| connector.class | Yes | The Java class for the GCS Sink connector. | |
| tasks.max | Yes | The maximum number of tasks that should be created for this connector. | |
| topics | No | A comma-separated list of Kafka topics to consume from, Only one of topics or topics.regex should be specified. | |
| topics.regex | No | Regular expression giving topics to consume. Under the hood, the regex is compiled to a java.util.regex.Pattern. Only one of topics or topics.regex should be specified. | |
| gcs.bucket.name | Yes | The name of the GCS bucket where the data will be stored. | |
| gcs.credentials.json | No | The GCP credentials in JSON format. If not provided, the connector will use the default application credentials. | |
| gcs.credentials.path | No | The path to a GCP credentials file. Cannot be set together with “gcs.credentials.json or “gcs.credentials.default. | |
| gcs.credentials.default | No | Whether to connect using default the GCP SDK default credential discovery. When set to null (the default) or false, will fall back to connecting with No Credentials.Cannot be set together with “gcs.credentials.json” or “gcs.credentials.path”. | |
| gcs.object.content.encoding | No | The GCS object metadata value of Content-Encoding. | |
| gcs.endpoint | No | Explicit GCS Endpoint Address, mainly for testing. | |
| gcs.retry.backoff.initial.delay.ms | No | Initial retry delay in milliseconds. The default value is 1000. | 1000 |
| gcs.retry.backoff.max.delay.ms | No | Maximum retry delay in milliseconds. The default value is 32000. | 32000 |
| gcs.retry.backoff.delay.multiplier | No | Retry delay multiplier. The default value is 2.0. | 2.0 |
| gcs.retry.backoff.max.attempts | No | Retry max attempts. The default value is 6. | 6 |
| gcs.retry.backoff.total.timeout.ms | No | Retry total timeout in milliseconds. The default value is 50000. | 50000 |
| gcs.user.agent | No | A custom user agent used while contacting google. | ”Google GCS Sink/3.4.1 (GPN: Aiven;)“ |
| file.name.prefix | No | The prefix to be added to the name of each file put on GCS. | |
| file.name.template | No | The template for file names on GCS. Supports {{ variable }} placeholders for substituting variables. Currently supported variables are topic, partition, and start_offset (the offset of the first record in the file). | {{topic}}-{{partition:padding=false}}-{{start_offset:padding=false}} |
| file.compression.type | No | The compression type used for files put on GCS. The supported values are: ‘none’, ‘gzip’, ‘snappy’, ‘zstd’. | none |
| file.max.records | No | The maximum number of records to put in a single file. Must be a non-negative integer number. 0 is interpreted as “unlimited”, which is the default. | 0 |
| file.name.timestamp.timezone | No | Specifies the timezone in which the dates and time for the timestamp variable will be treated. Use standard shot and long names. Default is UTC. | UTC |
| file.name.timestamp.source | No | Specifies the the timestamp variable source. Default is wall-clock. | WALLCLOCK |
| format.output.type | No | The format type of output contentThe supported values are: ‘avro’, ‘csv’, ‘json’, ‘jsonl’, ‘parquet’. | csv |
| format.output.fields | No | Fields to put into output files. The supported values are: ‘key’, ‘value’, ‘offset’, ‘timestamp’, ‘headers’. | value |
| format.output.fields.value.encoding | No | The type of encoding for the value field. The supported values are: ‘none’, ‘base64’. | base64 |
| format.output.envelope | No | Whether to enable envelope for entries with single field. | true |
| errors.deadletterqueue.topic.name | No | 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. The topic name is blank by default, which means that no messages are to be recorded in the DLQ. | |
| errors.deadletterqueue.topic.replication.factor | No | Replication factor used to create the dead letter queue topic when it doesn’t already exist. | 3 |
| errors.deadletterqueue.context.headers.enable | No | If true, add headers containing error context to the messages written to the dead letter queue. To avoid clashing with headers from the original record, all error context header keys, all error context header keys will start with __connect.errors. | false |
