minutes).
- Max poll records: Set the maximum number of records to consume from Kafka in a single request. Use this property to improve connector performance in cases when the connector cannot send records to the sink system. The default is 500 records.
Transforms
- Single Message Transforms: To add a new SMT, see Add transforms. For more information about unsupported SMTs, see Unsupported transformations.
Click Continue.

Sizing

Based on the number of topic partitions you select, you will be provided with a recommended number of tasks.

To change the number of recommended tasks, enter the number of tasks for the connector to use in the Tasks field.
Click Continue.

Review and launch

Verify the connection details.
Click Launch.
The status for the connector should go from Provisioning to Running.

Step 5: Check the results in BigQuery

From the Google Cloud Console, go to your BigQuery project.
Query your datasets and verify that new records are being added.

For more information and examples to use with the Confluent Cloud API for Connect, see the Confluent Cloud API for Connect Usage Examples section.

Tip

When you launch a connector, a Dead Letter Queue topic is automatically created. See View Connector Dead Letter Queue Errors in Confluent Cloud for details.

Using the Confluent CLI

Complete the following steps to set up and run the connector using the Confluent CLI.

Note

Be sure you have all the prerequisites completed.

Step 1: List the available connectors

Enter the following command to list available connectors:

confluent connect plugin list

Step 2: List the connector configuration properties

Enter the following command to show the connector configuration properties:

confluent connect plugin describe <connector-plugin-name>

The command output shows the required and optional configuration properties.

Step 3: Create the connector configuration file

Create a JSON file that contains the connector configuration properties. The following example shows example connector properties.

{
    "name" : "confluent-bigquery-storage-sink",
    "connector.class" : "BigQueryStorageSink",
    "kafka.auth.mode": "KAFKA_API_KEY",
    "kafka.api.key" : "<my-kafka-api-key>",
    "kafka.api.secret" : "<my-kafka-api-secret>",
    "keyfile" : "....",
    "project" : "<my-BigQuery-project>",
    "datasets" : "<my-BigQuery-dataset>",
    "ingestion.mode" : "STREAMING"
    "input.data.format" : "AVRO",
    "auto.create.tables" : "DISABLED"
    "sanitize.topics" : "true"
    "sanitize.field.names" : "false"
    "sanitize.field.names.in.array" : "false"
    "tasks.max" : "1"
    "topics" : "pageviews",
}

Note the following property definitions:

"name": Sets a name for the new connector.
"connector.class": Identifies the connector plugin name.

"kafka.auth.mode": Identifies the connector authentication mode you want to use. There are two options: SERVICE_ACCOUNT or KAFKA_API_KEY (the default). To use an API key and secret, specify the configuration properties kafka.api.key and kafka.api.secret, as shown in the example configuration (above). To use a service account, specify the Resource ID in the property kafka.service.account.id=<service-account-resource-ID>. To list the available service account resource IDs, use the following command:
```
confluent iam service-account list
```
For example:
```
confluent iam service-account list

   Id     | Resource ID |       Name        |    Description
+---------+-------------+-------------------+-------------------
   123456 | sa-l1r23m   | sa-1              | Service account 1
   789101 | sa-l4d56p   | sa-2              | Service account 2
```

"topics": Identifies the topic name or a comma-separated list of topic names.
"keyfile": This contains the contents of the downloaded Google Cloud credentials JSON file for a Google Cloud service account. See Format service account keyfile credentials for details about how to format and use the contents of the downloaded credentials file as the "keyfile" property value.
"input.data.format": Sets the input Kafka record value format (data coming from the Kafka topic). Valid entries are AVRO, JSON_SR, PROTOBUF, or JSON. You must have Confluent Cloud Schema Registry configured if using a schema-based message format (for example, Avro, JSON_SR (JSON Schema), or Protobuf).

"ingestion.mode": Sets the connector to use either STREAMING or BATCH LOADING. Defaults to STREAMING.

Review the following table for details about the differences between using BATCH LOADING versus STREAMING mode with the BigQuery API. For more information, see Introduction to loading data.

BATCH LOADING	STREAMING
Records are available after a stream commit and may not be real time	Records are available for reading immediately after append call (minimal latency)
Requires creation of application streams	Default stream is used
May cost less	May cost more
More API quota limits (max streams, buffering, etc.)	Less quota limits

Note that BigQuery API quotas apply.

The following are additional properties you can use. See Configuration Properties for all property values and definitions.

"auto.create.tables": Designates whether to automatically create BigQuery tables. Defaults to DISABLED. Note that this property is applicable for AVRO, JSON_SR, and PROTOBUF message formats only. The other options available are listed below.
- NON-PARTITIONED: The connector creates non-partitioned tables.
- PARTITION by INGESTION TIME: The connector creates tables partitioned by ingestion time. Uses partitioning.type to set partitioning.
- PARTITION by FIELD: The connector creates tables partitioned using a field in a Kafka record value. The field name is set using the property timestamp.partition.field.name. Uses partitioning.type to set partitioning.
Note
New tables and schema updates may take a few minutes to be detected by the Google Client Library. For more information see the Google Cloud BigQuery API guide.
"auto.update.schemas": Defaults to DISABLED. Designates whether or not to automatically update BigQuery schemas. If ADD NEW FIELDS is selected, new fields are added with mode NULLABLE in the BigQuery schema. Note that this property is applicable for AVRO, JSON_SR, and PROTOBUF message formats only.
- "partitioning.type": The time partitioning type to use when creating new partitioned tables. Existing tables are not altered to use this partitioning type. Defaults to DAY.
- "timestamp.partition.field.name": The name of the field in the value that contains the timestamp to partition by in BigQuery. Used when PARTITION by FIELD is used.
"sanitize.topics": Designates whether to automatically sanitize topic names before using them as table names. If not enabled, topic names are used as table names. If enabled, the table names created may be different from the topic names. Source topic names must comply with BigQuery naming conventions even if sanitize.topics is set to true.
"sanitize.field.names": Designates whether to automatically sanitize field names before using them as column names in BigQuery. BigQuery specifies that field names can only contain letters, numbers, and underscores. The sanitizer replaces invalid symbols with underscores. If the field name starts with a digit, the sanitizer adds an underscore in front of the field name. Defaults to false.
Caution
Fields a.b and a_b will have the same value after sanitizing, which could cause a key duplication error. If not used, field names are used as column names.
sanitize.field.names.in.array: Determines whether to sanitize field names that are nested inside array-type objects. By default, the connector does not sanitize fields inside array-type objects during record ingestion. To enable sanitization for these fields, set sanitize.field.names.in.array to true.
Note
This property requires sanitize.field.names=true. To sanitize all fields, including those in arrays, you must set both properties to true.
"use.date.time.formatter": Defaults to false and the connector uses the SimpleDateFormat. If set to true, the connector uses DateTimeFormatter, which supports a wide range of epochs. Use DateTimeFormatter for time-based field such as Logical TIMESTAMP, Logical TIME, Logical DATE, Debezium Date , Debezium MicroTime, Debezium Time , Debezium MicroTimestamp, Debezium TIMESTAMP, or Debezium ZonedTimestamp.
The output can vary for the same input across the two formatters. For example, with the epoch -62104147200000, the connector performs the following:
- With SimpleDateFormat: Writes the timestamp as 0002-01-02T00:00:00 to the destination.
- With DateTimeFormatter: Writes the timestamp as 0001-12-31 00:00:00.000 to the destination.

