forward

The in_forward Input plugin listens to a TCP socket to receive the event stream. It also listens to a UDP socket to receive heartbeat messages. See also the protocol section for implementation details.

This plugin is mainly used to receive event logs from other Fluentd instances, the fluent-cat command, or Fluentd client libraries. This is by far the most efficient way to retrieve the records.

If you want to receive events from raw TCP payload, use in_tcp plugin instead.

It is included in Fluentd's core.

Example Configuration

<source>
  @type forward
  port 24224
  bind 0.0.0.0
</source>

Refer to the Configuration File article for the basic structure and syntax of the configuration file.

Plugin Helpers

• server

Parameters

• Common Parameters

• Transport Section

@type

The value must be forward.

port

type	default	version
integer	24224	0.14.0

The port to listen to.

bind

type	default	version
string	0.0.0.0 (all addresses)	0.14.0

The bind address to listen to.

tag

type	default	version
string	nil	1.5.0

in_forward uses incoming event's tag by default (See Protocol Section). If the tag parameter is set, its value is used instead.

add_tag_prefix

type	default	version
string	nil	1.5.0

Adds the prefix to the incoming event's tag.

Here is an example:

<source>
  @type forward
  add_tag_prefix prod
</source>

With this configuration, the emitted tag is prod.INCOMING_TAG, e.g. prod.app.log.

linger_timeout

type	default	version
integer	0	0.14.0

The timeout used to set the linger option.

This parameter is deprecated since v1.14.6. Use <transport> directive instead.

resolve_hostname

type	default	version
bool	false	0.14.10

Tries to resolve hostname from IP addresses or not.

deny_keepalive

type	default	version
bool	false	0.14.5

The connections will be disconnected right after receiving a message, if true.

send_keepalive_packet

type	default	version
bool	false	1.4.2

Enables the TCP keepalive for sockets. See socket article for more details.

chunk_size_limit

type	default	version
size	nil (no limit)	0.14.0

The size limit of the received chunk. If the chunk size is larger than this value, the received chunk is dropped.

chunk_size_warn_limit

type	default	version
size	nil (no warning)	0.14.0

The warning size limit of the received chunk. If the chunk size is larger than this value, a warning message will be sent.

skip_invalid_event

type	default	version
bool	true	0.14.0

Skips the invalid incoming event.

This option is useful for forwarder, not aggregator. Since v1.19.0, the default value is changed to true.

source_address_key

type	default	version
string	nil (no adding address)	0.14.11

The field name of the client's source address. If set, the client's address will be set to its key.

source_hostname_key

type	default	version
string	nil (no adding hostname)	0.14.4

The field name of the client's hostname. If set, the client's hostname will be set to its key.

This iterates incoming events. So, if you send larger chunks to in_forward, it needs additional processing time.

<transport> Section

This section is for setting TLS transport or some general transport configurations.

General configuration

linger_timeout

type	default	available transport type	version
integer	0	tcp, tls	1.14.6

The timeout (seconds) to set SO_LINGER.

The default value 0 is to send RST rather than FIN to avoid lots of connections sitting in TIME_WAIT on closing on non-Windows.

You can set positive value to send FIN on closing on non-Windows.

On Windows, Fluentd sends FIN without depending on this setting.

<transport tcp>
  linger_timeout 1
</transport>

TLS configuration

<transport tls>
  cert_path /path/to/fluentd.crt
  # other parameters
</transport>

See How to Enable TLS Encryption section for how to use and see Configuration Example for all supported parameters.

Without <transport tls>, in_forward uses raw TCP.

<security> Section

required	multi	version
false	false	0.14.5

This section contains parameters related to authentication:

• self_hostname

• shared_key

• user_auth

• allow_anonymous_source

self_hostname

type	default	version
string	required parameter	0.14.5

The hostname.

shared_key

type	default	version
string	required parameter	0.14.5

The shared key for authentication.

user_auth

type	default	version
bool	false	0.14.5

If true, user-based authentication is used.

allow_anonymous_source

type	default	version
bool	true	0.14.5

Allows the anonymous source. <client> sections are required, if disabled.

<user> section

required	multi	version
false	true	0.14.5

This section contains user-based authentication:

• username

• password

This section can be used in <security>.

username

type	default	version
string	required parameter	0.14.5

The username for authentication.

password

type	default	version
string	required parameter	0.14.5

The password for authentication.

<client> section

required	multi	version
false	true	0.14.5

This section contains client IP/Network authentication and shared key per host:

• host

• network

• shared_key

• users

This section can be used in <security>

host

type	default	version
string	nil	0.14.5

The IP address or hostname of the client.

