KSQL REST API Reference

REST Endpoint

The default REST API endpoint is http://0.0.0.0:8088/.

Change the server configuration that controls the REST API endpoint by setting the listeners parameter in the KSQL server config file. For more info, see listeners. To configure the endpoint to use HTTPS, see Configuring KSQL for HTTPS.

Content Types

The KSQL REST API uses content types for requests and responses to indicate the serialization format of the data and the API version. Currently, the only serialization format supported is JSON. The only version supported is v1. Your request should specify this serialization format and version in the Accept header as:

Accept: application/vnd.ksql.v1+json

The less specific application/json content type is also permitted. However, this is only for compatibility and ease of use, and you should use the versioned value if possible.

The server also supports content negotiation, so you may include multiple, weighted preferences:

Accept: application/vnd.ksql.v1+json; q=0.9, application/json; q=0.5

For example, content negotiation is useful when a new version of the API is preferred, but you are not sure if it is available yet.

Here's an example request that returns the results from the LIST STREAMS command:

curl -X "POST" "http://localhost:8088/ksql" \
     -H "Content-Type: application/vnd.ksql.v1+json; charset=utf-8" \
     -d $'{
  "ksql": "LIST STREAMS;",
  "streamsProperties": {}
}'

Here's an example request that retrieves streaming data from TEST_STREAM:

curl -X "POST" "http://localhost:8088/query" \
     -H "Content-Type: application/vnd.ksql.v1+json; charset=utf-8" \
     -d $'{
  "ksql": "SELECT * FROM TEST_STREAM;",
  "streamsProperties": {}
}'

Errors

All API endpoints use a standard error message format for any requests that return an HTTP status indicating an error (any 4xx or 5xx statuses):

HTTP/1.1 <Error Status>
Content-Type: application/json

{
    "error_code": <Error code>
    "message": <Error Message>
}

Some endpoints may include additional fields that provide more context for handling the error.

Get the Status of a KSQL Server

The /info resource gives you information about the status of a KSQL server, which can be useful for health checks and troubleshooting. You can use the curl command to query the /info endpoint:

curl -sX GET "http://localhost:8088/info" | jq '.'

Your output should resemble:

{
  "KsqlServerInfo": {
    "version": "5.3.1",
    "kafkaClusterId": "j3tOi6E_RtO_TMH3gBmK7A",
    "ksqlServiceId": "default_"
  }
}

Run a KSQL Statement

The /ksql resource runs a sequence of KSQL statements. All statements, except those starting with SELECT, can be run on this endpoint. To run SELECT statements use the /query endpoint.

Note

If you use the SET or UNSET statements to assign query properties by using the REST API, the assignment is scoped only to the current request. In contrast, SET and UNSET assignments in the KSQL CLI persist throughout the CLI session.

POST /ksql

Run a sequence of KSQL statements.

JSON Parameters:
 
  • ksql (string) -- A semicolon-delimited sequence of KSQL statements to run.
  • streamsProperties (map) -- Property overrides to run the statements with. Refer to the Config Reference for details on properties that can be set.
  • streamsProperties[property-name] (string) -- The value of the property named by property-name. Both the value and property-name should be strings.
  • commandSequenceNumber (long) -- Optional. If specified, the statements will not be run until all existing commands up to and including the specified sequence number have completed. If unspecified, the statements are run immediately. (Note: When a command is processed, the result object contains its sequence number.)

The response JSON is an array of result objects. The result object contents depend on the statement that it is returning results for. The following sections detail the contents of the result objects by statement.

Common Fields The following fields are common to all responses.

Response JSON Object:
 
  • statementText (string) -- The KSQL statement whose result is being returned.
  • warnings (array) -- A list of warnings about conditions that may be unexpected by the user, but don't result in failure to execute the statement.
  • warnings[i].message (string) -- A message detailing the condition being warned on.

CREATE, DROP, TERMINATE

Response JSON Object:
 
  • commandId (string) -- A string that identifies the requested operation. You can use this ID to poll the result of the operation using the status endpoint.
  • commandStatus.status (string) -- One of QUEUED, PARSING, EXECUTING, TERMINATED, SUCCESS, or ERROR.
  • commandStatus.message (string) -- Detailed message regarding the status of the execution statement.
  • commandSequenceNumber (long) -- The sequence number of the requested operation in the command queue, or -1 if the operation was unsuccessful.

