HTTP Source V2 Connector for Confluent Cloud

The fully-managed HTTP Source V2 connector for Confluent Cloud integrates Apache Kafka® with an API using HTTP or HTTPS. It allows you to configure one or more APIs seamlessly with an OpenAPI/Swagger specification file, reducing overall configuration time and helping you achieve better performance when compared to the HTTP Source Connector for Confluent Cloud. In this page, you will find all the features the HTTP Source V2 connector offers and discover everything you need to begin using the connector.

Note

If you require private networking for fully-managed connectors, make sure to set up the proper networking beforehand. For more information, see Manage Networking for Confluent Cloud Connectors.

Features

The HTTP Source V2 connector includes the following features:

Multiple API path support: The connector allows you to configure up to 15 API paths per connector having the same base URL and authentication mechanism.
OpenAPI Specification-based configuration: The connector provides seamless configuration through an OpenAPI specification file.
Secure access and data exchange: The connector supports the following authentication mechanisms:
- Basic
- Bearer
- OAuth 2.0 Client Credentials grant flow
- API key (only supported in Confluent CLI)
Client-side field level encryption (CSFLE) support: The connector supports CSFLE for sensitive data. For more information about CSFLE setup, see the Manage CSFLE for connectors.
API error reporting management: You can configure the connector to notify you when an API error occurs through email or through the Confluent Cloud user interface. You also can configure the connector to ignore when an API error occurs.
API validation: The connector allows you to test the API using a test record and view the test API response in the Confluent Cloud user interface.
Template variables: The connector allows you to use the ${offset} template variable and gives you the ability to substitute template variables in parameters, headers, and body content.
At-least-once delivery: The connector guarantees that records are delivered at-least-once to the Kafka topic.
Supported data formats: The connector supports Schema Registry-based formats: Avro, JSON Schema, and Protobuf data formats. Schema Registry must be enabled to use a Schema Registry-based format. For more information, see Schema Registry Enabled Environments.
Schema Registry and Schema Context support: The connector allows you to map an API to a specific schema context so that you can use the schema context feature in different environments.
Configurable retry functionality: The connector allows you to customize retry settings based on your requirements.

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

Manage CSFLE

If you plan to enable CSFLE for the HTTP Sink V2 connector, take care of the following sensitive information that may get written to your Kafka topics:

Warning

Error topic: The error topic may include sensitive data from the API response.

Limitations

Be sure to review the following information.

For connector limitations, see HTTP Source V2 Connector limitations.
If you plan to use one or more Single Message Transformations (SMTs), see SMT Limitations.
If you plan to use Confluent Cloud Schema Registry, see Schema Registry Enabled Environments.

Cursor pagination with Timestamp mode

Cursor pagination with timestamp mode combines cursor-based pagination with timestamp-based filtering. This mode is particularly useful for APIs that:

Use cursor tokens to paginate through results.
Support timestamp-based filtering parameters to fetch data within specific time windows.
Return timestamp metadata that can be used to construct subsequent requests.

The connector can handle pagination using the Absolute URL mode.

Absolute URL mode

When enabled (is.next.page.token.absolute.url=true), the connector treats the next page token as a complete, absolute URL that can be used directly for the next request.

Configuration payload

The following configuration properties are required for this mode:

api1.http.offset.mode: Indicates how offsets are computed and how requests are generated. For Absolute URL mode, this must be set to CURSOR_PAGINATION.
api1.is.next.page.token.absolute.url: Specifies whether the next page token from the API is a full URL (true) or a relative path (false). For Absolute URL mode, set this to true.
api1.http.next.page.json.pointer: The JSON pointer to the value in the response which corresponds to the next page reference (either a page token, or a full URL).
initial.start.timestamp.value: Sets the initial start timestamp value used in place of ${start_time} for the first request.
initial.end.timestamp.value: Sets the initial end timestamp value used in place of ${end_time} for the first request. Defaults to current time value (Instant.now().toEpochMilli()).
http.timestamp.format: Timestamp format used by the API (format of ${start_time} and ${end_time}). Defaults to UNIX_TIMESTAMP_MILLISECONDS. If set to SIMPLE_DATE_FORMAT, the interval format must be DAYS.
http.timestamp.interval.format: Specifies the timestamp format of the interval to define the gap between start time and end time parameters. Defaults to SECONDS.
http.timestamp.interval: Timestamp interval between start time and end time parameters. The default value is 86400 seconds (24 hours).
time.range.request.delay: The time, in milliseconds, to delay a timestamp-based pagination request to avoid potential time drift. Defaults to 0; the maximum value is 60 seconds.

