Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

docs: filtering spans tutorial #5214

Merged
merged 2 commits into from
Oct 28, 2024
Merged

docs: filtering spans tutorial #5214

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
285d919
Create span_filtering_tutorial.ipynb
Jgilhuly Oct 28, 2024
b19ab64
Style fixes
Jgilhuly Oct 28, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
283 changes: 283 additions & 0 deletions tutorials/tracing/span_filtering_tutorial.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,283 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<center>\n",
" <p style=\"text-align:center\">\n",
" <img alt=\"phoenix logo\" src=\"https://storage.googleapis.com/arize-phoenix-assets/assets/phoenix-logo-light.svg\" width=\"200\"/>\n",
" <br>\n",
" <a href=\"https://docs.arize.com/phoenix/\">Docs</a>\n",
" |\n",
" <a href=\"https://github.com/Arize-ai/phoenix\">GitHub</a>\n",
" |\n",
" <a href=\"https://join.slack.com/t/arize-ai/shared_invite/zt-1px8dcmlf-fmThhDFD_V_48oU7ALan4Q\">Community</a>\n",
" </p>\n",
"</center>\n",
"<h1 align=\"center\">Filter OpenTelemetry Spans</h1>\n",
"\n",
"This tutorial shows how to filter OpenTelemetry spans based on a condition.\n",
"\n",
"If you're using multiple OTEL-compatible libraries, you may want to filter spans based on a condition. This can prevent you from sending spans to multiple backends unnecessarily.\n",
"\n",
"There are multiple approaches to do this, in this tutorial we'll show how to do this using a custom SpanProcessor."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Install Dependencies"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"!pip install -q uv\n",
"!uv pip install --system -q opentelemetry-sdk opentelemetry-exporter-otlp-proto-http openinference-instrumentation-openai openai python-dotenv"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"\n",
"from dotenv import load_dotenv\n",
"from opentelemetry import trace as trace_api\n",
"from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter\n",
"from opentelemetry.sdk.trace import TracerProvider\n",
"from opentelemetry.sdk.trace.export import ConsoleSpanExporter, SpanExporter, SpanProcessor\n",
"\n",
"load_dotenv()\n",
"\n",
"from getpass import getpass\n",
"\n",
"import openai\n",
"from openinference.instrumentation.openai import OpenAIInstrumentor"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Define a custom SpanProcessor\n",
"\n",
"We'll extend the SpanProcessor class to add a condition that determines whether a span should be exported."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"class ConditionalSpanProcessor(SpanProcessor):\n",
" def __init__(self, exporter: SpanExporter, condition: callable):\n",
" self.exporter = exporter\n",
" self.condition = condition\n",
"\n",
" def on_start(self, span, parent_context):\n",
" pass\n",
"\n",
" def on_end(self, span):\n",
" # Only export spans that meet the condition\n",
" if self.condition(span):\n",
" self.exporter.export([span])\n",
"\n",
" def shutdown(self):\n",
" self.exporter.shutdown()\n",
"\n",
" def force_flush(self, timeout_millis=None):\n",
" self.exporter.force_flush(timeout_millis)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Along with this class, we'll define two conditions: one for sending spans to the console, and one for sending spans to Phoenix."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Define conditions for sending spans to specific exporters\n",
"def console_condition(span):\n",
" return \"console\" in span.name # Example: send to Console if \"console\" is in the span name\n",
"\n",
"\n",
"def phoenix_condition(span):\n",
" # return \"phoenix\" in span.name # Example: send to Phoenix if \"phoenix\" is in the span name\n",
" return not console_condition(\n",
" span\n",
" ) # Example: send to Phoenix if \"console\" is not in the span name"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Use our custom SpanProcessor to set up instrumentation\n",
"In this example, we'll use Phoenix as one of our destinations, and the console as the other. You could instead add any other exporters you'd like in this approach.\n",
"\n",
"If you need to set up an API key for Phoenix, you can do so [here](https://app.phoenix.arize.com/)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"if not (phoenix_api_key := os.getenv(\"PHOENIX_API_KEY\")):\n",
" phoenix_api_key = getpass(\"Enter your Phoenix API Key: \")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Define the console exporter"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"tracer_provider = TracerProvider()\n",
"\n",
"# Create the Console exporter\n",
"console_exporter = ConsoleSpanExporter()\n",
"\n",
"# Add the Console exporter to the tracer provider\n",
"tracer_provider.add_span_processor(ConditionalSpanProcessor(console_exporter, console_condition))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Define the Phoenix exporter"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Add Phoenix API Key to the headers for tracing and API access\n",
"os.environ[\"PHOENIX_CLIENT_HEADERS\"] = f\"api_key={phoenix_api_key}\"\n",
"\n",
"# Create the Phoenix exporter\n",
"otlp_exporter = OTLPSpanExporter(endpoint=\"https://app.phoenix.arize.com/v1/traces\")\n",
"\n",
"# Add the Phoenix exporter to the tracer provider\n",
"tracer_provider.add_span_processor(ConditionalSpanProcessor(otlp_exporter, phoenix_condition))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Add the exporters to the tracer provider"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Set the tracer provider\n",
"trace_api.set_tracer_provider(tracer_provider)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Run our app and view traces"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"if not (openai_api_key := os.getenv(\"OPENAI_API_KEY\")):\n",
" os.environ[\"OPENAI_API_KEY\"] = getpass(\"Enter your OpenAI API Key: \")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This first span will be exported to the console. Here we're using manual instrumentation to create the span."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Create a tracer\n",
"tracer = trace_api.get_tracer(__name__)\n",
"\n",
"# Example of creating and exporting spans\n",
"with tracer.start_as_current_span(\"console-span\"):\n",
" print(\"This span will be exported to Console only.\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This next span will be exported only to Phoenix. In this case, we're using our auto-instrumentor for OpenAI to generate the span."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Auto-instrumentors can still be used with this setup\n",
"OpenAIInstrumentor().instrument(tracer_provider=tracer_provider, skip_dep_check=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"client = openai.OpenAI()\n",
"client.chat.completions.create(\n",
" model=\"gpt-4o-mini\", messages=[{\"role\": \"user\", \"content\": \"Hello, world!\"}]\n",
")"
]
}
],
"metadata": {
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Loading