LIST STREAMS, SHOW STREAMS

Response JSON Object:
 
  • streams (array) -- List of streams.
  • streams[i].name (string) -- The name of the stream.
  • streams[i].topic (string) -- The topic backing the stream.
  • streams[i].format (string) -- The serialization format of the data in the stream. One of JSON, AVRO, or DELIMITED.

LIST TABLES, SHOW TABLES

Response JSON Object:
 
  • tables (array) -- List of tables.
  • tables[i].name (string) -- The name of the table.
  • tables[i].topic (string) -- The topic backing the table.
  • tables[i].format (string) -- The serialization format of the data in the table. One of JSON, AVRO, or DELIMITED.

LIST QUERIES, SHOW QUERIES

Response JSON Object:
 
  • queries (array) -- List of queries.
  • queries[i].queryString (string) -- The text of the statement that started the query.
  • queries[i].sinks (string) -- The streams and tables being written to by the query.
  • queries[i].id (string) -- The query ID.

LIST PROPERTIES, SHOW PROPERTIES

Response JSON Object:
 
  • properties (map) -- The KSQL server query properties.
  • properties[property-name] (string) -- The value of the property named by property-name.

DESCRIBE

Response JSON Object:
 
  • sourceDescription.name (string) -- The name of the stream or table.
  • sourceDescription.readQueries (array) -- The queries reading from the stream or table.
  • sourceDescription.writeQueries (array) -- The queries writing into the stream or table
  • sourceDescription.fields (array) -- A list of field objects that describes each field in the stream/table.
  • sourceDescription.fields[i].name (string) -- The name of the field.
  • sourceDescription.fields[i].schema (object) -- A schema object that describes the schema of the field.
  • sourceDescription.fields[i].schema.type (string) -- The type the schema represents. One of INTEGER, BIGINT, BOOLEAN, DOUBLE, STRING, MAP, ARRAY, or STRUCT.
  • sourceDescription.fields[i].schema.memberSchema (object) -- A schema object. For MAP and ARRAY types, contains the schema of the map values and array elements, respectively. For other types this field is not used and its value is undefined.
  • sourceDescription.fields[i].schema.fields (array) -- For STRUCT types, contains a list of field objects that descrbies each field within the struct. For other types this field is not used and its value is undefined.
  • sourceDescription.type (string) -- STREAM or TABLE
  • sourceDescription.key (string) -- The name of the key column.
  • sourceDescription.timestamp (string) -- The name of the timestamp column.
  • sourceDescription.format (string) -- The serialization format of the data in the stream or table. One of JSON, AVRO, or DELIMITED.
  • sourceDescription.topic (string) -- The topic backing the stream or table.
  • sourceDescription.extended (boolean) -- A boolean that indicates whether this is an extended description.
  • sourceDescription.statistics (string) -- A string that contains statistics about production and consumption to and from the backing topic (extended only).
  • sourceDescription.errorStats (string) -- A string that contains statistics about errors producing and consuming to and from the backing topic (extended only).
  • sourceDescription.replication (int) -- The replication factor of the backing topic (extended only).
  • sourceDescription.partitions (int) -- The number of partitions in the backing topic (extended only).

EXPLAIN

Response JSON Object:
 
  • queryDescription.statementText (string) -- The KSQL statement for which the query being explained is running.
  • queryDescription.fields (array) -- A list of field objects that describes each field in the query output.
  • queryDescription.fields[i].name (string) -- The name of the field.
  • queryDescription.fields[i].schema (object) -- A schema object that describes the schema of the field.
  • queryDescription.fields[i].schema.type (string) -- The type the schema represents. One of INTEGER, BIGINT, BOOLEAN, DOUBLE, STRING, MAP, ARRAY, or STRUCT.
  • queryDescription.fields[i].schema.memberSchema (object) -- A schema object. For MAP and ARRAY types, contains the schema of the map values and array elements, respectively. For other types this field is not used and its value is undefined.
  • queryDescription.fields[i].schema.fields (array) -- For STRUCT types, contains a list of field objects that descrbies each field within the struct. For other types this field is not used and its value is undefined.
  • queryDescription.sources (array) -- The streams and tables being read by the query.
  • queryDescription.sources[i] (string) -- The name of a stream or table being read from by the query.
  • queryDescription.sinks (array) -- The streams and tables being written to by the query.
  • queryDescription.sinks[i] (string) -- The name of a stream or table being written to by the query.
  • queryDescription.executionPlan (string) -- They query execution plan.
  • queryDescription.topology (string) -- The Kafka Streams topology that the query is running.
  • overriddenProperties (map) -- The property overrides that the query is running with.