For more information, see Configuration Properties.

In this mode:

The connector uses a URL (For example, https://api.example.com/abcd) from previous the HTTP response as the complete URl for the next request.
No additional parameters are appended to this URL.
The configured request parameters are ignored for subsequent requests.

Chaining pagination with Timestamp mode

Chaining pagination with timestamp mode combines chaining-based pagination with timestamp-based filtering. This mode is particularly useful for APIs that:

Use chaining-based pagination to traverse through results sequentially.
Support timestamp-based filtering parameters to fetch data within specific time windows.
Return timestamp metadata that can be used to construct subsequent requests.
Provide either relative or global ordering guarantees for records.

Prerequisites

To use chaining pagination with timestamp mode, you must configure the following properties:

Offset mode: Must be set to CHAINING.
Chaining Timestamp Mode: Set chaining timestamp mode to true (chaining.with.timestamp.mode=true).
Record Ordering: Choose between GLOBAL or RELATIVE ordering.

Record ordering modes

The chaining.with.timestamp.mode.record.ordering configuration determines how records are ordered across API queries. Chaining pagination with timestamp mode supports two types of record ordering:

Global
Relative

Global ordering

To use global ordering, set the chaining.with.timestamp.mode.record.ordering to GLOBAL. You can use global ordering when the source API guarantees a consistent ordering across all queries, regardless of time windows.

Configuration payload

chaining.with.timestamp.mode.record.ordering (required): Specifies the record ordering mode. This must be set to GLOBAL.
http.offset.json.primary.key.pointer (required): Specifies the JSON pointer to the primary key field in the HTTP response. The extracted value is used as the query parameter for http.chaining.pagination.parameter.name.
http.chaining.pagination.parameter.name (required): Specifies the name of the query parameter used in the request for chaining with timestamp pagination. This is used when fetching the next page of results.
initial.start.timestamp.value (required): Sets the initial start timestamp value used in place of ${start_time} for the first request.
initial.end.timestamp.value (optional): Sets the initial end timestamp value used in place of ${end_time} for the first request.
timestamp.format (required): Timestamp format used in the time related request parameters. It should match the format of initial.start.timestamp.value and initial.end.timestamp.value Supported Formats:
- UNIX_TIMESTAMP_MILLISECONDS: “1640995200000” (milliseconds since epoch).
- UNIX_TIMESTAMP_SECONDS: “1640995200” (seconds since epoch).
- SIMPLE_DATE_FORMAT: “2022-01-01” (yyyy-MM-dd format).
- ISO8601: “2022-01-01T00:00:00Z”, “2022-01-01T14:30:00Z”, “2022-01-01T14:30:00+05:30” (with time zone support).
http.timestamp.interval (optional): Timestamp interval between start time and end time parameters. The default value is 86400 seconds (24 hours).
http.timestamp.interval.format (optional): Specifies the timestamp format of the interval to define the gap between start time and end time parameters. Defaults to SECONDS.
time.range.request.delay (optional): The time in milliseconds to delay a timestamp-based pagination request to avoid potential time drift. Defaults to 0. The maximum value is 60 seconds.

Relative ordering

To use relative ordering, set the chaining.with.timestamp.mode.record.ordering to RELATIVE. You can use relative ordering when the source API provides ordering only within individual queries or time windows, where the order may vary across separate queries.

Configuration payload

chaining.with.timestamp.mode.record.ordering (required): Specifies the record ordering mode. This must be set to RELATIVE.
http.offset.json.primary.key.pointer (required): Specifies the JSON pointer to the primary key field in the HTTP response. The extracted value is used as the query parameter for http.chaining.pagination.parameter.name.
time.range.request.delay (optional): The time in milliseconds to delay a timestamp-based pagination request to avoid potential time drift. Defaults to 0. The maximum value is 60 seconds.