Note

(Optional) To enable CSFLE for data encryption, specify the following properties:

csfle.enabled: Flag to indicate whether the connector honors CSFLE rules.
sr.service.account.id: A Service Account to access the Schema Registry and associated encryption rules or keys with that schema.
csfle.onFailure: Configures the connector behavior (ERROR or NONE) on data decryption failure. If set to ERROR, the connector fails and writes the encrypted data in the DLQ. If set to NONE, the connector writes the encrypted data in the target system without decryption.

Warning

Security Risk: Dead Letter Queue (DLQ) with CSFLE

When using CSFLE with connectors that route failed messages to a Dead Letter Queue (DLQ), be aware that data sent to the DLQ is written in plaintext (unencrypted). This poses a significant security risk as sensitive data that should be encrypted may be exposed in the DLQ.

Do not use DLQ with CSFLE in the current version. If you need error handling for CSFLE-enabled data, use alternative approaches such as:

Setting the connector behavior to ERROR to throw exceptions instead of routing to DLQ
Implementing custom error handling in your applications
Using NONE to pass encrypted data through without decryption

For more information on CSFLE setup, see Manage CSFLE for connectors.

Single Message Transforms: See the Single Message Transforms (SMT) documentation for details about adding SMTs using the CLI. See Unsupported transformations for a list of SMTs that are not supported with this connector.

See Configuration Properties for all property values and definitions.

Step 4: Load the configuration file and create the connector

Enter the following command to load the configuration and start the connector:

confluent connect cluster create --config-file <file-name>.json

For example:

confluent connect cluster create --config-file bigquery-storage-sink-config.json

Example output:

Created connector confluent-bigquery-storage-sink lcc-ix4dl

Step 5: Check the connector status

Enter the following command to check the connector status:

confluent connect cluster list

Example output:

ID          |       Name                      | Status  | Type
+-----------+---------------------------------+---------+------+
lcc-ix4dl   | confluent-bigquery-storage-sink | RUNNING | sink

Step 6: Check the results in BigQuery.

From the Google Cloud Console, go to your BigQuery project.
Query your datasets and verify that new records are being added.

For more information and examples to use with the Confluent Cloud API for Connect, see the Confluent Cloud API for Connect Usage Examples section.

Tip

When you launch a connector, a Dead Letter Queue topic is automatically created. See View Connector Dead Letter Queue Errors in Confluent Cloud for details.

Connector authentication

You can use a Google Cloud service account or Google OAuth to authenticate the connector with BigQuery. Note that OAuth is only available when using the Confluent Cloud UI to create the connector.

Create a Service Account

For details about how to create a Google Cloud service account and get a JSON credentials file, see Create and delete service account keys. Note the following when creating the service account:

The service account must have write access to the BigQuery project containing the dataset.

To use the Storage Write API, the connector must have bigquery.tables.updateData permissions. The minimum permissions are:

bigquery.datasets.get
bigquery.tables.create
bigquery.tables.get
bigquery.tables.getData
bigquery.tables.list
bigquery.tables.update
bigquery.tables.updateData

You create and download a key when creating a service account. The key must be downloaded as a JSON file. It resembles the example below:

{
  "type": "service_account",
  "project_id": "confluent-123456",
  "private_key_id": ".....",
  "private_key": "-----BEGIN PRIVATE ...omitted... =\n-----END PRIVATE KEY-----\n",
  "client_email": "confluent2@confluent-123456.iam.gserviceaccount.com",
  "client_id": "....",
  "auth_uri": "https://accounts.google.com/oauth2/auth",
  "token_uri": "https://oauth2.googleapis.com/token",
  "auth_provider_x509_cert_url": "https://www.googleapis.com/oauth2/certs",
  "client_x509_cert_url": "https://www.googleapis.com/robot/metadata/confluent2%40confluent-123456.iam.gserviceaccount.com"
}

Format service account keyfile credentials

Formatting the keyfile is only required when using the CLI or Terraform to create the connector, where you must add the service account key directly into the connector configuration.

The contents of the downloaded service account credentials JSON file must be converted to string format before it can be used in the connector configuration.

Convert the JSON file contents into string format.

Add the escape character \ before all \n entries in the Private Key section so that each section begins with \\n (see the highlighted lines below). The example below has been formatted so that the \\n entries are easier to see. Most of the credentials key has been omitted.

Tip

A script is available that converts the credentials to a string and also adds the additional escape characters where needed. See Stringify Google Cloud Credentials.

