in_tcp: add source address even when format is none

[jadeKim1993]

Modify to add source_address even when input format is none instead of JSON

---

Enter [N/A] in the box, if an item is not applicable to your change.

Testing
Before we can approve your change; please submit the following in a comment:

• Example configuration file for the change

[INPUT]
    Name        tcp
    Listen      0.0.0.0
    Port        55555
    Format      none
    Source_Address_Key source_address

[OUTPUT]
    Name        stdout
    Match       *

• Debug log output from testing the change

[2026/01/08 14:42:27] [ info] Configuration:
[2026/01/08 14:42:27] [ info]  flush time     | 1.000000 seconds
[2026/01/08 14:42:27] [ info]  grace          | 5 seconds
[2026/01/08 14:42:27] [ info]  daemon         | 0
[2026/01/08 14:42:27] [ info] ___________
[2026/01/08 14:42:27] [ info]  inputs:
[2026/01/08 14:42:27] [ info]      tcp
[2026/01/08 14:42:27] [ info] ___________
[2026/01/08 14:42:27] [ info]  filters:
[2026/01/08 14:42:27] [ info] ___________
[2026/01/08 14:42:27] [ info]  outputs:
[2026/01/08 14:42:27] [ info]      stdout.0
[2026/01/08 14:42:27] [ info] ___________
[2026/01/08 14:42:27] [ info]  collectors:
[2026/01/08 14:42:27] [ info] [fluent bit] version=4.0.2, commit=3a17947d29, pid=63465
[2026/01/08 14:42:27] [debug] [engine] coroutine stack size: 24576 bytes (24.0K)
[2026/01/08 14:42:27] [ info] [storage] ver=1.5.3, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2026/01/08 14:42:27] [ info] [simd    ] disabled
[2026/01/08 14:42:27] [ info] [cmetrics] version=1.0.0
[2026/01/08 14:42:27] [ info] [ctraces ] version=0.6.4
[2026/01/08 14:42:27] [ info] [input:tcp:tcp.0] initializing
[2026/01/08 14:42:27] [ info] [input:tcp:tcp.0] storage_strategy='memory' (memory only)
[2026/01/08 14:42:27] [debug] [tcp:tcp.0] created event channels: read=25 write=26
[2026/01/08 14:42:27] [debug] [downstream] listening on 0.0.0.0:55555
[2026/01/08 14:42:27] [debug] [stdout:stdout.0] created event channels: read=28 write=29
[2026/01/08 14:42:27] [ info] [sp] stream processor started
[2026/01/08 14:42:27] [ info] [output:stdout:stdout.0] worker #0 started
[2026/01/08 14:42:31] [debug] [task] created task=0x7fde14029420 id=0 OK
[2026/01/08 14:42:31] [debug] [output:stdout:stdout.0] task_id=0 assigned to thread #0
[0] tcp.0: [[1767850950.628888995, {}], {"log"=>"Hello TCP", "source_address"=>"tcp://127.0.0.1:60994"}]
[2026/01/08 14:42:31] [debug] [out flush] cb_destroy coro_id=0
[2026/01/08 14:42:31] [debug] [task] destroy task=0x7fde14029420 (task_id=0)

• Attached Valgrind output that shows no leaks or memory corruption was found

==63490== Memcheck, a memory error detector
==63490== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==63490== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright info
==63490== Command: ./fluent-bit -c test.conf
==63490==
Fluent Bit v4.0.2
* Copyright (C) 2015-2025 The Fluent Bit Authors
* Fluent Bit is a CNCF sub-project under the umbrella of Fluentd
* https://fluentbit.io

______ _                  _    ______ _ _             ___  _____
|  ___| |                | |   | ___ (_) |           /   ||  _  |
| |_  | |_   _  ___ _ __ | |_  | |_/ /_| |_  __   __/ /| || |/' |
|  _| | | | | |/ _ \ '_ \| __| | ___ \ | __| \ \ / / /_| ||  /| |
| |   | | |_| |  __/ | | | |_  | |_/ / | |_   \ V /\___  |\ |_/ /
\_|   |_|\__,_|\___|_| |_|\__| \____/|_|\__|   \_/     |_(_)___/