Error records and status codes

The connector may report status_code = -1 in error records when no HTTP response code is available. This usually happens with client-side or transport failures that occur before the server can respond. For example, DNS failures, connection refused errors, socket connect/read timeouts, or TLS handshake errors.

When the server responds with a non-2xx status, the connector reports the actual HTTP code, for example, 400 or 500.

If the connector receives a 2xx response but fails to parse or process the response body, it resets the internal status to -1 for error reporting in source error records. The status_code then appears as null. For diagnostics, use accompanying fields like reason_phrase, response_content, error_message, and exception.

Test API

Use the Test API functionality to test the API with a sample record and view the logs directly in the Confluent Cloud user interface.

Important

This feature is only available for publicly accessible endpoints.
Invoking the Test API on an API may change data on the end system, depending on the API’s behavior.

When using this feature with the HTTP Source V2 connector, configure the following required parameter in the user interface: ${offset}. This is the offset that the connector should read from. The connector uses this value to replace the ${offset} template variable wherever it is configured.

API Chaining

API Chaining refers to obtaining the desired data, by navigating through multiple APIs. The response from the first API determines the value for the template placeholders in the URL for the next API, thereby creating a ‘chain.’ For example, let’s assume there is a SaaS system that supports the following APIs:

Parent API

  GET /users : fetches all the users registered in the service

Child API

  GET /users/${id}/places : fetches all the places visited by a particular user

If the requirement is to create a connector that captures all the places visited by registered users in the system, the steps would broadly include:

Retrieve all users in the system using the /users endpoint.
Obtain all places visited by calling the /users/${id}/places API for each user fetched in the first step.

The configuration property, api.chaining.parent.child.relationship can be used to define the relationship between the APIs. For more information, see Configuration Properties.

Considerations

Important points to note with respect to API Chaining:

This feature can only be used with API/CLI/TF.
When this feature is enabled, an additional metadata topic is created in your cluster, and you are responsible for the associated costs.
The value for the placeholders in the child API path (like ${id} in above example) and the offset field values are stored in the metadata topic in unencrypted form. If you are using CSFLE with the HTTP Source V2 connector, you need to ensure that the placeholder and the offset fields do not contain Personally Identifiable Information (PII) data.
Currently, only one level of chaining is supported. This means that a child API can not have a child of its own.

Limitations

The ordering between the records may not be maintained in API chaining.
The connector may produce duplicate records in the Kafka topic.
Custom offsets are not supported in API chaining (for both parent and child APIs).
The APIs used for API chaining must support pagination.

Permissions

You need to provide permissions for additional metadata topic and consumer group created as part of API chaining:

Metadata topic is prefixed with HTTPv2-work-queue-.
Consumer group is prefixed with HTTPv2-work-queue-consumers-.

SNAPSHOT_PAGINATION mode

Some API’s doesn’t allow offset pagination (like /api/path/${offset}), instead it provides a static URL which can have incremental records over time as shown in the example below:

GET /api/path/users at time t1       → R1 to R10
GET /api/path/users at time t2 (>t1) → R1 to R20
GET /api/path/users at time t3 (>t2) → R1 to R27

In such cases, you can use SNAPSHOT_PAGINATION mode to paginate the data. You can use the api<index number>.http.offset.mode configuration to set its value.

Note

All the records should be in sorted order according to some unique key. The same key is used to set under api<index number>.http.offset.json.pointer configuration.

Configuration Properties

Use the following configuration properties with the fully-managed HTTP V2 Source connector. When configuring your APIs, use the configuration properties listed in the API-1 Config section. You’ll just need to update the prefix name for each of the configurations accordingly. For example, for the http.request.headers parameter, it would look similar to: api<indexnumber>.http.request.headers (for example, api2.http.request.headers, api3.http.request.headers, and so forth)

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

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

Output messages

output.data.format

Sets the output Kafka record value format. Valid entries are AVRO, JSON_SR, or PROTOBUF. Please configure Confluent Cloud Schema Registry.

Type: string
Default: JSON_SR
Importance: high

Number of tasks for this connector

tasks.max

Maximum number of tasks for the connector.

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

Behavior on error

behavior.on.error

Error handling behavior setting for handling error response from HTTP requests.

