Mask Span Attributes

There are scenarios where adjusting the observability level of your tracing is necessary. For example, you might want to prevent sensitive data from being logged for security purposes, or you might aim to reduce the payload size by limiting the size of base64 encoded images logged.

The traceAI Specification provides a set of environment variables that you can configure to meet your observability requirements. Additionally, the traceAI auto-instrumentors allow you to specify a trace configuration directly in your code, eliminating the need to set environment variables if you prefer.

The available settings include:

Environment Variable Name	Description	Type	Default
FI_HIDE_INPUTS	Hides input values, all input messages, and embedding input text	bool	False
FI_HIDE_OUTPUTS	Hides output values and all output messages	bool	False
FI_HIDE_INPUT_MESSAGES	Hides all input messages and embedding input text	bool	False
FI_HIDE_OUTPUT_MESSAGES	Hides all output messages	bool	False
FI_HIDE_INPUT_IMAGES	Hides images from input messages	bool	False
FI_HIDE_INPUT_TEXT	Hides text from input messages and input embeddings	bool	False
FI_HIDE_OUTPUT_TEXT	Hides text from output messages	bool	False
FI_HIDE_EMBEDDING_VECTORS	Hides returned embedding vectors	bool	False
FI_BASE64_IMAGE_MAX_LENGTH	Hides the character count of a base64 encoded image	int	32,000

To configure these settings, you can:

Set the environment variables as outlined above
Define the configuration within your code as demonstrated below
Opt for the default values by taking no action

You can use a combination of these methods, with the following order of precedence:

Values defined in the TraceConfig within your code
Environment variables
Default values

Below is an example of how to configure these settings in code using our OpenAI Python instrumentor. This configuration is applicable to all of our auto-instrumentors.

from fi_instrumentation import TraceConfig


config = TraceConfig(        
    hide_inputs=False, 
    hide_outputs=False,
    hide_input_messages=False,
    hide_output_messages=False,
    hide_input_images=False,
    hide_input_text=False,
    hide_output_text=False,
    hide_embedding_vectors=False,
    base64_image_max_length=32000, 
)

from traceai_openai import OpenAIInstrumentor
OpenAIInstrumentor().instrument(
    tracer_provider=trace_provider, 
    config=config,
)

SDK Reference