[2026/01/08 14:54:08] [ info] Configuration:
[2026/01/08 14:54:08] [ info]  flush time     | 1.000000 seconds
[2026/01/08 14:54:08] [ info]  grace          | 5 seconds
[2026/01/08 14:54:08] [ info]  daemon         | 0
[2026/01/08 14:54:08] [ info] ___________
[2026/01/08 14:54:08] [ info]  inputs:
[2026/01/08 14:54:08] [ info]      tcp
[2026/01/08 14:54:08] [ info] ___________
[2026/01/08 14:54:08] [ info]  filters:
[2026/01/08 14:54:08] [ info] ___________
[2026/01/08 14:54:08] [ info]  outputs:
[2026/01/08 14:54:08] [ info]      stdout.0
[2026/01/08 14:54:08] [ info] ___________
[2026/01/08 14:54:08] [ info]  collectors:
[2026/01/08 14:54:08] [ info] [fluent bit] version=4.0.2, commit=3a17947d29, pid=63490
[2026/01/08 14:54:08] [debug] [engine] coroutine stack size: 24576 bytes (24.0K)
[2026/01/08 14:54:08] [ info] [storage] ver=1.5.3, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2026/01/08 14:54:08] [ info] [simd    ] disabled
[2026/01/08 14:54:08] [ info] [cmetrics] version=1.0.0
[2026/01/08 14:54:08] [ info] [output:stdout:stdout.0] worker #0 started
[2026/01/08 14:54:08] [ info] [ctraces ] version=0.6.4
[2026/01/08 14:54:08] [ info] [input:tcp:tcp.0] initializing
[2026/01/08 14:54:08] [ info] [input:tcp:tcp.0] storage_strategy='memory' (memory only)
[2026/01/08 14:54:08] [debug] [tcp:tcp.0] created event channels: read=25 write=26
[2026/01/08 14:54:08] [debug] [downstream] listening on 0.0.0.0:55555
[2026/01/08 14:54:08] [debug] [stdout:stdout.0] created event channels: read=28 write=29
[2026/01/08 14:54:08] [ info] [sp] stream processor started
[0] tcp.0: [[1767851652.262490636, {}], {"log"=>"Hello TCP", "source_address"=>"tcp://127.0.0.1:59940"}]
[2026/01/08 14:54:12] [debug] [task] created task=0x54af260 id=0 OK
[2026/01/08 14:54:12] [debug] [output:stdout:stdout.0] task_id=0 assigned to thread #0
[2026/01/08 14:54:12] [debug] [out flush] cb_destroy coro_id=0
[2026/01/08 14:54:12] [debug] [task] destroy task=0x54af260 (task_id=0)
^C[2026/01/08 14:54:15] [engine] caught signal (SIGINT)
[2026/01/08 14:54:15] [ warn] [engine] service will shutdown in max 5 seconds
[2026/01/08 14:54:15] [ info] [engine] service has stopped (0 pending tasks)
[2026/01/08 14:54:15] [ info] [output:stdout:stdout.0] thread worker #0 stopping...
[2026/01/08 14:54:15] [ info] [output:stdout:stdout.0] thread worker #0 stopped
==63490==
==63490== HEAP SUMMARY:
==63490==     in use at exit: 0 bytes in 0 blocks
==63490==   total heap usage: 1,701 allocs, 1,701 frees, 900,312 bytes allocated
==63490==
==63490== All heap blocks were freed -- no leaks are possible
==63490==
==63490== For lists of detected and suppressed errors, rerun with: -s
==63490== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)

If this is a change to packaging of containers or native binaries then please confirm it works for all targets.

• [N/A] Run local packaging test showing all targets (including any new ones) build.
• [N/A] Set ok-package-test label to test for all targets (requires maintainer to do).