Type: string
Default: FAIL
Importance: low

Authentication

http.api.base.url

The HTTP API Base URL. For example: http://example.com/absenceManagement/v1.

Type: string
Importance: high

auth.type

Authentication type of the endpoint. Valid values are NONE, BASIC, OAUTH2 (Client Credentials grant type only), BEARER, API_KEY

Type: string
Default: NONE
Importance: high

connection.user

The username to be used with an endpoint requiring basic authentication.

Type: string
Importance: medium

connection.password

The password to be used with an endpoint requiring basic authentication.

Type: password
Importance: medium

bearer.token

The bearer authentication token to be used with an endpoint requiring bearer token based authentication.

Type: password
Importance: medium

oauth2.token.url

The URL to be used for fetching the OAuth2 token. Client Credentials is the only supported grant type.

Type: string
Importance: medium

oauth2.client.id

The client id used when fetching the OAuth2 token.

Type: string
Importance: medium

oauth2.client.secret

The client secret used when fetching the OAuth2 token.

Type: password
Importance: medium

oauth2.token.property

The name of the property containing the OAuth2 token returned by the OAuth2 token URL (defaults to access_token).

Type: string
Default: access_token
Importance: medium

oauth2.client.scope

The scope parameter sent to the service when fetching the OAuth2 token.

Type: string
Default: any
Importance: medium

oauth2.client.auth.mode

Specifies how to encode client_id and client_secret in the OAuth2 authorization request. If set to header, the credentials are encoded as an ‘Authorization: Basic <base-64 encoded client_id:client_secret>’ HTTP header. If set to ‘url’, then client_id and client_secret are sent as URL encoded parameters.

Type: string
Default: header
Importance: medium

oauth2.client.headers

HTTP headers to be included in the OAuth2 client endpoint. Individual headers should be separated by |. This configuration can be set via CLI only.

Type: password
Importance: low

oauth2.client.header.separator

Separator character used in OAuth2 Client Headers. This configuration can be set via CLI only.

Type: string
Default: |
Importance: low

https.ssl.enabled

Controls whether to enforce TLSv1.3 for HTTPS connections. If false, the connector uses the JVM default TLS configuration (typically TLSv1.2). If true, TLSv1.3 is enforced. Note that, even when disabled, HTTPS connections still use TLS. This setting only controls TLS version enforcement.

Type: boolean
Default: false
Importance: medium

https.ssl.keystorefile

The keystore that contains the client certificate and private key. Supported formats are JKS or PKCS12. File system paths are not supported.

Type: password
Default: [hidden]
Importance: low

https.ssl.keystore.password

The store password for the key store file.

Type: password
Importance: high

https.ssl.key.password

The password for the private key in the key store file.

Type: password
Importance: high

https.ssl.truststorefile

The truststore that contains the server CA certificate(s). Supported formats are JKS or PKCS12. File system paths are not supported.

Type: password
Default: [hidden]
Importance: high

https.ssl.truststore.password

The trust store password containing a server CA certificate.

Type: password
Importance: high

https.ssl.protocol

The protocol to use for SSL connections

Type: string
Default: TLSv1.3
Importance: medium

api.key.location

Specifies where the API key is included in the HTTP request. Choose HEADER to send the API key as an HTTP header (for example, X-API-KEY: your-api-key), or QUERY to include it as a query parameter in the URL (for example, ?api_key=your-api-key). The default is HEADER which is the most common and secure approach.

Type: string
Default: HEADER
Importance: medium

api.key.name

The identifier for the API key used in authentication. The format depends on the api.key.location configuration. If location is HEADER, use the HTTP header name (for example, X-API-KEY). If location is QUERY, use the query parameter name (for example, api_key, app_id). The default value X-API-KEY is commonly used for header-based authentication.

Type: string
Default: X-API-KEY
Importance: medium

api.key.value

The API key to be used with an endpoint that requires API key-based authentication

Type: password
Importance: medium

http.proxy.host

The host or IP address of the HTTP proxy.

Type: string
Importance: medium

http.proxy.port

The port number of the HTTP proxy.

Type: int
Valid Values: [0,…,65535]
Importance: medium

http.proxy.user

The username to be used for proxy authentication.

Type: string
Importance: medium

http.proxy.password

The password to be used for proxy authentication.