  {
      "name" : "confluent-bigquery-sink",
      "connector.class" : "BigQueryStorageSink",
      "kafka.api.key" : "<my-kafka-api-key>",
      "kafka.api.secret" : "<my-kafka-api-secret>",
      "topics" : "pageviews",
      "keyfile" : "{\"type\":\"service_account\",\"project_id\":\"connect-
      1234567\",\"private_key_id\":\"omitted\",
      \"private_key\":\"-----BEGIN PRIVATE KEY-----
      \\nMIIEvAIBADANBgkqhkiG9w0BA
      \\n6MhBA9TIXB4dPiYYNOYwbfy0Lki8zGn7T6wovGS5\opzsIh
      \\nOAQ8oRolFp\rdwc2cC5wyZ2+E+bhwn
      \\nPdCTW+oZoodY\\nOGB18cCKn5mJRzpiYsb5eGv2fN\/J
      \\n...rest of key omitted...
      \\n-----END PRIVATE KEY-----\\n\",
      \"client_email\":\"pub-sub@connect-123456789.iam.gserviceaccount.com\",
      \"client_id\":\"123456789\",\"auth_uri\":\"https:\/\/accounts.google.com\/o\/oauth2\/
      auth\",\"token_uri\":\"https:\/\/oauth2.googleapis.com\/
      token\",\"auth_provider_x509_cert_url\":\"https:\/\/
      www.googleapis.com\/oauth2\/v1\/
      certs\",\"client_x509_cert_url\":\"https:\/\/www.googleapis.com\/
      robot\/v1\/metadata\/x509\/pub-sub%40connect-
      123456789.iam.gserviceaccount.com\"}",
      "project": "<my-BigQuery-project>",
      "datasets":"<my-BigQuery-dataset>",
      "data.format":"AVRO",
      "tasks.max" : "1"
  }

Add all the converted string content to the "keyfile" credentials section of your configuration file as shown in the example above.

Set up Google OAuth (Bring your own app)

Complete the following steps to set up Google OAuth and get the Client ID and Client secret. This is applicable for the OAuth 2.0 (Bring your own app) BigQuery authentication option. For additional information, see Setting up OAuth 2.0.

Create a Project in Google Cloud Console. If you haven’t already, create a new project in the Google Cloud Console.
Enable the BigQuery API.
- Navigate to the API Library in the Google Cloud Console.
- Search for the BigQuery API.
- Select it and enable it for your project.
Configure the OAuth Consent Screen.
- Go to the OAuth consent screen in the Credentials section of the Google Cloud Console.
- Select Internal when prompted to set the User Type. This means only users within your organization can consent to the connector accessing data in BigQuery.
- In the Scopes section of the Consent screen, enter the following scope:
```
https://www.googleapis.com/auth/bigquery
```
- Fill in the remaining required fields and save.
Create OAuth 2.0 Credentials.
- In the Google Cloud Console, go to the Credentials page.
- Click on Create Credentials and select OAuth client ID.
- Choose the application type Web application.
- Set up the authorized redirect URI. Enter the following Confluent-provided URI:
```
https://confluent.cloud/api/connect/oauth/callback
```
Click Create. You are provided with the Client ID and Client secret.

Authorization Code grant flow

The following are the two stages of the authorization code grant flow when using OAuth 2.0.

Get the authorization code:
- The Google OAuth API endpoint is invoked by passing query parameters (for example, client_id, redirect_uri, and scope).
- The user handles authentication (AuthN) and authorization (AuthZ) with Google. Note that Confluent does not have access to the username and password.
- Google delivers the authorization code as a query parameter to the redirect_uri passed in the original API request.
Exchange the authorization code for an access_token and refresh_token.
- After invoking the Google OAuth API endpoint and passing the authorization code and other parameters, an access_token is returned and is used as the Bearer Token to access Google BigQuery.
- When the access_token expires, a new token is requested by passing the refresh_token request to the OAuth API endpoint. No human intervention occurs.

OAuth Limitations

Note the following limitations when using OAuth for the connector.

If you enter the wrong Client ID when authenticating, the UI attempts to launch the consent screen, and then displays the error for an invalid client. Click the back button and re-enter the connector configuration. This is applicable for the OAuth 2.0 (Bring your own app) BigQuery authentication option.
If you enter the correct Client ID but enter the wrong Client secret, the UI successfully launches the consent screen and the user is able to authorize. However, the API request fails. The user is redirected to the connect UI provisioning screen. This is applicable for the OAuth 2.0 (Bring your own app) BigQuery authentication option.
You must manually revoke permissions for the connector from Third-party apps & services if you delete the connector or switch to a different auth mechanism.
Google Cloud allows a maximum of 100 refresh tokens for a single user using a single OAuth 2.0 client ID. For example, a single user can create a maximum of 100 connectors using the same client ID. When this user creates connector 101 using the same client ID, the refresh token for the first connector created is revoked. If you want to run more than 100 connectors, you can use multiple Client IDs or User IDs. Or, you can use a service account.

Legacy to V2 Connector Migration

The BigQuery Sink Legacy connector uses the BigQuery legacy streaming API and will reach end of life (EOL) on January 9, 2026. Confluent recommends migrating to the BigQuery Sink V2 connector. The BigQuery Sink V2 connector uses the BiqQuery storage write API, which offers higher throughput, improved reliability, and better schema evolution support than the legacy connector.

Due to fundamental differences between the underlying Google Cloud APIs, a direct in-place upgrade is not possible. The following sections provide the migration approach with no data loss, including the known API changes, and mappings for key configuration properties.

You can choose one of two paths for your migration:

Automated migration (recommended): Use the Bigquery V2 Sink migration tool provided by Confluent. This is the simplest and fastest method, as it automates the transfer of configurations and offsets.
Manual migration: Set up a parallel pipeline, validate data, and manually cut over from the legacy to the V2 connector. This approach offers more granular control but requires more hands-on effort.

Prerequisites

Confluent Cloud: An active account with access to the legacy connector.
Google Cloud: A Google Cloud service account with the necessary BigQuery permissions.
Automated migration script: If you choose to use the automated script, ensure your local machine has Python 3.6 or later installed.

Using the automated migration

For a streamlined approach, use the Bigquery V2 Sink migration tool. This tool is designed to simplify the process by automating the transfer of existing offsets and topic mappings from your legacy connector configuration to the new V2 connector. The migration tool, migrate-to-bq-v2-sink.py, is available in the official Confluent GitHub repository.

