how to set otel logs attribute

3 min read 23-01-2025

OpenTelemetry (OTel) is a powerful tool for collecting and exporting telemetry data, including logs. Adding attributes to your logs provides crucial context, making them much more valuable for debugging, monitoring, and analysis. This article will guide you through setting log attributes in various popular programming languages and frameworks.

Understanding OTel Log Attributes

Before diving into the specifics, let's clarify what log attributes are. Attributes are key-value pairs associated with a log record. They provide additional information beyond the log message itself, such as:

Environment: environment: "production"
User ID: user_id: "12345"
Request ID: request_id: "abcdef123456"
Database Name: db_name: "mydatabase"

These attributes allow you to filter, aggregate, and analyze your logs more effectively. For example, you can easily find all logs related to a specific user or a particular database.

Setting Log Attributes: Practical Examples

The exact method for setting log attributes varies depending on the programming language and the OTel library you're using. However, the underlying principle remains consistent: you associate key-value pairs with the log record before exporting it.

Python

Python's OpenTelemetry library provides a straightforward way to add attributes. Here's an example using the opentelemetry-instrumentation package:

from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPTraceExporter
from opentelemetry.sdk.resources import SERVICE_NAME, Resource
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor
from opentelemetry.trace import get_tracer_provider

trace.set_tracer_provider(
    TracerProvider(resource=Resource.get_empty().merge({"service.name": "my-service"}))
)
trace.get_tracer_provider().add_span_processor(
    BatchSpanProcessor(OTLPTraceExporter())
)
tracer = trace.get_tracer(__name__)

with tracer.start_as_current_span("my_operation"):
    with tracer.start_as_current_span("my_sub_operation", attributes={"attribute1":"value1", "attribute2": "value2"}):
        print("This is a log message.")
        logger.info("Another log message")

This code snippet shows how to add attributes directly to a span. The attributes will automatically be associated with any log records created within that span.

Node.js

In Node.js, you can use the @opentelemetry/sdk-node package. Here's a basic example:

const { trace } = require('@opentelemetry/sdk-node');
const { NodeTracerProvider } = require('@opentelemetry/sdk-trace-node');
const { SimpleSpanProcessor } = require('@opentelemetry/sdk-trace-base');
const { ConsoleSpanExporter } = require('@opentelemetry/sdk-trace-node');

const provider = new NodeTracerProvider();
provider.addSpanProcessor(new SimpleSpanProcessor(new ConsoleSpanExporter()));
provider.register();

const tracer = provider.getTracer('my-service');

const span = tracer.startSpan('myOperation', { attributes: { 'attribute1': 'value1', 'attribute2': 'value2' } });
// ... your logging code ...
span.end();

This example demonstrates how to set attributes when initiating a span. Any logs generated within this span will include these attributes.

Java

In Java, using the OpenTelemetry API, you would typically do something similar to this (using the opentelemetry-api and a suitable exporter):

import io.opentelemetry.api.common.Attributes;
import io.opentelemetry.api.trace.Span;
import io.opentelemetry.api.trace.Tracer;
import io.opentelemetry.context.Scope;
import io.opentelemetry.sdk.trace.SdkTracerProvider;
import io.opentelemetry.sdk.trace.export.SimpleSpanProcessor;
import io.opentelemetry.sdk.trace.export.SpanExporter;


// ... your OpenTelemetry configuration ...

Tracer tracer = tracerProvider.get("my-service");

Span span = tracer.spanBuilder("myOperation").setAttributes(Attributes.of("attribute1", "value1", "attribute2", "value2")).startSpan();
try (Scope scope = span.makeCurrent()) {
    // ... your logging code that will include the attributes ...
    log.info("my log message");
} finally {
    span.end();
}

Remember to replace placeholders like "my-service" with your actual service name.

Best Practices for Setting Log Attributes

Use meaningful names: Choose descriptive attribute keys that clearly communicate the data's purpose.
Maintain consistency: Use the same attribute names consistently across your application.
Avoid sensitive data: Don't include personally identifiable information (PII) or other sensitive data in your log attributes.
Use structured logging: Structure your logs using a format like JSON to make them easier to parse and analyze.
Keep it concise: Limit the number of attributes per log record to avoid excessive overhead.

By following these guidelines and leveraging the examples provided, you can effectively utilize OTel log attributes to enhance the value and usability of your logs. Remember to consult the specific documentation for your chosen programming language and OTel libraries for detailed instructions and best practices.