Errors

If KSQL fails to execute a statement, it returns a response with an error status code (4xx/5xx). Even if an error is returned, the server may have been able to successfully execute some statements in the request. In this case, the response includes the error_code and message fields, a statementText field with the text of the failed statement, and an entities field that contains an array of result objects:

Response JSON Object:
 
  • statementText (string) -- The text of the KSQL statement where the error occurred.
  • entities (array) -- Result objects for statements that were successfully executed by the server.

The /ksql endpoint may return the following error codes in the error_code field:

  • 40001 (BAD_STATEMENT): The request contained an invalid KSQL statement.
  • 40002 (QUERY_ENDPOINT): The request contained a statement that should be issued to the /query endpoint.

Example request

POST /ksql HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "ksql": "CREATE STREAM pageviews_home AS SELECT * FROM pageviews_original WHERE pageid='home'; CREATE STREAM pageviews_alice AS SELECT * FROM pageviews_original WHERE userid='alice';",
  "streamsProperties": {
    "ksql.streams.auto.offset.reset": "earliest"
  }
}

Example response

HTTP/1.1 200 OK
Content-Type: application/vnd.ksql.v1+json

[
  {
    "statementText":"CREATE STREAM pageviews_home AS SELECT * FROM pageviews_original WHERE pageid='home';",
    "commandId":"stream/PAGEVIEWS_HOME/create",
    "commandStatus": {
      "status":"SUCCESS",
      "message":"Stream created and running"
    },
    "commandSequenceNumber":10
  },
  {
    "statementText":"CREATE STREAM pageviews_alice AS SELECT * FROM pageviews_original WHERE userid='alice';",
    "commandId":"stream/PAGEVIEWS_ALICE/create",
    "commandStatus": {
      "status":"SUCCESS",
      "message":"Stream created and running"
    },
    "commandSequenceNumber":11
  }
]

Coordinate Multiple Requests

To submit multiple, interdependent requests, there are two options. The first is to submit them as a single request, similar to the example request above:

POST /ksql HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "ksql": "CREATE STREAM pageviews_home AS SELECT * FROM pageviews_original WHERE pageid='home'; CREATE TABLE pageviews_home_count AS SELECT userid, COUNT(*) FROM pageviews_home GROUP BY userid;"
}

The second method is to submit the statements as separate requests and incorporate the interdependency by using commandSequenceNumber. Send the first request:

POST /ksql HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "ksql": "CREATE STREAM pageviews_home AS SELECT * FROM pageviews_original WHERE pageid='home';"
}

Make note of the commandSequenceNumber returned in the response:

HTTP/1.1 200 OK
Content-Type: application/vnd.ksql.v1+json

[
  {
    "statementText":"CREATE STREAM pageviews_home AS SELECT * FROM pageviews_original WHERE pageid='home';",
    "commandId":"stream/PAGEVIEWS_HOME/create",
    "commandStatus": {
      "status":"SUCCESS",
      "message":"Stream created and running"
    },
    "commandSequenceNumber":10
  }
]

Provide this commandSequenceNumber as part of the second request, indicating that this request should not execute until after command number 10 has finished executing:

POST /ksql HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "ksql": "CREATE TABLE pageviews_home_count AS SELECT userid, COUNT(*) FROM pageviews_home GROUP BY userid;",
  "commandSequenceNumber":10
}

Run A Query And Stream Back The Output

The /query resource lets you stream the output records of a SELECT statement via a chunked transfer encoding. The response is streamed back until the LIMIT specified in the statement is reached, or the client closes the connection. If no LIMIT is specified in the statement, then the response is streamed until the client closes the connection.

POST /query

Run a SELECT statement and stream back the results.

