Skip to main content
Version: 1.11.x (Next)

Logs

Langflow produces logs for individual flows and the Langflow application itself using the structlog library for logging.

By default, OSS Langflow writes logs to stdout. To write logs to a file, set LANGFLOW_LOG_FILE to the desired path.

Log storage​

The LANGFLOW_CONFIG_DIR environment variable controls where Langflow stores its configuration and data files. The default location depends on your operating system:

  • Langflow Desktop:

    • macOS: /Users/<username>/Library/Logs/com.LangflowDesktop
    • Windows: C:\Users\<username>\AppData\Local\com.LangflowDesktop\logs

  • OSS Langflow (all installation methods, including uv pip install and git clone):

    • macOS: /Users/<username>/Library/Caches/langflow
    • Linux: /home/<username>/.cache/langflow
    • Windows/WSL: C:\Users\<username>\AppData\Local\langflow\langflow\Cache

To customize log storage locations and behaviors, set the following Langflow environment variables in your Langflow .env file, and then start Langflow with uv run langflow run --env-file .env:

VariableFormatDefaultDescription
LANGFLOW_​CONFIG_​DIRStringVariesSet the Langflow configuration directory where files and logs are stored. Default path depends on your installation, as described in the preceding list.
LANGFLOW_​ENABLE_​LOG_​RETRIEVALBooleanFalseEnables retrieval of logs from your Langflow instance with Logs endpoints.
LANGFLOW_​ENVIRONMENTStringNot setAdds an environment field to every JSON log record when set (staging, production, etc.). Omitted otherwise.
LANGFLOW_​LOG_​ENVStringdefaultPrimary log format controller. container or container_​json: JSON format via JSONRenderer. container_​csv: Key-value format. default or unset: uses LANGFLOW_​PRETTY_​LOGS to determine format. See JSON log format.
LANGFLOW_​LOG_​FILEStringNot setSets the log file storage location. For example, LANGFLOW_​LOG_​FILE=path/to/logfile.log. If not set, logs are written to stdout.
LANGFLOW_​LOG_​FORMATStringNot setSet to key_​value for key-value format or console for ConsoleRenderer. Only applies when LANGFLOW_​LOG_​ENV=default and LANGFLOW_​PRETTY_​LOGS=true.
LANGFLOW_​LOG_​LEVELStringERRORGlobal log level. One of DEBUG, INFO, WARNING, ERROR, or CRITICAL.
LANGFLOW_​LOG_​LEVELSStringNot setPer-logger level overrides as name=LEVEL,name=LEVEL. Overrides LANGFLOW_​LOG_​LEVEL for specific loggers without changing the global default. Malformed entries raise a UserWarning at startup. Example: sqlalchemy.engine=WARNING,httpx=INFO.
LANGFLOW_​LOG_​REDACT_​KEYSStringNot setComma-separated extra keys to scrub from JSON log records, on top of the defaults (password, api_​key, token, authorization, cookie, etc.).
LANGFLOW_​LOG_​RETRIEVER_​BUFFER_​SIZEInteger10000Buffer size for log retrieval when LANGFLOW_​ENABLE_​LOG_​RETRIEVAL=True. Must be greater than 0.
LANGFLOW_​LOG_​ROTATIONString1 dayControls log file rotation by time (1 day, 12 hours, 1 week) or size (10 MB, 1 GB). Set to None to disable rotation.
LANGFLOW_​LOG_​TRACE_​LOCALSBooleanfalseInclude frame locals in structured tracebacks. Off by default because locals can leak secrets; enable only for local debugging.
LANGFLOW_​NATIVE_​TRACINGBooleantrueEnables the tracer to record execution traces in the Langflow database for use in Trace View. Set to false to disable.
LANGFLOW_​PRETTY_​LOGSBooleanTrueControls output format when LANGFLOW_​LOG_​ENV=default or unset. true: ConsoleRenderer (human-readable). false: JSON format.
LANGFLOW_​SERVICE_​NAMEStringlangflowValue of the service field on every JSON log record. Used as a label by log aggregators. See Grafana and Loki.
LANGFLOW_​VERSIONStringNot setAdds a version field to every JSON log record when set. Omitted otherwise.

View logs in real-time​

By default, OSS Langflow writes logs to stdout. To write logs to a file that you can tail, do the following:

  1. Set LANGFLOW_LOG_FILE to a file path in your .env file:


    _10
    LANGFLOW_LOG_FILE=/path/to/langflow.log

  2. Start Langflow with your .env file:


    _10
    uv run langflow run --env-file .env

  3. Change to the directory where you set LANGFLOW_LOG_FILE:


    _10
    cd /Users/**USERNAME**/Library/Caches/langflow

  4. Tail the log file:


    _10
    tail -f langflow.log

    If you don't see new log entries, check that Langflow is running, and perform some actions to generate logs events. You can also check the terminal where you started Langflow to see if logs are being printed there.

