Contents
The TIBCO StreamBase® Output Adapter for Apache Kafka Producer allows StreamBase applications to connect to an Apache Kafka Broker and to send messages to the broker on specific topics.
Messages can be sent in various formats such as tuple, string, blob, or a custom format that you provide. The input tuple determines what the data type is for the message and key. The key for each message can also be specified as any format, just like the data.
You can provide a partition number to indicate into which partition this message should be stored, or you can leave it blank for the Kafka broker to manage partition placement. When a message is delivered to the broker, an output tuple is sent from the producer with the actual partition and offset for that message. This occurs when you enable the Enable Pass Through Fields option. The Kafka broker guarantees the message order.
The Apache Kafka adapter suite is implemented against the version of the Kafka libraries listed on the Supported Configurations page.
This section describes the properties you can set for this adapter, using the various tabs of the Properties view in StreamBase Studio.
Name: Use this required field to specify or change the name of this instance of this component, which must be unique in the current EventFlow module. The name must contain only alphabetic characters, numbers, and underscores, and no hyphens or other special characters. The first character must be alphabetic or an underscore.
Adapter: A read-only field that shows the formal name of the adapter.
Start with application: If this field is set to Yes (default) or to a module parameter that evaluates to true
, this instance of this adapter starts as part of the JVM engine that runs this EventFlow module. If this field is set to
No or to a module parameter that evaluates to false
, the adapter instance is loaded with the engine, but does not start until you send an sbadmin resume command, or until you start the component with StreamBase Manager.
Enable Error Output Port: Select this check box to add an Error Port to this component. In the EventFlow canvas, the Error Port shows as a red output port, always the last port for the component. See Using Error Ports to learn about Error Ports.
Description: Optionally enter text to briefly describe the component's purpose and function. In the EventFlow canvas, you can see the description by pressing Ctrl while the component's tooltip is displayed.
Property | Type | Description |
---|---|---|
Brokers | string | A comma-separated list of address:port[config=value|config=value] Kafka brokers. Default value is localhost:9092 This value can also be a Azure event hubs connection string. The config=value section of the brokers list allows you to specify
advance configuration directly in the broker list. For example if you required a security.protocol and security.mechanism
you can specify a broker list like: test.com:9093[security.protocol=SASL_SSL,sasl.mechanism=PLAIN],test2.com:9093 |
Connect On Startup | check box | When enabled, the adapter attempts to connect to a Kafka broker on startup. If not enabled, you must manually send a connect command; if not sent, any published tuples fail until connected. |
Enable Command Port | check box | When enabled, the command port allows input tuples to control the adapter. For more information see Command Port. |
Enable Status Port | check box | When enabled, the adapter sends informational data on the status port about various adapter states. For more information see Status Port. |
Enable Metrics Port | check box | When enabled, the adapter allows output for the Kafka connection metrics when a metrics command is input. For more information see Metrics Port. |
Enable Passthrough Fields | check box | When enabled, the adapter outputs the data tuple when the message is successfully published to the Kafka broker. |
Close Connection On Error | check box | When enabled, the producer connection closes when any exception occurs. |
Synchronous Send | check box | When enabled, the producer blocks until sending completes. |
Log Level | INFO | Controls the level of verbosity the adapter uses to issue informational traces to the console. This setting is independent of the containing application's overall log level. Available values, in increasing order of verbosity, are: OFF, ERROR, WARN, INFO, DEBUG, TRACE. |
Property | Type | Description |
---|---|---|
Serializer Type | Enum | This value specifies how the input message field will be serialized. If | is selected and a serializer class is supplied, it is used. If no serializer class is supplied then the input field type determines how the message is serialized (tuple, blob, string). If is selected, the input message field must be a tuple, which is serialized to an Avro record.
Serializer Class | string |
Specifies the class that implements the There are a few values that are passed to the
|
Serializer Properties | string key:value |
A set of key value properties that are passed into the custom serializer by the configuration method. |
Key Serializer Type | Enum | This value specifies how the input key field will be serialized. If custom and a user serialize class is supplied, it is used. If no serialize class is supplied then input field type determines how the key is serialized (tuple, blob, string). If Avro is selected the input key field must be a tuple and will serialize to an Avro record. |
Key Serializer Class | string |
Specifies the class that implements the The values passed the |
Key Serializer Properties | string key:value |
A set of key value properties that are passed into the custom key serializer by the configuration method. |
Compression Type | drop-down list | The compression type for all data generated by this producer instance. Compression consists of full batches of data, so the efficacy of batching also affects the compression ratio (more batching means better compression). Available values: None, GZip, Snappy. |
Advanced Config | String Array | This key value set of strings allows you to configure any Kafka Producer Property. |
Use the Concurrency tab to specify parallel regions for this instance of this component, or multiplicity options, or both. The Concurrency tab settings are described in Concurrency Options, and dispatch styles are described in Dispatch Styles.
