Versions | v0.12 (td-agent2) | v0.10 (td-agent1)

Collect Log Files into Kinesis Stream in Real-Time

This article explains how to use Fluentd’s Amazon Kinesis Output plugin (out_kinesis) to aggregate semi-structured logs in real-time. Kinesis plugin is developed and published by Amazon Web Services officially.

Table of Contents


Fluentd is an advanced open-source log collector originally developed at Treasure Data, Inc. Because Fluentd can collect logs from various sources, Amazon Kinesis is one of the popular destinations for the output.

Amazon Kinesis is a platform for streaming data on AWS, offering powerful services to make it easy to load and analyze streaming data, and also providing the ability for you to build custom streaming data applications for specialized needs.

This article will show you how to use Fluentd to import Apache logs into Amazon Kinesis.


In this example, Fluentd does 3 things:

  1. It continuously “tails” the access log file.
  2. It parses the incoming log entries into meaningful fields (such as ip, path, etc.) and buffers them.
  3. It writes the buffered data to Amazon Kinesis periodically.


For simplicity, this article will describe how to set up an one-node configuration. Please install the following software on the same node.

  • Fluentd
  • Apache (with the Combined Log Format)

You can install Fluentd via major packaging systems.

Install Kinesis Plugin

Since Amazon Kinesis plugin is not bundled with td-agent package, plase install it manually.

$ sudo td-agent-gem install fluent-plugin-kinesis


Let’s start configuring Fluentd. If you used the deb/rpm package, Fluentd’s config file is located at /etc/td-agent/td-agent.conf. Otherwise, it is located at /etc/fluentd/fluentd.conf.

Tail Input

For the input source, we will set up Fluentd to track the recent Apache logs (typically found at /var/log/apache2/access_log) The Fluentd configuration file should look like this:

  type tail
  format apache2
  path /var/log/apache2/access_log
  pos_file /var/log/td-agent/apache2.access_log.pos
  tag kinesis.apache.access
Please make sure that your Apache outputs are in the default 'combined' format. `format apache2` cannot parse custom log formats. Please see the in_tail article for more information.

Let’s go through the configuration line by line.

  1. type tail: The tail Input plugin continuously tracks the log file. This handy plugin is included in Fluentd’s core.
  2. format apache2: Uses Fluentd’s built-in Apache log parser.
  3. path /var/log/apache2/access_log: The location of the Apache log. This may be different for your particular system.
  4. tag kinesis.apache.access: kinesis.apache.access is used as the tag to route the messages within Fluentd.

That’s it! You should now be able to output a JSON-formatted data stream for Fluentd to process.

Amazon Kinesis Output

The output destination will be Amazon Kinesis. The output configuration should look like this:

<match **>
  # plugin type
  type kinesis

  # your kinesis stream name
  stream_name <KINESIS_STREAM_NAME>

  # AWS credentials
  aws_key_id <AWS_KEY_ID>
  aws_sec_key <AWS_SECRET_KEY>

  # AWS region
  region us-east-1

  # Use random value for the partition key
  random_partition_key true

  # Frequency of ingestion
  flush_interval 5s

  # Parallelism of ingestion
  num_threads 16

The match section specifies the regexp used to look for matching tags. If a matching tag is found in a log, then the config inside <match>...</match> is used (i.e. the log is routed according to the config inside). In this example, the kinesis.apache.access tag (generated by tail) is always used.

The ** in match.** matches zero or more period-delimited tag parts (e.g. match/match.a/match.a.b).

flush_interval specifies how often the data is written to Kinesis.

random_partition_key true option will generate the partition key via UUID v3 (source). Kinesis Stream consists of shards, and the processing power of each shard is limited. This partition key will be used by Kinesis, to determine which shard wll be assigned to for the specific record.

For additional configuration parameters, please see the Kinesis Output plugin README.
For those who are interested in security, all communication between Fluentd and Amazon Kinesis are done via HTTPS. If you don't want to have AES keys in the configuration file, IAM Role based authentication is available too for EC2 nodes.


Please restart td-agent process first, to make the configuration change available.

$ sudo /etc/init.d/td-agent stop
$ sudo /etc/init.d/td-agent configtest
$ sudo /etc/init.d/td-agent start

To test the configuration, just have a couple of accesses to your Apache server. This example uses the ab (Apache Bench) program.

$ ab -n 100 -c 10 http://localhost/


Why we need Fluentd, while Kinesis also offers client libraries?

A lot of people use Fluentd + Kinesis, simply because they want to have more choice for inputs and outputs. For inputs, Fluentd has a lot more community contributed plugins and libraries. For outputs, you can send not only Kinesis, but multiple destinations like Amazon S3, local file storage, etc.


Fluentd + Amazon Kinesis makes real-time log collection simple, easy, and robust.

Learn More

Last updated: 2016-06-28 12:40:54 UTC

Versions | v0.12 (td-agent2) | v0.10 (td-agent1)

If this article is incorrect or outdated, or omits critical information, please let us know. Fluentd is a open source project under Cloud Native Computing Foundation (CNCF), originally invented by Treasure Data, Inc. All components are available under the Apache 2 License.

Interested in the Fluentd Newsletters?