Quick start

Clone or download the repository containing the script.

Run the migration tool from your local machine:

python3 migrate-to-bq-v2-sink.py --legacy_connector "<YOUR_LEGACY_CONNECTOR_NAME>" --environment "<YOUR_ENVIRONMENT_NAME>" --cluster_id "<YOUR_KAFKA_CLUSTER_ID>"

The tool will guide you through configuring the new V2 connector, including:

Check V1 Sink connector status: The tool shows your connector’s current status.
- If testing on dummy tables: You can keep the existing connector running.
- For production tables: It’s recommended to pause the V1 connector to avoid data duplication.
Get environment details: Fetch the environment name and cluster ID from your Kafka cluster’s URL in Confluent Cloud.
Set credentials: The tool supports three methods for providing your Confluent Cloud credentials.
- Environment variables: Export your email and password as environment variables. Note that the environment variables can be visible in process lists and command history.
```
export EMAIL="your-email@example.com"
export PASSWORD="your-password"
```
- Credentials file (recommended): Create a JSON file with your credentials.
```
{
  "email": "your-email@example.com",
  "password": "your-password"
}
```
- Secure input: Enter credentials interactively when prompted.
Follow the interactive prompts: The tool will guide you through configuring the new V2 connector, including:
- Connector Name: Enter a name for the new V2 connector.
- Ingestion Mode: Select a mode for ingestion. Defaults to STREAMING.
  - STREAMING: Lower latency, higher cost (default)
  - BATCH LOADING: Higher latency, lower cost
  - UPSERT: For upsert operations (requires key fields)
  - UPSERT_DELETE: For upsert and delete operations (requires key fields)
- Int8/Int16 casting for Byte and Short fields: Choose between FLOAT (default) or INTEGER.
- Commit Interval: For BATCH LOADING mode, set interval between 60-14400 seconds.
- Auto Create Tables: Configure table creation (defaults to DISABLED).
  - DISABLED: Do not auto-create tables (default)
  - NON-PARTITIONED: Create tables without partitioning
  - PARTITION by INGESTION TIME: Create time-partitioned tables
  - PARTITION by FIELD: Create field-partitioned tables
- Partitioning: Set partitioning type and field (if applicable).
  - HOUR: Partition by hour
  - DAY: Partition by day
  - MONTH: Partition by month
  - YEAR: Partition by year
- Topic to Table Mapping: For testing purposes, you can configure topic2table.map to redirect data to different tables.
  - Use existing mapping: If already configured in the legacy connector.
  - Configure new mapping: For testing without affecting production tables. For example, my-topic:my-test-table,another-topic:another-test-table.
  - Skip mapping: Use default table names.
- Date Time Formatter: Choose between SimpleDateFormat or DateTimeFormatter.
- GCP Service Account Keyfile: Provide your credentials via file path, environment variable, or direct input.
  - File Path: Provide the path to your JSON keyfile
  - Environment Variable: Set GCP_KEYFILE_PATH environment variable
  - Direct Input: Paste the JSON content directly
- Credentials: Choose a secure method for Confluent Cloud authentication.
Review and Confirm: The tool displays the final configuration and asks for confirmation before creating the connector.

Considerations

Preserve offsets: The V2 connector will start from the latest offset of your legacy connector.
Prevent data loss: The tool automatically transfers offsets to avoid data duplication.
Review breaking changes: Before you migrate, review the breaking changes section.
Test first: Always test with a small dataset first before upgrading a production connector.
Monitor after migration: Monitor the new connector for any issues after migration.
Check connector status: The tool shows the connector status and recommends actions based on whether you’re testing or migrating production data.

Using the manual migration

Follow these steps to migrate your pipelines from the legacy connector to the V2 connector.

Understand the key changes: Review the differences in API behavior, data type handling, and configuration properties. Pay close attention to how TIMESTAMP values are handled.
Set up a parallel pipeline: Never attempt to upgrade a production connector in place. A parallel environment is essential for validation without impacting your production data flow. Create a new BigQuery Sink V2 connector in a test environment, pointing it to a new, separate BigQuery table for validation.
To directly compare fields, data types, and stored data between legacy and V2, you can continue using the same Kafka topic as your production connector.
Validate data integrity: Run the legacy and V2 connectors in parallel, writing to separate tables. Compare the data in both tables to ensure consistency.
- Confirm that TIMESTAMP, DATE, and DATETIME fields are handled correctly.
- Ensure all fields are mapped as NULLABLE in BigQuery, as expected.
- Check for any inconsistencies with nested or complex data schemas.
Cutover to V2: After validating the V2 pipeline, perform the following:
- Gracefully stop the legacy BigQuery sink connector.
- Start or resume the new V2 connector from the last committed offsets of the V1 connector.
- After confirming the V2 connector is running stably, decommission the legacy connector and update any monitoring systems to use the new connector.

Breaking API changes

The following table lists API differences that were identified during connector development and testing. Be aware that there may be additional breaking changes that were not identified.

BigQuery Schema Types	Legacy InsertAll API	Storage Write API (V2)
TIMESTAMP	Input value is considered seconds since epoch	Input value is considered microseconds since epoch
DATE	INT is not supported	INT in range -719162 (0001-01-01) to 2932896 (9999-12-31) is supported
TIMESTAMP, JSON, INTEGER, BIGNUMERIC, NUMERIC, STRING, DATE, TIME, DATETIME, BYTES	Supports most primitive data type inputs	See Supported data types
DATE, TIME, DATETIME, TIMESTAMP	Supports entire datetime canonical format	Supports a subset of the datetime canonical format

New V2 properties

These properties provide new functionality that wasn’t available in the legacy connector.

