> For the complete documentation index, see [llms.txt](https://docs.fluentbit.io/manual/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.fluentbit.io/manual/data-pipeline/inputs/tcp.md). # TCP {% hint style="info" %} **Supported event types:** `logs` {% endhint %} The *TCP* input plugin lets you retrieve structured JSON or raw messages over a TCP network interface (TCP port). ## Configuration parameters The plugin supports the following configuration parameters: | Key | Description | Default | | -------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------- | | `buffer_size` | Specify the maximum buffer size in KB to receive a JSON message. If not set, the default is the value of `chunk_size`. | `chunk_size` | | `chunk_size` | The default buffer to store the incoming JSON messages. It doesn't allocate the maximum memory allowed; instead it allocates memory when required. The rounds of allocations are set by `chunk_size`. If not set, `chunk_size` is equal to 32 (32 KB). | `32` | | `format` | Specify the expected payload format. Supported values: `json` and `none`. When set to `json` it expects JSON objects or arrays and, in v5.1.3 or later, skips other values. When set to `none`, every record splits using the defined `separator`. | `json` | | `listen` | Listener network interface. | `0.0.0.0` | | `parser` | Optional [parser](/manual/data-pipeline/parsers.md) for line-delimited records. Requires `format` to be `none`; if `format` is set to anything else, it's automatically switched to `none` and a warning is logged. | *none* | | `port` | TCP port to listen for connections. | `5170` | | `separator` | When `format` is set to `none`, Fluent Bit needs a separator string to split the records. | `LF` or `0x10` (break line) | | `source_address_key` | Specify the key to inject the source address. | *none* | | `threaded` | Indicates whether to run this input in its own [thread](/manual/administration/multithreading.md#inputs). | `false` | | `workers` | The number of listener workers that accept and process incoming connections on the port. | `1` | In v5.1.3 or later, when `format` is `json` and a payload can't be encoded or ingested, the plugin logs the warning `skipping unprocessable JSON payload` and consumes the affected bytes instead of processing them again. Any records from that payload that were already routed before the failure remain delivered; consuming the bytes prevents them from being duplicated on a later read. ## Get started To receive JSON messages over TCP, you can run the plugin from the command line or through the configuration file. ### Command line From the command line you can let Fluent Bit listen for JSON messages with the following options: ```shell fluent-bit -i tcp -o stdout ``` By default, the service will listen an all interfaces (`0.0.0.0`) through TCP port `5170`. Optionally you can change this directly: ```shell fluent-bit -i tcp://192.168.3.2:9090 -o stdout ``` In the example the JSON messages will only arrive through the network interface at `192.168.3.2` address and TCP Port `9090`. ### Configuration file In your main configuration file append the following sections: {% tabs %} {% tab title="fluent-bit.yaml" %} ```yaml pipeline: inputs: - name: tcp listen: 0.0.0.0 port: 5170 chunk_size: 32 buffer_size: 64 format: json outputs: - name: stdout match: '*' ``` {% endtab %} {% tab title="fluent-bit.conf" %} ``` [INPUT] Name tcp Listen 0.0.0.0 Port 5170 Chunk_Size 32 Buffer_Size 64 Format json [OUTPUT] Name stdout Match * ``` {% endtab %} {% endtabs %} ## Test the configuration When Fluent Bit is running, you can send some messages using `netcat`: ```shell echo '{"key 1": 123456789, "key 2": "abcdefg"}' | nc 127.0.0.1 5170 ``` Run Fluent Bit: ```shell fluent-bit -i tcp -o stdout -f 1 ``` You should see the following output: ``` ... [0] tcp.0: [1570115975.581246030, {"key 1"=>123456789, "key 2"=>"abcdefg"}] ... ``` ## Performance considerations When receiving payloads in JSON format, there are high performance penalties. Parsing JSON is a very expensive task so you could expect your CPU usage increase under high load environments. To get faster data ingestion, consider to use the option `format none` to avoid JSON parsing if not needed. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.fluentbit.io/manual/data-pipeline/inputs/tcp.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.