JSON Parameters:
 
  • ksql (string) -- The SELECT statement to run.
  • streamsProperties (map) -- Property overrides to run the statements with. Refer to the Config Reference for details on properties that can be set.
  • streamsProperties[property-name] (string) -- The value of the property named by property-name. Both the value and property-name should be strings.

Each response chunk is a JSON object with the following format:

Response JSON Object:
 
  • row (object) -- A single row being returned. This will be null if an error is being returned.
  • row.columns (array) -- The values contained in the row.
  • row.columns[i] (?) -- The value contained in a single column for the row. The value type depends on the type of the column.
  • finalMessage (string) -- If this field is non-null, it contains a final message from the server. No additional rows will be returned and the server will end the response.
  • errorMessage (string) -- If this field is non-null, an error has been encountered while running the statement. No additional rows are returned and the server will end the response.

Example request

POST /query HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "ksql": "SELECT * FROM pageviews;",
  "streamsProperties": {
    "ksql.streams.auto.offset.reset": "earliest"
  }
}

Example response

HTTP/1.1 200 OK
Content-Type: application/vnd.ksql.v1+json
Transfer-Encoding: chunked

...
{"row":{"columns":[1524760769983,"1",1524760769747,"alice","home"]},"errorMessage":null}
...

Get the Status of a CREATE, DROP, or TERMINATE

CREATE, DROP, and TERMINATE statements returns an object that indicates the current state of statement execution. A statement can be in one of the following states:

  • QUEUED, PARSING, EXECUTING: The statement was accepted by the server and is being processed.
  • SUCCESS: The statement was successfully processed.
  • ERROR: There was an error processing the statement. The statement was not executed.
  • TERMINATED: The query started by the statement was terminated. Only returned for CREATE STREAM|TABLE AS SELECT.

If a CREATE, DROP, or TERMINATE statement returns a command status with state QUEUED, PARSING, or EXECUTING from the /ksql endpoint, you can use the /status endpoint to poll the status of the command.

GET /status/(string: commandId)

Get the current command status for a CREATE, DROP, or TERMINATE statement.

Parameters:
  • commandId (string) -- The command ID of the statement. This ID is returned by the /ksql endpoint.
Response JSON Object:
 
  • status (string) -- One of QUEUED, PARSING, EXECUTING, TERMINATED, SUCCESS, or ERROR.
  • message (string) -- Detailed message regarding the status of the execution statement.

Example request

GET /status/stream/PAGEVIEWS/create HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

Example response

HTTP/1.1 200 OK
Content-Type application/vnd.ksql.v1+json

{
  "status": "SUCCESS",
  "message":"Stream created and running"
}

Terminate a Cluster

If you don't need your KSQL cluster anymore, you can terminate the cluster and clean up the resources using this endpoint. To terminate a KSQL cluster, first shutdown all the servers except one. Then, send the TERMINATE CLUSTER request to the /ksql/terminate endpoint in the last remaining server. When the server receives a TERMINATE CLUSTER request at /ksql/terminate endpoint, it writes a TERMINATE CLUSTER command into the command topic. Note that TERMINATE CLUSTER request can only be sent via the /ksql/terminate endpoint and you cannot send it via the CLI. When the server reads the TERMINATE CLUSTER command, it takes the following steps:

-Sets the KSQL engine mode to NOT ACCEPTING NEW STATEMENTS so no new statements will be passed to the engine for execution. -Terminates all persistent and transient queries in the engine and performs the required clean up for each query. -Deletes the command topic for the cluster.

Example request

POST /ksql/terminate HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{}

You can customize the clean up process if you want to delete some or all of the Kafka topics too:

Provide a List of Kafka Topics to Delete You can provide a list of kafka topic names or regular expressions for kafka topic names along with your TERMINATE CLUSTER request. The KSQL server will delete all topics with names that are in the list or that match any of the regular expressions in the list. Only the topics that were generated by KSQL queries will be considered for deletion. Topics that were not generated by KSQL queries will not be deleted even if they match the provided list. The following example will delete topic FOO along with all topics with prefix bar.

POST /ksql/terminate HTTP/1.1
Accept: application/vnd.ksql.v1+json
Content-Type: application/vnd.ksql.v1+json

{
  "deleteTopicList": ["FOO", "bar.*"]
}