V2 property	Default value	Description
`ingestion.mode`	`STREAMING`	Selects the data ingestion strategy. `STREAMING` offers lower latency and higher cost, while `BATCH LOADING` offers higher latency and lower cost. `UPSERT` and `UPSERT_DELETE` modes handle record-level updates and deletes.
`auto.create.tables`	`DISABLED`	Provides more granular control over table creation. Options include: `NON-PARTITIONED`, `PARTITION by INGESTION TIME`, and `PARTITION by FIELD`.
`use.date.time.formatter`	`false` (uses `SimpleDateFormat`)	Allows switching from the default `SimpleDateFormat` to the more modern `DateTimeFormatter` for handling date and time types.
`use.integer.for.int8.int16`	`FLOAT`	Determines how `INT8` and `INT16` Kafka types are cast.

Discontinued properties

The following legacy configuration properties have been removed in V2.

Discontinued legacy property	V2 behavior & equivalent
`allowBigQueryRequiredFieldRelaxation`	Not supported. All fields in the V2 connector are created as `NULLABLE` by default. To make a field `REQUIRED`, you must manually alter the table schema in the BigQuery UI.
`allowSchemaUnionization`	Removed. This functionality is now part of the `auto.update.schemas` property, which handles schema evolution for both primitive and complex types (structs and arrays).
`bigQueryRetry` and `bigQueryRetryWait`	Not user-configurable. The V2 connector uses built-in retry mechanisms with a default retry count of 10 and a wait time of 1,000 ms.
`defaultDataset`	Removed. Datasets are specified directly in the connector configuration.
`schemaRetriever`	Removed. Its functionality for creating and updating tables is now handled by the `auto.create.tables` and `auto.update.schemas` properties.

Properties with changed behavior

The following configuration properties exist in both the legacy and V2 versions but have different implementations or capabilities in V2. Note the legacy to V2 mapping for the key configuration properties:

Configuration property	Legacy behavior	V2 behavior
`auto.update.schemas`	Allowed automatic updates for new primitive fields only.	Supports schema updates for both primitive and complex field types (structs and arrays), making it more robust.
`all.bq.fields.nullable`	Could be set to `false` to enforce `REQUIRED` fields based on the schema.	Removed. All fields are created as `NULLABLE` by default, with no override configuration available.
`partitioning.type` and `timestamp.partition.field.name`	Offered basic partitioning options.	These configurations offer more granular control and work in tandem with the enhanced `auto.create.tables` options.
`convert.double.special.values`	User-configurable	Not exposed in Confluent Cloud. The default behavior is `false`, passing non-finite values (`NaN`, `+Infinity`, `-Infinity`) as-is without conversion.

Migration FAQs

The frequently asked questions (FAQs) address common questions and issues encountered while migrating from the legacy to BigQuery Sink V2 connector for Confluent Cloud.

Why should I migrate to the V2 connector?

The V2 connector, which uses the underlying Storage Write API, offers significant benefits:

Higher throughput: Sustains much higher data ingestion rates.
Future-proof: Aligns with Google Cloud’s official recommendation for streaming data into BigQuery.
Improved reliability: Provides better error handling and built-in retry mechanisms.
Expanded type support: Natively supports NUMERIC and BIGNUMERIC data types.
Enhanced schema evolution: More robustly handles changes to Kafka topic schemas.
Flexible write strategies: Supports multiple delivery modes, including STREAMING, BATCH LOADING, UPSERT, and UPSERT_DELETE.

What happens if I don’t migrate before the legacy connector EOL?

The legacy BigQuery streaming API is already deprecated, and in alignment with this, the legacy BigQuery connector is scheduled for EOL on January 9, 2026.

After this date:

You can no longer create new legacy BigQuery connectors.
Existing connectors may continue to function, but they will not receive any updates or fixes. To ensure long-term stability and support, migrate to the BigQuery Sink V2 connector as soon as possible.

Are these two connectors just different versions of the same software?

No. They are two completely different connectors from an architectural standpoint. The legacy connector is built on BigQuery’s old InsertAll API, while the V2 connector was written from the ground up to leverage the modern BiqQuery storage write API. This fundamental difference in the underlying technology is why a direct, in-place upgrade is not possible and requires a parallel migration strategy.

Is there a price difference between the legacy and V2 connectors?

No, Confluent prices both the legacy and V2 connectors the same.

Are there any changes to my Google Cloud costs?

While Confluent’s pricing for the connector remains the same, the underlying Google Cloud APIs have different pricing models. The legacy InsertAll API and the new Storage Write API are billed differently by Google. You should review Google Cloud’s BigQuery pricing documentation to estimate the cost impact for your specific workload.

Do I need to stop my legacy connector before starting V2 connector?

No. During the validation phase, you should run them in parallel. However, once you cut over to the V2 connector for production traffic, you must stop the legacy connector to prevent duplicate data writes.

Will my old offsets work with the V2 connector?

Yes, but you should not reuse them manually. Offsets are stored per partition and can be reused by the V2 connector to ensure continuity. The Bigquery V2 Sink migration tool leverages these offsets so the V2 connector starts from the last committed offset of each partition, avoiding skipped data or duplication. You should always use the migration tool to correctly map offsets.

Can I use the BigQuery V2 Sink migration tool mentioned in the docs?

Yes. Confluent provides a Bigquery V2 Sink migration tool to automate the migration process for users looking for a more automated approach. However, we still strongly recommend that you understand the manual steps and perform thorough validation before switching production workloads.

Does the connector support a single topic with multiple schemas to write to multiple tables?

No. The BigQuery V2 Sink connector does not support splitting data from a single Kafka topic into multiple BigQuery tables based on different schemas. Each Kafka topic maps to one dedicated BigQuery table. However, a single connector instance can consume from multiple Kafka topics, with each topic mapping to its own respective table and schema.

Is `allow.schema.unionization` configuration property available in V2 connector?

No. The allow.schema.unionization property is removed in V2. In the V1 connector, you needed this property enabled to update complex fields (structs and arrays). In the V2 connector, the auto.update.schemas property handles schema updates for both primitive and complex field types.

Why are all fields mapped as `NULLABLE` in V2?

The V2 connector uses this design to ensure maximum compatibility and prevent load failures that occur when a source record is missing an optional field. Unlike the legacy connector (all.bq.fields.nullable), V2 does not currently offer a configuration to override this behavior. If you require a field to be REQUIRED, manually update the table schema in the BigQuery UI after the table is created.