Type: password
Importance: medium

APIs

apis.num

The number of http(s) APIs to configure. This value should be less than or equal to 15

Type: int
Default: 1
Importance: high

api.chaining.parent.child.relationship

Comma separated list of parent-child relationship in case of API Chaining. For ex - If api1 is the parent API, and api2,api3 are the child APIs of api1, then the value will be set to api1::api2,api1::api3.

Type: string
Default: “”
Importance: high

Reporter

reporter.error.topic.name

The name of the topic to produce errant records after each unsuccessful API operation. Defaults to ‘error-${connector}’ if not set. You can provide ${connector} in the value to use it as a placeholder for the logical cluster ID.

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

report.errors.as

Dictates the content of records produced to the error topic. If set to Error string the value would be a human readable string describing the failure. The value will include some or all of the following information if available: http response code, reason phrase, submitted payload, url, response content, exception and error message. If set to http_response, the value would be the plain response content for the request which failed to write the record. In both modes, any information about the failure will also be included in the error records headers.

Type: string
Default: Error string
Importance: low

Additional Configs

header.converter

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

Type: string
Importance: low

producer.override.compression.type

The compression type for all data generated by the producer. Valid values are none, gzip, snappy, lz4, and zstd.

Type: string
Importance: low

producer.override.linger.ms

The producer groups together any records that arrive in between request transmissions into a single batched request. More details can be found in the documentation: https://docs.confluent.io/platform/current/installation/configuration/producer-configs.html#linger-ms.

Type: long
Valid Values: [100,…,1000]
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: none
Importance: low

key.converter.key.subject.name.strategy

How to construct the subject name for key schema registration.

Type: string
Default: TopicNameStrategy
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.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

API-1 Configs

api1.http.api.path

The HTTP API path together with the ‘http.api.base.url’ will form the complete HTTP(S) URL. This path can be templated with offset information. For example: /resource1/${offset} where ${offset} will be substituted with the offset generated from the previous request’s response (or if it’s the first request, from ‘http.initial.offset’).

Type: string
Importance: high

api1.topics

The name of the Kafka topics to publish data to.

Type: string
Importance: high

api1.http.request.headers

HTTP headers to be included in each request. Header names and values should be separated by :. Distinct headers should be separated by |. For example: From:abcxyz@confluent.io|Content-Length:348.

Type: string
Importance: medium

api1.http.request.method

HTTP Request Method. Valid options are GET and POST.

Type: string
Default: GET
Importance: high

api1.http.request.parameters

HTTP parameters to be added to each request. Parameter names and values should be separated by =. Distinct parameters should be separated by &. Parameter values can be templated with offset information (for example: search_after=${offset}) where ${offset} will be substituted with the offset generated from the previous request’s response (or if it’s the first request, from ‘http.initial.offset’). The parameters are only set if ‘http.request.method’ = GET.

Type: string
Importance: medium

api1.http.connect.timeout.ms

The time in milliseconds to wait for a connection to be established

Type: int
Default: 30000 (30 seconds)
Valid Values: [1,…,600000]
Importance: medium

api1.http.request.body

The payload to be sent along with each HTTP request. The value can be templated with offset (for example: search_after: ${offset}) where ${offset} will be substituted with the offset generated from the previous request’s response (or if it’s the first request, from ‘http.initial.offset’). The body is only set if ‘http.request.method’ = POST.

Type: string
Importance: medium

api1.http.request.timeout.ms

The time in milliseconds to wait for a request response from the server

Type: int
Default: 30000 (30 seconds)
Valid Values: [1,…,600000]
Importance: medium

api1.http.offset.mode

This config indicates how offsets are computed and how requests are generated. If set to SIMPLE_INCREMENTING, the ${offset} used to generate requests is simply the previous offset (or http.initial.offset) incremented by 1 per sourced record. In this mode, http.initial.offset needs to be set to an integer value. If set to CHAINING, the config ‘http.offset.json.pointer’ needs to be set, and the offset for a record is set to the value for the configured key in the response data. If the value is CURSOR_PAGINATION, then the config ‘http.next.page.json.pointer’ needs to be set and the offset for the last record in each page will be set to the next page value. If set to SNAPSHOT_PAGINATION, only the config ‘http.offset.json.pointer’ needs to be set.