This is exclusive with network.

network

type	default	version
string	nil	0.14.5

The network address specification.

This is exclusive with host.

shared_key

type	default	version
string	nil	0.14.5

The shared key per client.

users

type	default	version
array	[]	0.14.5

The array of usernames.

Protocol

This plugin accepts both JSON or MessagePack messages and automatically detects which one is used. Internally, Fluentd uses MessagePack as it is more efficient than JSON.

The time value is an EventTime or a platform-specific integer and is based on the output of Ruby's Time.now.to_i function. On Linux, BSD, and Mac systems, this is the number of seconds since 1970.

Multiple messages may be sent on the same connection:

stream:
  message...

message:
  [tag, time, record]
  or
  [tag, [[time,record], [time,record], ...]]

example:
  ["myapp.access", 1308466941, {"a":1}]["myapp.messages", 1308466942, {"b":2}]
  ["myapp.access", [[1308466941, {"a":1}], [1308466942, {"b":2}]]]

For more details, see Fluentd Forward Protocol Specification (v1).

Tips and Tricks

How to Enable TLS Encryption

Since v0.14.12, Fluentd includes a built-in TLS support. Here we present a quick tutorial for setting up TLS encryption:

First, generate a self-signed certificate using the following command:

$ openssl req -new -x509 -sha256 -days 1095 -newkey rsa:2048 \
              -keyout fluentd.key -out fluentd.crt
# Note that during the generation, you will be asked for:
#  - a password (to encrypt the private key), and
#  - subject information (to be included in the certificate)

Move the generated certificate and private key to a safer place. For example:

# Move files into /etc/td-agent
$ sudo mkdir -p /etc/td-agent/certs
$ sudo mv fluentd.key fluentd.crt /etc/td-agent/certs

# Set strict permissions
$ sudo chown td-agent:td-agent -R /etc/td-agent/certs
$ sudo chmod 700 /etc/td-agent/certs/
$ sudo chmod 400 /etc/td-agent/certs/fluentd.key

Then, add the following settings to td-agent.conf and restart the service:

<source>
  @type forward
  <transport tls>
    cert_path /etc/td-agent/certs/fluentd.crt
    private_key_path /etc/td-agent/certs/fluentd.key
    private_key_passphrase YOUR_PASSPHRASE
  </transport>
</source>
<match debug.**>
  @type stdout
</match>

To test your encryption settings, execute the following command in your terminal. If the encryption is working properly, you should see a line containing {"foo":"bar"} in the log file:

$ echo -e '\x93\xa9debug.tls\xceZr\xbc1\x81\xa3foo\xa3bar' | \
  openssl s_client -connect localhost:24224

If you can confirm TLS encryption has been set up correctly, please proceed to the configuration of the out_forward server.

How to Enable TLS Mutual Authentication

Since v1.1.1, Fluentd supports TLS mutual authentication (i.e. client certificate auth). If you want to use this feature, please set the client_cert_auth and ca_path options like this:

<source>
  @type forward
  <transport tls>
    ...
    client_cert_auth true
    ca_path /path/to/ca/cert
  </transport>
</source>

When this feature is enabled, Fluentd will check all the incoming requests for a client certificate signed by the trusted CA. Requests with an invalid client certificate will fail.

To check if mutual authentication is working properly, issue the following command:

$ openssl s_client -connect localhost:24224 \
  -key path/to/client.key \
  -cert path/to/client.crt \
  -CAfile path/to/ca.crt

If the connection gets established successfully, your setup is working fine.

+For fluentd and fluent-bit combination, see Banzai Cloud article: Secure logging on Kubernetes with Fluentd and Fluent Bit.

How to Enable Password Authentication

Fluentd is equipped with a password-based authentication mechanism, which allows you to verify the identity of each client using a shared secret key.

To enable this feature, you need to add a <security> section to your configuration file like this:

<source>
  @type forward
  <security>
    self_hostname YOUR_SERVER_NAME
    shared_key PASSWORD
  </security>
</source>

Once the setup is complete, you have to configure your clients accordingly. For example, if you have an out_forward instance running on another server, configure it by following these instructions.

Multi-process Environment

If you use this plugin under the multi-process environment, the port will be shared.

<system>
  workers 3
</system>

<source>
  @type forward
  port 24224
</source>

With this configuration, the three (3) workers share the port 24224. No need for an additional port. Incoming data will be routed to the workers automatically.

FAQ

How to parse incoming events?

in_forward does not provide parsing mechanism unlike in_tail or in_tcp because in_forward is mainly for efficient log transfer. If you want to parse an incoming event, use parser filter in your pipeline.

See Docker Logging driver use case.

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