Why is the `convert.double.special.values` configuration not available in BigQuery Sink V2?

The V2 connector supports the convert.double.special.values configuration internally, but it is not exposed in Confluent Cloud. The default value is false, which means non-finite double values (NaN, +Infinity, -Infinity) are passed through as-is and are not converted to maximum (Double.MAX_VALUE) or minimum values (Double.MIN_VALUE).

Does the BigQuery V2 connector require users to include `_CHANGE_TYPE` in Kafka records?

No. In UPSERT_DELETE mode, the connector automatically interprets a record containing only primary key fields (with all other fields as null) as a delete operation. The connector’s backend then internally sets _CHANGE_TYPE=DELETE before applying the change in BigQuery.

Will there be a V2 connector for self-managed Confluent Platform?

At this time, there are no plans to release a self-managed V2 connector. We are currently evaluating customer demand. If you need this capability, submit a Feature Request through your Confluent account team so we can track and prioritize interest.

Configuration Properties

Use the following configuration properties with this connector.

Which topics do you want to get data from?

topics.regex

A regular expression that matches the names of the topics to consume from. This is useful when you want to consume from multiple topics that match a certain pattern without having to list them all individually.

Type: string
Importance: low

topics

Identifies the topic name or a comma-separated list of topic names.

Type: list
Importance: high

errors.deadletterqueue.topic.name

The name of the topic to be used as the dead letter queue (DLQ) for messages that result in an error when processed by this sink connector, or its transformations or converters. Defaults to ‘dlq-${connector}’ if not set. The DLQ topic will be created automatically if it does not exist. You can provide ${connector} in the value to use it as a placeholder for the logical cluster ID.

Type: string
Default: dlq-${connector}
Importance: low

Schema Config

schema.context.name

Add a schema context name. A schema context represents an independent scope in Schema Registry. It is a separate sub-schema tied to topics in different Kafka clusters that share the same Schema Registry instance. If not used, the connector uses the default schema configured for Schema Registry in your Confluent Cloud environment.

Type: string
Default: default
Importance: medium

Input messages

input.key.format

Sets the input Kafka record key format. Valid entries are AVRO, BYTES, JSON, JSON_SR, PROTOBUF. Note that you need to have Confluent Cloud Schema Registry configured if using a schema-based message format like AVRO, JSON_SR, and PROTOBUF

Type: string
Default: BYTES
Valid Values: AVRO, BYTES, JSON, JSON_SR, PROTOBUF, STRING
Importance: high

input.data.format

Sets the input Kafka record value format. Valid entries are AVRO, JSON_SR, PROTOBUF, and JSON. Note that you need to have Confluent Cloud Schema Registry configured if using a schema-based message format like AVRO, JSON_SR, or PROTOBUF.

Type: string
Default: JSON
Importance: high

How should we connect to your data?

name

Sets a name for your connector.

Type: string
Valid Values: A string at most 64 characters long
Importance: high

Kafka Cluster credentials

kafka.auth.mode

Kafka Authentication mode. It can be one of KAFKA_API_KEY or SERVICE_ACCOUNT. It defaults to KAFKA_API_KEY mode.

Type: string
Default: KAFKA_API_KEY
Valid Values: KAFKA_API_KEY, SERVICE_ACCOUNT
Importance: high

kafka.api.key

Kafka API Key. Required when kafka.auth.mode==KAFKA_API_KEY.

Type: password
Importance: high

kafka.service.account.id

The Service Account that will be used to generate the API keys to communicate with Kafka Cluster.

Type: string
Importance: high

kafka.api.secret

Secret associated with Kafka API key. Required when kafka.auth.mode==KAFKA_API_KEY.

Type: password
Importance: high

GCP credentials

provider.integration.id

Select an existing integration that has access to your resource. In case you need to integrate a new Google Service Account, use provider integration

Type: string
Importance: high

authentication.method

Select how you want to authenticate with BigQuery.

Type: string
Default: Google cloud service account
Importance: high

keyfile

GCP service account JSON file with write permissions for BigQuery.

Type: password
Importance: high

oauth.client.id

Client ID of your Google OAuth application.

Type: string
Importance: high

oauth.client.secret

Client secret of your Google OAuth application.

Type: password
Importance: high

oauth.refresh.token

OAuth 2.0 refresh token for BigQuery.

Type: password
Importance: high

BigQuery details

project

ID for the GCP project where BigQuery is located.

Type: string
Importance: high

datasets

Name of the BigQuery dataset where table(s) is located.

Type: string
Importance: high

Ingestion Mode details

ingestion.mode

Select a mode to ingest data into the table. Select STREAMING for reduced latency. Select BATCH LOADING for cost savings. Select UPSERT for upserting records. Select UPSERT_DELETE for upserting and deleting records.

Type: string
Default: STREAMING
Importance: high

Insertion and DDL support

commit.interval

The interval, in seconds, the connector attempts to commit streamed records. Set the interval between 60 seconds (1 minute) and 14,400 seconds (4 hours). Be careful when setting the commit interval as on every commit interval, a task calls the CreateWriteStream API which is subject to a quota <https://cloud.google.com/bigquery/quotas#write-api-limits:~:text=handle%20unexpected%20demand.-,CreateWriteStream,-requests>. For example, if you have five tasks (may belong to different connectors also) set to commit every 60 seconds to a project, there will be five calls to CreateWriteStream API on that project every minute. If the count exceeds the allowed quota, some tasks may fail.

Type: int
Default: 60
Valid Values: [60,…,14400]
Importance: high

topic2table.map

Map of topics to tables (optional). The required format is comma-separated tuples. For example, <topic-1>:<table-1>,<topic-2>:<table-2>,… Note that a topic name must not be modified using a regex SMT while using this option. If this property is used, sanitize.topics is ignored. Also, if the topic-to-table map doesn’t contain the topic for a record, the connector creates a table with the same name as the topic name.

Type: string
Default: “”
Importance: medium

sanitize.topics

Designates whether to automatically sanitize topic names before using them as table names in BigQuery. If not enabled, topic names are used as table names.