Type: string
Default: SIMPLE_INCREMENTING
Importance: high

api1.max.retries

The maximum number of times to retry on errors before failing the task.

Type: int
Default: 5
Importance: medium

api1.http.initial.offset

The initial offset to be used to generate the first request. This needs to be set if either one or more of the configs - ‘url’, ‘http.request.parameters’, or ‘http.request.body’ contain the template variable ${offset}.

Type: string
Default: “”
Importance: high

api1.retry.backoff.policy

The backoff policy to use in terms of retry - CONSTANT_VALUE or EXPONENTIAL_WITH_JITTER

Type: string
Default: EXPONENTIAL_WITH_JITTER
Importance: medium

api1.http.response.data.json.pointer

The JSON Pointer to the entity in the JSON response containing the actual data that should be written to Kafka as records. The entity can be an array (multiple records in a single response) or an object / scalar value (single record).

Type: string
Importance: high

api1.retry.backoff.ms

The initial duration in milliseconds to wait following an error before a retry attempt is made. Subsequent backoff attempts can be a constant value or exponential with jitter (can be configured using api*.retry.backoff.policy parameter). Jitter adds randomness to the exponential backoff algorithm to prevent synchronized retries.

Type: int
Default: 3000 (3 seconds)
Valid Values: [20,…]
Importance: medium

api1.http.offset.json.pointer

The JSON Pointer to the value in each record that corresponds to the offset for that record (it is relative to ‘http.response.data.json.pointer’). The offset will be available to the subsequent request as ${offset} and it will also be used for checkpointing and recovery in case of connector failures or restarts. This config should only be set if ‘http.offset.mode’ is set to CHAINING.

Type: string
Importance: medium

api1.retry.on.status.codes

Comma-separated list of HTTP status codes or range of codes to retry on. Ranges are specified with start and optional end code. Range boundaries are inclusive. For instance, 400- includes all codes greater than or equal to 400. 400-500 includes codes from 400 to 500, including 500. Multiple ranges and single codes can be specified together to achieve fine-grained control over retry behavior. For example, 404,408,500- will retry on 404 NOT FOUND, 408 REQUEST TIMEOUT, and all 5xx error codes. Note that some status codes will always be retried, such as unauthorized, timeouts and too many requests.

Type: string
Default: 400-
Importance: medium

api1.http.next.page.json.pointer

The JSON pointer to the value in the response which corresponds to the next page reference (either a page token, a full URL or a URL fragment). This will be stored as the offset and will be available to the subsequent request via the template variable ${offset}. This config should only be set if ‘http.offset.mode’ is set to CURSOR_PAGINATION. The referenced value can be either a string or an integer.

Type: string
Importance: medium

api1.http.request.headers.separator

The character that separates multiple distinct headers within http.request.headers. This configuration can be set via CLI only.

Type: string
Default: |
Importance: medium

api1.http.request.parameters.separator

The character that separates multiple distinct request parameters within http.request.parameters. This configuration can be set via CLI only.

Type: string
Default: &
Importance: medium

api1.request.interval.ms

The time in milliseconds to wait between consecutive requests.

Type: int
Default: 60000 (1 minute)
Valid Values: [100,…]
Importance: medium

api1.enable.cursor.pagination.with.timestamp.mode

Whether to use cursor pagination with timestamp mode.

Type: boolean
Default: false
Importance: medium

api1.http.timestamp.json.pointer

JSON pointer to the timestamp field in the response. It is used to construct time related parameters for the subsequent API calls.

Type: string
Default: “”
Importance: medium

api1.initial.start.timestamp.value

The initial start timestamp to be used to generate the first request. This needs to be set when using cursor.pagination.with.timestamp.mode or chaining.with.timestamp.mode with chaining.with.timestamp.mode.record.ordering set to Global.

Type: string
Default: “”
Importance: medium

api1.http.request.sensitive.headers

Sensitive HTTP headers to include in each request. Sensitive header names and values should be separated by :. Distinct headers should be separated by |. For example: From:abcxyz@confluent.io|Content-Length:348. If a key is present in both regular and sensitive headers, the value in sensitive headers takes priority. This is a CLI-only feature.

Type: password
Importance: medium

api1.http.cursor.pagination.parameter.name

Name of the request parameter which holds the next page token.