Documentation

• Documentation required for this feature

Please refer to the official documentation at the following link
(https://github.com/fluent/fluent-bit-docs/blob/master/pipeline/inputs/tcp.md)

Backporting

• [N/A] Backport to latest stable release.

---

Fluent Bit is licensed under Apache 2.0, by submitting this pull request I understand that this code will be released under the terms of that license.

Summary by CodeRabbit

• New Features

  • TCP input plugin now supports capturing the source address from incoming client connections when using format=none payload format. Configure this feature using the source_address_key option to store the source address in a custom field.

• Tests

  • Added test coverage verifying source address capture functionality works correctly with format=none.

[github-actions]

This PR is stale because it has been open 45 days with no activity. Remove stale label or comment or this will be closed in 10 days.

[jadeKim1993]

I have updated the PR description. Please review it at your convenience.
thank you.

[leonardo-albertovich]

@edsiper this PR is pretty simple and it's covered by a test case, it's ready for the next merge window.

[leonardo-albertovich]

@edsiper could you please make the test cases run so the code doesn't go stale for the next release cycle? It's a really simple change that shouldn't cause any problems.

[edsiper]

not sure why CI is now not running, will check.

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Swish!

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR extends the TCP input plugin to optionally include the remote source address in output records when format=none is configured. It introduces conditional logic in parse_payload_none() to append a source address field to the output record and includes a test case validating this behavior.

Changes

Cohort / File(s)	Summary
TCP Payload Parsing plugins/in_tcp/tcp_conn.c	Modified parse_payload_none() to conditionally retrieve and append source address information to the output record when ctx->source_address_key is configured.
Runtime Tests tests/runtime/in_tcp.c	Added new test flb_test_format_none_with_source_address() that validates source address field inclusion in TCP messages with format=none configuration.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

• #11531 – Also modifies the parse_payload_none() function in the same file with type/bounds adjustments.

Suggested reviewers

• edsiper

Poem

🐰 A whisker-twitching tcp tale,
Source addresses now set sail,
In none-formatted streams so bright,
Connection origins, shining light! ✨

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: adding source address support to the TCP input plugin when format is set to 'none', which is the primary focus of both the code and test changes.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

🧹 Nitpick comments (1)

tests/runtime/in_tcp.c (1)

687-687: Make the assertion less order-dependent to avoid brittle test failures.

Line 687 expects "log" immediately followed by "source_host". Prefer checking only "source_host":"tcp:// here (the log field is already covered by flb_test_format_none) or validating both fields independently.

♻️ Minimal diff to reduce ordering sensitivity

-    cb_data.data = "\"log\":\"message\",\"source_host\":\"tcp://";
+    cb_data.data = "\"source_host\":\"tcp://";

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@tests/runtime/in_tcp.c` at line 687, The test currently sets cb_data.data to
a string that requires the "log" field to appear immediately before
"source_host" which makes the assertion order-dependent; change the test so
cb_data.data only asserts the presence of the "source_host":"tcp://" fragment
(or validate "log" and "source_host" independently) instead of the exact
ordering—update the assignment to cb_data.data in tests/runtime/in_tcp.c and
adjust any assertions that rely on exact field order (note: flb_test_format_none
already covers "log", so keep that check separate).

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Nitpick comments:
In `@tests/runtime/in_tcp.c`:
- Line 687: The test currently sets cb_data.data to a string that requires the
"log" field to appear immediately before "source_host" which makes the assertion
order-dependent; change the test so cb_data.data only asserts the presence of
the "source_host":"tcp://" fragment (or validate "log" and "source_host"
independently) instead of the exact ordering—update the assignment to
cb_data.data in tests/runtime/in_tcp.c and adjust any assertions that rely on
exact field order (note: flb_test_format_none already covers "log", so keep that
check separate).

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: f192e424-ced0-4ae8-8ab2-c0ffa81c1781

📥 Commits

Reviewing files that changed from the base of the PR and between 680f14c and f05593e.

📒 Files selected for processing (2)

• plugins/in_tcp/tcp_conn.c
• tests/runtime/in_tcp.c