Flow and component logs​

After you run a flow, you can inspect the logs for each component and flow run. For example, you can inspect Message objects ingested and generated by Input and Output components.

View flow logs​

In the visual editor, click Logs to view logs for the entire flow:

Logs pane

Then, click the cells in the inputs and outputs columns to inspect the Message objects. For example, the following Message data could be the output from a Chat Input component:


_12
"messages": [
_12
{
_12
"message": "What's the recommended way to install Docker on Mac M1?",
_12
"sender": "User",
_12
"sender_name": "User",
_12
"session_id": "Session Apr 21, 17:37:04",
_12
"stream_url": null,
_12
"component_id": "ChatInput-4WKag",
_12
"files": [],
_12
"type": "text"
_12
}
_12
],

In the case of Input/Output components, the original input might not be structured as a Message object. For example, a language model component can pass a raw text response to a Chat Output component that is then transformed into a Message object.

You can find .log files for flows at your Langflow installation's log storage location. For filepaths, see Log storage.

View chat logs​

In the Playground, you can inspect the chat history for each chat session. For more information, see View chat history.

View output from a single component​

When debugging issues with the format or content of a flow's output, it can help to inspect each component's output to determine where data is being lost or malformed.

To view the output produced by a single component during the most recent run, click Inspect output on the component in the visual editor.

Access Langflow Desktop logs​

If you encounter issues with Langflow Desktop, you might need to access startup logs for debugging. Follow the steps for your operating system.

  1. Open Terminal and run:


    _10
    cd ~/Library/Logs/com.LangflowDesktop

  2. To open the folder and view the log files, run the command:


    _10
    open .

  3. Locate the langflow.log file.

You can use the log file to investigate the issue on your own, add context to a GitHub Issue, or send it to support for debugging assistance.

The log file is only created when Langflow Desktop runs. If you don't see a log file, try starting Langflow Desktop first, then check for the log file.

Log format​

By default, structlog uses ConsoleRenderer to produce human-readable output with the following structure:

For example:


_10
2026-05-17T18:18:29Z [info ] incoming request logger=langflow.api.run flow_id=flow-718 user_id=user-4823

JSON log format​

Set LANGFLOW_LOG_ENV=container to switch structlog's terminal processor to JSONRenderer, so every line written to stdout is a JSON object containing the event message, level, timestamp, logger name, exception structure, and service metadata.

A typical record in JSON mode looks like the following:


_18
{
_18
"event": "incoming request",
_18
"level": "info",
_18
"logger": "langflow.api.run",
_18
"timestamp": "2026-05-17T18:18:29.100798Z",
_18
"service": "langflow",
_18
"version": "1.10.0",
_18
"environment": "production",
_18
"user_id": "user-4823",
_18
"flow_id": "flow-718",
_18
"authorization": "***",
_18
"cookie": "***",
_18
"request_body": {
_18
"input": "Tell me a joke",
_18
"api_key": "***",
_18
"session_id": "Session Apr 21, 17:37:04"
_18
}
_18
}

When a log record includes an exception, the exception field contains a structured traceback instead of a plain string:


_16
{
_16
"event": "flow run failed",
_16
"level": "error",
_16
"logger": "langflow.api.run",
_16
"timestamp": "2026-05-17T18:18:32.412Z",
_16
"service": "langflow",
_16
"exception": [
_16
{
_16
"exc_type": "ConnectionError",
_16
"exc_value": "upstream returned 503",
_16
"frames": [
_16
{"filename": "/app/.../runner.py", "lineno": 142, "name": "run_flow"}
_16
]
_16
}
_16
]
_16
}

Frame locals, which are local variables at each stack frame, are intentionally excluded by default because they can contain API keys, request bodies, and environment values. Set LANGFLOW_LOG_TRACE_LOCALS=true to include them for local debugging.

Stdlib loggers​

Third-party libraries that use Python's standard logging module (uvicorn, sqlalchemy, httpx, langchain, asyncio) are routed through the same JSON stream in container mode. Their original logger name is preserved on the logger field, so log queries that work for application loggers also work for library output.

OpenTelemetry trace correlation​

If opentelemetry-api is installed and an active span is present, every log record automatically picks up trace_id and span_id. OpenTelemetry is not a hard dependency; if it is not installed, the processor is a no-op and nothing else changes.

See also​

Search