Caution
Concurrency settings are not suitable for every application, and using these settings requires a thorough analysis of your application. For details, see Execution Order and Concurrency, which includes important guidelines for using the concurrency options.
You can optionally enable the command input port for this adapter instance by means of the Enable Command Port property in the Adapter Properties page of the Properties view. Use the command port to send action commands to the adapter.
-
command, string, the command to send to the adapter.
The valid values are: connect, disconnect, metrics, updateBrokers
-
connect — Connect to the specified Kafka broker.
-
disconnect — Disconnect from the broker.
-
metrics — If this command port is enabled, the adapter produces a list of Kafka metrics associated with the current connection. Most values are null until at least one message is sent.
-
updateBrokers — Update the list of brokers to which this adapter will connect. This command updates only the broker list. You must disconnect and reconnect for the new broker list to take effect.
-
-
brokers (optional), list[tuple[string,integer]] — a list of brokers to use with the updateBrokers command.
The data port is the default input port for the Kafka Producer adapter, and is always enabled. Use the data port to send message data to the connected Apache Broker.
The default schema for the data input port is:
-
topic, string. The topic for this message.
-
message, string, tuple, blob, or custom. The message content.
-
key (optional), string, tuple, blob, or custom. The key for this message.
-
partition (optional), integer. The partition where this message is stored. If not supplied, the broker determines the partition.
You can optionally enable the status output port for this adapter instance by means of the Enable Status Port property in the Adapter Properties page of the Properties view. Use the status port to retrieve status output messages from the adapter.
-
type, string. The type of status information emitted. Status types are:
-
Error — This message relates to an error that occurred.
-
Warn — This message relates to a warning that the user should be aware of.
-
Info — This message relates to extra status information.
-
-
action, string. Valid values are:
-
Send — This message relates to an attempt to send a message to the broker.
-
Connected — The adapter has successfully connected to the broker.
-
Connecting — The adapter is currently trying to connect to the broker.
-
Disconnecting — The adapter is currently disconnecting from the broker.
-
Update Brokers — The broker list is being updated.
-
Command — This message relates to an input command sent on the command port.
-
-
object, string. This value may be null. If not null, it contains a value relevant to the status message.
-
message, string. This is a formatted human readable message that explains the status message.
-
time, timestamp. The time the status message occurred.
You can optionally enable the metrics output port for this adapter instance by means of the Enable Metrics Port property in the Adapter Properties page of the Properties view. Use the metrics port to retrieve metrics output messages sent from the adapter in response to a metrics command sent to its command port. A set of perhaps 40 or 50 metrics messages are sent in response to this command. Most metrics values are null until at least one message has been sent through the currently connected broker.
When you select Custom for either the message or key serialization, you must provide a class that implements the org.apache.kafka.common.serialization.Serializer
interface. For example:
package com.streambase.sb.adapter.kafka; import java.util.EnumSet; import java.util.Map; import org.slf4j.Logger; import com.streambase.sb.Schema; import com.streambase.sb.Tuple; import com.streambase.sb.TupleJSONUtil; import com.streambase.sb.TupleJSONUtil.Options; public class DemoSerializer implements org.apache.kafka.common .serialization.Serializer<Tuple> { private static final String CONFIG_SCHEMA = "schema"; private static final String CONFIG_LOGGER = "logger"; private Schema schema; private Logger logger; @Override public void close() { } @Override public void configure(Map<String, ?> configs, boolean isKey) { if (configs.containsKey(CONFIG_SCHEMA)) { schema = (Schema)configs.get(CONFIG_SCHEMA); } if (configs.containsKey(CONFIG_LOGGER)) { logger = (Logger) configs.get(CONFIG_LOGGER); } } @Override public byte[] serialize(String topic, Tuple tuple) { EnumSet<Options> options = EnumSet.of(Options.INCLUDE_NULL_FIELDS, Options .INCLUDE_NULL_FIELDS, Options.PREFER_MAP); String jsonTuple = TupleJSONUtil.toJSONString(tuple, options, ""); return jsonTuple.getBytes(); } }