Supported Sources, Sinks, and Channels

The following tables list the only currently-supported sources, sinks, and channels. For more information, including information on developing custom components, see the documents listed under Viewing the Flume Documentation.

Sources

Type

Description

Implementation Class

avro

Avro Netty RPC event source. Listens on Avro port and receives events from external Avro streams.

AvroSource

netcat

Netcat style TCP event source. Listens on a given port and turns each line of text into an event.

NetcatSource

seq

Monotonically incrementing sequence generator event source

SequenceGeneratorSource

exec

Execute a long-lived Unix process and read from stdout.

ExecSource

syslogtcp

Reads syslog data and generates flume events. Creates a new event for a string of characters separated by carriage return ( \n ).

SyslogTcpSource

syslogudp

Reads syslog data and generates flume events. Treats an entire message as a single event.

SyslogUDPSource

org.apache.flume.source.avroLegacy. AvroLegacySource

Allows the Flume 1.x agent to receive events from Flume 0.9.4 agents over avro rpc.

AvroLegacySource

org.apache.flume.source.thriftLegacy. ThriftLegacySource

Allows the Flume 1.x agent to receive events from Flume 0.9.4 agents over thrift rpc.

ThriftLegacySource

org.apache.flume.source.StressSource

Mainly for testing purposes. Not meant for production use. Serves as a continuous source of events where each event has the same payload.

StressSource

org.apache.flume.source.scribe. ScribeSource

Scribe event source. Listens on Scribe port and receives events from Scribe.

ScribeSource

multiport_syslogtcp

Multi-port capable version of the SyslogTcpSource.

MultiportSyslogTCPSource

spooldir

Ingests data by placing files to be ingested into a "spooling" directory on disk.

SpoolDirectorySource

http

Accepts Flume events by HTTP POST and GET. GET should be used for experimentation only.

HTTPSource

org.apache.flume.source.jms.JMSSource

Reads messages from a JMS destination such as a queue or topic.

JMSSource

org.apache.flume.agent.embedded. EmbeddedSource

Used only by the Flume embedded agent. See Flume Developer Guide for more details.

EmbeddedSource

org.apache.flume.source.kafka.KafkaSource Streams data from Kafka to Hadoop or from any Flume source to Kafka. KafkaSource
org.apache.flume.source.taildir.TaildirSource

Watches specified files, and tails them in near real-time when it detects appends to these files.

  • This source is reliable and does not miss data, even when the tailing files rotate.
  • It periodically writes the last read position of each file in a position file using the JSON format.
  • If Flume is stopped or down for some reason, it can restart tailing from the position written in the existing position file.
  • It can add event headers to each tailing file group.
TaildirSource

Sinks

Type

Description

Implementation Class

logger

Log events at INFO level using configured logging subsystem (log4j by default)

LoggerSink

avro

Sink that invokes a pre-defined Avro protocol method for all events it receives (when paired with an avro source, forms tiered collection)

AvroSink

hdfs

Writes all events received to HDFS (with support for rolling, bucketing, HDFS-200 append, and more)

HDFSEventSink

file_roll

Writes all events received to one or more files.

RollingFileSink

org.apache.flume.hbase.HBaseSink

A simple sink that reads events from a channel and writes them synchronously to HBase. The AsyncHBaseSink is recommended. See Importing Data Into HBase.

HBaseSink

org.apache.flume.sink.hbase.AsyncHBaseSink

A simple sink that reads events from a channel and writes them asynchronously to HBase. This is the recommended HBase sink, but note that it does not support Kerberos. See Importing Data Into HBase.

AsyncHBaseSink

org.apache.flume.sink.solr.morphline.MorphlineSolrSink

Extracts and transforms data from Flume events, and loads it into Apache Solr servers. See the section on MorphlineSolrSink in the Flume User Guide listed under Viewing the Flume Documentation.

MorphlineSolrSink

org.apache.flume.sink.kafka.KafkaSink Used to send data to Kafka from a Flume source. You can use the Kafka sink in addition to Flume sinks such as HBase or HDFS. KafkaSink

Channels

Type

Description

Implementation Class

memory

In-memory, fast, non-durable event transport

MemoryChannel

jdbc

JDBC-based, durable event transport (Derby-based)

JDBCChannel

file

File-based, durable event transport

FileChannel

org.apache.flume.channel.kafka.KafkaChannel Use the Kafka channel:
  • To write to Hadoop directly from Kafka without using a source.
  • To write to Kafka directly from Flume sources without additional buffering.
  • As a reliable and highly available channel for any source/sink combination.
KafkaChannel

Providing for Disk Space Usage

It's important to provide plenty of disk space for any Flume File Channel. The largest consumers of disk space in the File Channel are the data logs. You can configure the File Channel to write these logs to multiple data directories. The following space will be consumed by default in each data directory:

  • Current log file (up to 2 GB)
  • Last log file (up to 2 GB)
  • Pending delete log file (up to 2 GB)

Events in the queue could cause many more log files to be written, each of them up 2 GB in size by default.

You can configure both the maximum log file size (MaxFileSize) and the directories the logs will be written to (DataDirs) when you configure the File Channel; see the File Channel section of the Flume User Guide for details.