Type: string
Default: “”
Importance: medium

api1.initial.end.timestamp.value

Sets the initial end timestamp to generate the first request when using enable.cursor.pagination.with.timestamp.mode or chaining.with.timestamp.mode with chaining.with.timestamp.mode.record.ordering set to Global.

Type: string
Default: “”
Importance: medium

api1.timestamp.format

Timestamp format used in the time related request parameters. It should match the format of initial.start.timestamp.value and initial.end.timestamp.value

Type: string
Default: Unix Timestamp Milliseconds
Importance: medium

api1.timestamp.interval.format

Format of the timestamp interval provided.

Type: string
Default: Seconds
Importance: medium

api1.http.timestamp.interval

Timestamp interval between start time and end time.

Type: long
Default: 86400
Importance: medium

api1.time.range.request.delay

The time in milliseconds to delay a timestamp-based pagination request to avoid potential time drift.

Type: long
Default: 0
Valid Values: [0,…,600000]
Importance: medium

api1.is.next.page.token.absolute.url

Whether next page token is an absolute url or not.

Type: boolean
Default: true
Importance: medium

api1.http.path.parameters

HTTP path parameters to be added to the request. Parameter names and values should be separated by :. Distinct parameters should be separated by |. Parameter values can be templated with different template values like ${offset}.

Type: string
Importance: medium

api1.http.path.parameters.separator

The character that separates multiple distinct path parameters within http.path.parameters. This configuration can be set via CLI only.

Type: string
Default: |
Importance: medium

api1.http.response.schema.enforce

Controls how strictly the connector enforces the HTTP response schema. When set to STRICT, the connector validates that the response conforms to the defined OpenAPI response schema. For both missing non-optional and new fields, the connector handles the error (fail or ignore) based on the behavior.on.error setting. When set to NONE, the connector does not enforce the OpenAPI schema and incase non-optional field is missing, connector deduces the schema for the response dynamically otherwise connector registers new fields as optional strings.

Type: string
Default: NONE
Importance: medium

api1.chaining.with.timestamp.mode

Whether to use chaining with timestamp mode.

Type: boolean
Default: false
Importance: medium

api1.chaining.with.timestamp.mode.record.ordering

Specifies how records are ordered during chaining when using timestamp mode. Use Relative when the source API provides ordering only within a time window, where the order may vary across queries. Use Global when the source API guarantees a consistent ordering across all queries.

Type: string
Default: Global
Importance: medium

api1.http.chaining.pagination.parameter.name

Specifies the name of the query parameter used in the request for chaining with time window pagination. This is typically used when fetching the next page of results using values like starting_after.

Type: string
Default: “”
Importance: medium

api1.http.offset.json.primary.key.pointer

Defines the JSON Pointer to the offset value within each record (relative to http.response.data.json.pointer). This offset becomes ${primary_key} for subsequent requests and enables checkpointing and recovery if the connector fails or restarts. Only applicable when http.offset.mode is set to CHAINING.

Type: string
Default: “”
Importance: medium

Auto-restart policy

auto.restart.on.user.error

Enable connector to automatically restart on user-actionable errors.

Type: boolean
Default: true
Importance: medium

Egress allowlist

connector.egress.whitelist

Type: string
Default: “”
Importance: high

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.

HTTP Source V2 Connector for Confluent Cloud

Features

Manage CSFLE

Limitations

Manage custom offsets

Chaining offset mode

Simple incrementing offset mode

Cursor pagination offset mode

Cursor pagination with Timestamp mode

Absolute URL mode

Chaining pagination with Timestamp mode

Prerequisites

Record ordering modes

Global ordering

Relative ordering

Quick Start

Prerequisites

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 for records

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 properties file and create the connector

Step 5: Check the connector status

Step 6: Check for records

Error records and status codes

Test API

API Chaining

Considerations

Limitations

Permissions

SNAPSHOT_PAGINATION mode

Configuration Properties

How should we connect to your data?

Kafka Cluster credentials

Schema Config

Output messages

Number of tasks for this connector

Behavior on error

Authentication

APIs

Reporter

Additional Configs

API-1 Configs

Auto-restart policy

Egress allowlist

Suggested Reading

Next Steps