Type: boolean
Default: true
Importance: high

sanitize.field.names

Whether to automatically sanitize field names before using them as field names in BigQuery. BigQuery specifies that field names can only contain letters, numbers, and underscores. The sanitizer replaces invalid symbols with underscores. If the field name starts with a digit, the sanitizer adds an underscore in front of field name. Caution: Key duplication errors can occur if different fields are named a.b and a_b, for instance. After being sanitized, field names a.b and a_b will have same value.

Type: boolean
Default: false
Importance: high

auto.create.tables

Designates whether or not to automatically create BigQuery tables. Note: Supports AVRO, JSON_SR, and PROTOBUF message format only.

Type: string
Default: DISABLED
Importance: high

auto.update.schemas

Designates whether or not to automatically update BigQuery schemas. New fields in record schemas must be nullable. Note: Supports AVRO, JSON_SR, and PROTOBUF message format only.

Type: string
Default: DISABLED
Importance: high

sanitize.field.names.in.array

Whether to automatically sanitize field names inside arrays. When enabled, field names inside arrays will also be sanitized according to BigQuery naming rules. This setting only takes effect if ‘sanitize.field.names’ is also enabled.

Type: boolean
Default: false
Importance: medium

partitioning.type

The time partitioning type to use when creating new partitioned tables. Existing tables will not be altered to use this partitioning type.

Type: string
Default: DAY
Importance: low

timestamp.partition.field.name

The name of the field in the value that contains the timestamp to partition by in BigQuery. This also enables timestamp partitioning for each table.

Type: string
Importance: low

use.date.time.formatter

Specify whether to use a DateTimeFormatter to support a wide range of epochs. Setting this true will use DateTimeFormatter over default SimpleDateFormat. The output might vary for same input between the two formatters.

Type: boolean
Default: false
Importance: high

topic2clustering.fields.map

Maps topics to their corresponding table clustering fields (optional). Format: comma-separated tuples, e.g., topic1:[col1|col2], topic2:[col1|col2|..], … Specifies the fields used to cluster data in BigQuery. The order of the fields determines the clustering precedence.

Type: string
Default: “”
Importance: low

use.integer.for.int8.int16

Determines how the connector stores INT8 (BYTE) and INT16 (SHORT) data types in BigQuery during auto table creation and schema update. When set to false``(default), these values are stored as FLOAT. When set to ``true, INT8 (BYTE) and INT16 (SHORT) values are stored as INTEGER.

Type: boolean
Default: false
Importance: low

Consumer configuration

max.poll.interval.ms

The maximum delay between subsequent consume requests to Kafka. This configuration property may be used to improve the performance of the connector, if the connector cannot send records to the sink system. Defaults to 300000 milliseconds (5 minutes).

Type: long
Default: 300000 (5 minutes)
Valid Values: [60000,…,1800000] for non-dedicated clusters and [60000,…] for dedicated clusters
Importance: low

max.poll.records

The maximum number of records to consume from Kafka in a single request. This configuration property may be used to improve the performance of the connector, if the connector cannot send records to the sink system. Defaults to 500 records.

Type: long
Default: 500
Valid Values: [1,…,500] for non-dedicated clusters and [1,…] for dedicated clusters
Importance: low

Number of tasks for this connector

tasks.max

Maximum number of tasks for the connector.

Type: int
Valid Values: [1,…]
Importance: high

Additional Configs

consumer.override.auto.offset.reset

Defines the behavior of the consumer when there is no committed position (which occurs when the group is first initialized) or when an offset is out of range. You can choose either to reset the position to the “earliest” offset (the default) or the “latest” offset. You can also select “none” if you would rather set the initial offset yourself and you are willing to handle out of range errors manually. More details: https://docs.confluent.io/platform/current/installation/configuration/consumer-configs.html#auto-offset-reset

Type: string
Importance: low

consumer.override.isolation.level

Controls how to read messages written transactionally. If set to read_committed, consumer.poll() will only return transactional messages which have been committed. If set to read_uncommitted (the default), consumer.poll() will return all messages, even transactional messages which have been aborted. Non-transactional messages will be returned unconditionally in either mode. More details: https://docs.confluent.io/platform/current/installation/configuration/consumer-configs.html#isolation-level

Type: string
Importance: low

header.converter

The converter class for the headers. This is used to serialize and deserialize the headers of the messages.

Type: string
Importance: low

value.converter.allow.optional.map.keys

Allow optional string map key when converting from Connect Schema to Avro Schema. Applicable for Avro Converters.

Type: boolean
Importance: low

value.converter.auto.register.schemas

Specify if the Serializer should attempt to register the Schema.

Type: boolean
Importance: low

value.converter.connect.meta.data

Allow the Connect converter to add its metadata to the output schema. Applicable for Avro Converters.

Type: boolean
Importance: low

value.converter.enhanced.avro.schema.support

Enable enhanced schema support to preserve package information and Enums. Applicable for Avro Converters.

Type: boolean
Importance: low

value.converter.enhanced.protobuf.schema.support

Enable enhanced schema support to preserve package information. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.flatten.unions

Whether to flatten unions (oneofs). Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.generate.index.for.unions

Whether to generate an index suffix for unions. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.generate.struct.for.nulls

Whether to generate a struct variable for null values. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.int.for.enums

Whether to represent enums as integers. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.latest.compatibility.strict

Verify latest subject version is backward compatible when use.latest.version is true.

Type: boolean
Importance: low

value.converter.object.additional.properties

Whether to allow additional properties for object schemas. Applicable for JSON_SR Converters.

Type: boolean
Importance: low

value.converter.optional.for.nullables

Whether nullable fields should be specified with an optional label. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.optional.for.proto2

Whether proto2 optionals are supported. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.scrub.invalid.names

Whether to scrub invalid names by replacing invalid characters with valid characters. Applicable for Avro and Protobuf Converters.

Type: boolean
Importance: low

value.converter.use.latest.version

Use latest version of schema in subject for serialization when auto.register.schemas is false.

Type: boolean
Importance: low

value.converter.use.optional.for.nonrequired

Whether to set non-required properties to be optional. Applicable for JSON_SR Converters.

