Merge branch 'main' into ossf-scorecard

Author: xrmx

CHANGELOG.md

@@ -25,6 +25,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - `opentelemetry-instrumentation-dbapi`, `opentelemetry-instrumentation-django`,
   `opentelemetry-instrumentation-sqlalchemy`: Fix sqlcomment for non string query and composable object.
   ([#3113](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3113))
+- `opentelemetry-instrumentation-grpc` Fix error when using gprc versions <= 1.50.0 with unix sockets.
+  ([[#3393](https://github.com/open-telemetry/opentelemetry-python-contrib/issues/3393)])
 
 ## Version 1.31.0/0.52b0 (2025-03-12)
 

instrumentation/opentelemetry-instrumentation-botocore/README.rst

@@ -8,6 +8,44 @@ OpenTelemetry Botocore Tracing
 
 This library allows tracing requests made by the Botocore library.
 
+Extensions
+----------
+
+The instrumentation supports creating extensions for AWS services for enriching what is collected. We have extensions
+for the following AWS services:
+
+- Bedrock Runtime
+- DynamoDB
+- Lambda
+- SNS
+- SQS
+
+Bedrock Runtime
+***************
+
+This extension implements the GenAI semantic conventions for the following API calls:
+
+- Converse
+- ConverseStream
+- InvokeModel
+- InvokeModelWithResponseStream
+
+For the Converse and ConverseStream APIs tracing, events and metrics are implemented.
+
+For the InvokeModel and InvokeModelWithResponseStream APIs tracing, events and metrics implemented only for a subset of
+the available models, namely:
+
+- Amazon Titan models
+- Amazon Nova models
+- Anthropic Claude
+
+Tool calls with InvokeModel and InvokeModelWithResponseStream APIs are supported with:
+
+- Amazon Nova models
+- Anthropic Claude 3+
+
+If you don't have an application using Bedrock APIs yet, try our `zero-code examples <examples/bedrock-runtime/zero-code>`_.
+
 Installation
 ------------
 

instrumentation/opentelemetry-instrumentation-botocore/src/opentelemetry/instrumentation/botocore/__init__.py

@@ -78,40 +78,6 @@ def response_hook(span, service_name, operation_name, result):
     )
     ec2 = session.create_client("ec2", region_name="us-west-2")
     ec2.describe_instances()
-
-Extensions
-----------
-
-The instrumentation supports creating extensions for AWS services for enriching what is collected. We have extensions
-for the following AWS services:
-
-- Bedrock Runtime
-- DynamoDB
-- Lambda
-- SNS
-- SQS
-
-Bedrock Runtime
-***************
-
-This extension implements the GenAI semantic conventions for the following API calls:
-
-- Converse
-- ConverseStream
-- InvokeModel
-- InvokeModelWithResponseStream
-
-For the Converse and ConverseStream APIs tracing, events and metrics are implemented.
-
-For the InvokeModel and InvokeModelWithResponseStream APIs tracing, events and metrics implemented only for a subset of
-the available models, namely:
-- Amazon Titan models
-- Amazon Nova models
-- Anthropic Claude
-
-Tool calls with InvokeModel and InvokeModelWithResponseStream APIs are supported with:
-- Amazon Nova models
-- Anthropic Claude 3+
 """
 
 import logging

instrumentation/opentelemetry-instrumentation-grpc/src/opentelemetry/instrumentation/grpc/_server.py

@@ -239,7 +239,7 @@ def _start_span(
         # * ipv4:127.0.0.1:57284
         # * ipv4:10.2.1.1:57284,127.0.0.1:57284
         #
-        if context.peer() != "unix:":
+        if not context.peer().startswith("unix:"):
             try:
                 ip, port = (
                     context.peer()