Type: boolean
Importance: low

value.converter.wrapper.for.nullables

Whether nullable fields should use primitive wrapper messages. Applicable for Protobuf Converters.

Type: boolean
Importance: low

value.converter.wrapper.for.raw.primitives

Whether a wrapper message should be interpreted as a raw primitive at root level. Applicable for Protobuf Converters.

Type: boolean
Importance: low

errors.tolerance

Use this property if you would like to configure the connector’s error handling behavior. WARNING: This property should be used with CAUTION for SOURCE CONNECTORS as it may lead to dataloss. If you set this property to ‘all’, the connector will not fail on errant records, but will instead log them (and send to DLQ for Sink Connectors) and continue processing. If you set this property to ‘none’, the connector task will fail on errant records.

Type: string
Default: all
Importance: low

key.converter.key.subject.name.strategy

How to construct the subject name for key schema registration.

Type: string
Default: TopicNameStrategy
Importance: low

key.converter.replace.null.with.default

Whether to replace fields that have a default value and that are null to the default value. When set to true, the default value is used, otherwise null is used. Applicable for JSON Key Converter.

Type: boolean
Default: true
Importance: low

key.converter.schemas.enable

Include schemas within each of the serialized keys. Input message keys must contain schema and payload fields and may not contain additional fields. For plain JSON data, set this to false. Applicable for JSON Key Converter.

Type: boolean
Default: false
Importance: low

value.converter.decimal.format

Specify the JSON/JSON_SR serialization format for Connect DECIMAL logical type values with two allowed literals:

BASE64 to serialize DECIMAL logical types as base64 encoded binary data and

NUMERIC to serialize Connect DECIMAL logical type values in JSON/JSON_SR as a number representing the decimal value.

Type: string
Default: BASE64
Importance: low

value.converter.flatten.singleton.unions

Whether to flatten singleton unions. Applicable for Avro and JSON_SR Converters.

Type: boolean
Default: false
Importance: low

value.converter.ignore.default.for.nullables

When set to true, this property ensures that the corresponding record in Kafka is NULL, instead of showing the default column value. Applicable for AVRO,PROTOBUF and JSON_SR Converters.

Type: boolean
Default: false
Importance: low

value.converter.reference.subject.name.strategy

Set the subject reference name strategy for value. Valid entries are DefaultReferenceSubjectNameStrategy or QualifiedReferenceSubjectNameStrategy. Note that the subject reference name strategy can be selected only for PROTOBUF format with the default strategy being DefaultReferenceSubjectNameStrategy.

Type: string
Default: DefaultReferenceSubjectNameStrategy
Importance: low

value.converter.replace.null.with.default

Whether to replace fields that have a default value and that are null to the default value. When set to true, the default value is used, otherwise null is used. Applicable for JSON Converter.

Type: boolean
Default: true
Importance: low

value.converter.schemas.enable

Include schemas within each of the serialized values. Input messages must contain schema and payload fields and may not contain additional fields. For plain JSON data, set this to false. Applicable for JSON Converter.

Type: boolean
Default: false
Importance: low

value.converter.value.subject.name.strategy

Determines how to construct the subject name under which the value schema is registered with Schema Registry.

Type: string
Default: TopicNameStrategy
Importance: low

Auto-restart policy

auto.restart.on.user.error

Enable connector to automatically restart on user-actionable errors.

Type: boolean
Default: true
Importance: medium

Next Steps

For an example that shows fully-managed Confluent Cloud connectors in action with Confluent Cloud for Apache Flink, see the Cloud ETL Demo. This example also shows how to use Confluent CLI to manage your resources in Confluent Cloud.

Google BigQuery Sink V2 Connector for Confluent Cloud

Features

Limitations

Supported data types

General mapping

Table creation and schema update mapping

Quick Start

Using the Confluent Cloud Console

Step 1: Launch your Confluent Cloud cluster

Step 2: Add a connector

Step 3: Select your connector

Step 4: Enter the connector details

Step 5: Check the results in BigQuery

Using the Confluent CLI

Step 1: List the available connectors

Step 2: List the connector configuration properties

Step 3: Create the connector configuration file

Step 4: Load the configuration file and create the connector

Step 5: Check the connector status

Step 6: Check the results in BigQuery.

Connector authentication

Create a Service Account

Format service account keyfile credentials

Set up Google OAuth (Bring your own app)

Authorization Code grant flow

OAuth Limitations

Legacy to V2 Connector Migration

Using the automated migration

Quick start

Considerations

Using the manual migration

Breaking API changes

New V2 properties

Discontinued properties

Properties with changed behavior

Migration FAQs

Why should I migrate to the V2 connector?

What happens if I don’t migrate before the legacy connector EOL?

Are these two connectors just different versions of the same software?

Is there a price difference between the legacy and V2 connectors?

Are there any changes to my Google Cloud costs?

Do I need to stop my legacy connector before starting V2 connector?

Will my old offsets work with the V2 connector?

Can I use the BigQuery V2 Sink migration tool mentioned in the docs?

Does the connector support a single topic with multiple schemas to write to multiple tables?

Is allow.schema.unionization configuration property available in V2 connector?

Why are all fields mapped as NULLABLE in V2?

Why is the convert.double.special.values configuration not available in BigQuery Sink V2?

Does the BigQuery V2 connector require users to include _CHANGE_TYPE in Kafka records?

Will there be a V2 connector for self-managed Confluent Platform?

Configuration Properties

Which topics do you want to get data from?

Schema Config

Input messages

How should we connect to your data?

Kafka Cluster credentials

GCP credentials

BigQuery details

Ingestion Mode details

Insertion and DDL support

Consumer configuration

Number of tasks for this connector

Additional Configs

Auto-restart policy

Next Steps

Is `allow.schema.unionization` configuration property available in V2 connector?

Why are all fields mapped as `NULLABLE` in V2?

Why is the `convert.double.special.values` configuration not available in BigQuery Sink V2?

Does the BigQuery V2 connector require users to include `_CHANGE_TYPE` in Kafka records?