genai-function-calling: removes references to APM server from env and READMEs

genai-function-calling/README.md

@@ -45,14 +45,14 @@ sequenceDiagram
     activate Agent
     Note over Agent: invokes get_latest_elasticsearch_version(majorVersion=8)
 
-    Agent -->> LLM: [user, assistant, tool: "8.17.4"]
+    Agent -->> LLM: [user, assistant, tool: "8.18.0"]
     Note over Agent: LLM is stateless, the tool result is sent back with prior messages
     deactivate Agent
     activate LLM
 
-    LLM ->> Agent: content: "The latest version of Elasticsearch 8 is 8.17.4"
+    LLM ->> Agent: content: "The latest version of Elasticsearch 8 is 8.18.0"
     deactivate LLM
-    Note over Agent: "The latest version of Elasticsearch 8 is 8.17.4"
+    Note over Agent: "The latest version of Elasticsearch 8 is 8.18.0"
 ```
 
 The GenAI framework not only abstracts the above loop, but also LLM plugability
@@ -152,14 +152,14 @@ sequenceDiagram
     activate Agent
     Note over Agent: invokes get_latest_elasticsearch_version(majorVersion=8)
 
-    Agent -->> LLM: [user, assistant, tool: "8.17.4"]
+    Agent -->> LLM: [user, assistant, tool: "8.18.0"]
     Note over Agent: LLM is stateless, the tool result is sent back with prior messages
     deactivate Agent
     activate LLM
 
-    LLM ->> Agent: content: "The latest version of Elasticsearch 8 is 8.17.4"
+    LLM ->> Agent: content: "The latest version of Elasticsearch 8 is 8.18.0"
     deactivate LLM
-    Note over Agent: "The latest version of Elasticsearch 8 is 8.17.4"
+    Note over Agent: "The latest version of Elasticsearch 8 is 8.18.0"
 
     Agent ->> MCP: Close stdin
     activate MCP

genai-function-calling/openai-agents/README.md

@@ -14,12 +14,7 @@ Agents support is via [OpenInference][openinference].
 Copy [env.example](env.example) to `.env` and update its `OPENAI_API_KEY`.
 
 An OTLP compatible endpoint should be listening for traces, metrics and logs on
-`http://localhost:4317`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
-
-For example, if Elastic APM server is running locally, edit `.env` like this:
-```
-OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-```
+`http://localhost:4318`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
 
 ## Run with Docker
 
@@ -92,7 +87,7 @@ dotenv -f ../.env run -- pytest
 ## Notes
 
 The LLM should generate something like "The latest stable version of
-Elasticsearch is 8.17.4", unless it hallucinates. Just run it again, if you
+Elasticsearch is 8.18.0", unless it hallucinates. Just run it again, if you
 see something else.
 
 OpenAI Agents SDK's OpenTelemetry instrumentation is via

genai-function-calling/openai-agents/cassettes/test_main.yaml

@@ -834,9 +834,9 @@ interactions:
               "manifest": "https://artifacts.elastic.co/downloads/8.17.2.json"
             },
             {
-              "version": "8.17.3",
+              "version": "8.18.0",
               "public_release_date": "2025-03-04",
-              "manifest": "https://artifacts.elastic.co/downloads/8.17.3.json"
+              "manifest": "https://artifacts.elastic.co/downloads/8.18.0.json"
             },
             {
               "version": "8.2.0",
@@ -1007,7 +1007,7 @@ interactions:
           {
             "role": "tool",
             "tool_call_id": "call_pT4CJ0D2kmnTP5WoVhv2edrt",
-            "content": "8.17.3"
+            "content": "8.18.0"
           }
         ],
         "model": "gpt-4o-mini",
@@ -1083,7 +1083,7 @@ interactions:
               "index": 0,
               "message": {
                 "role": "assistant",
-                "content": "The latest version of Elasticsearch 8 is 8.17.3.",
+                "content": "The latest version of Elasticsearch 8 is 8.18.0.",
                 "refusal": null,
                 "annotations": []
               },

genai-function-calling/openai-agents/env.example

@@ -27,10 +27,11 @@ OPENAI_API_KEY=
 
 OTEL_SERVICE_NAME=genai-function-calling
 
-# OTEL_EXPORTER_* variables are not required. If you would like to change your
-# OTLP endpoint to Elastic APM server using HTTP, uncomment the following:
-# OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-# OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
+# Default to send logs, traces and metrics to an OpenTelemetry collector,
+# accessible via localhost. For example, Elastic Distribution of OpenTelemetry
+# (EDOT) Collector.
+OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
 
 # Change to 'false' to hide prompt and completion content
 OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT=true

genai-function-calling/openai-agents/main_test.py

@@ -14,4 +14,4 @@ async def test_main(default_openai_env, capsys):
 
     reply = capsys.readouterr().out.strip()
 
-    assert reply == "The latest version of Elasticsearch 8 is 8.17.3."
+    assert reply == "The latest version of Elasticsearch 8 is 8.18.0."

genai-function-calling/semantic-kernel-dotnet/README.md

@@ -11,12 +11,7 @@ of OpenTelemetry (EDOT) .NET, by prepending its command with `instrument.sh`.
 Copy [env.example](env.example) to `.env` and update its `OPENAI_API_KEY`.
 
 An OTLP compatible endpoint should be listening for traces, metrics and logs on
-`http://localhost:4317`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
-
-For example, if Elastic APM server is running locally, edit `.env` like this:
-```
-OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-```
+`http://localhost:4318`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
 
 ## Run with Docker
 
@@ -27,7 +22,7 @@ docker compose run --build --rm genai-function-calling
 ## Notes
 
 The LLM should generate something like "The latest stable version of
-Elasticsearch is 8.17.4", unless it hallucinates. Just run it again, if you
+Elasticsearch is 8.18.0", unless it hallucinates. Just run it again, if you
 see something else.
 
 Semantic Kernel .NET's OpenTelemetry instrumentation uses the following custom

genai-function-calling/semantic-kernel-dotnet/env.example

@@ -30,9 +30,10 @@ SEMANTICKERNEL_EXPERIMENTAL_GENAI_ENABLE_OTEL_DIAGNOSTICS=true
 SEMANTICKERNEL_EXPERIMENTAL_GENAI_ENABLE_OTEL_DIAGNOSTICS_SENSITIVE=true
 OTEL_DOTNET_AUTO_TRACES_ADDITIONAL_SOURCES="Microsoft.SemanticKernel*"
 
-# OTEL_EXPORTER_* variables are not required. If you would like to change your
-# OTLP endpoint to Elastic APM server using HTTP, uncomment the following:
-# OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-# OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
+# Default to send logs, traces and metrics to an OpenTelemetry collector,
+# accessible via localhost. For example, Elastic Distribution of OpenTelemetry
+# (EDOT) Collector.
+OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
 
 OTEL_SERVICE_NAME=genai-function-calling

genai-function-calling/spring-ai/README.md

@@ -14,12 +14,7 @@ of Spring AI.
 Copy [env.example](env.example) to `.env` and update its `OPENAI_API_KEY`.
 
 An OTLP compatible endpoint should be listening for traces, metrics and logs on
-`http://localhost:4317`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
-
-For example, if Elastic APM server is running locally, edit `.env` like this:
-```
-OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-```
+`http://localhost:4318`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
 
 ## Run with Docker
 

genai-function-calling/spring-ai/env.example

@@ -35,9 +35,10 @@ SPRING_AUTOCONFIGURE_EXCLUDE=org.springframework.boot.actuate.autoconfigure.obse
 # SPRING_AUTOCONFIGURE_EXCLUDE=org.springframework.ai.autoconfigure.openai.OpenAiAutoConfiguration
 OTEL_INSTRUMENTATION_MICROMETER_ENABLED=true
 
-# OTEL_EXPORTER_* variables are not required. If you would like to change your
-# OTLP endpoint to Elastic APM server using HTTP, uncomment the following:
-# OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-# OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
+# Default to send logs, traces and metrics to an OpenTelemetry collector,
+# accessible via localhost. For example, Elastic Distribution of OpenTelemetry
+# (EDOT) Collector.
+OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
 
 OTEL_SERVICE_NAME=genai-function-calling

genai-function-calling/vercel-ai/README.md

@@ -11,12 +11,7 @@ of OpenTelemetry (EDOT) Node.js, by requiring `@elastic/opentelemetry-node`
 Copy [env.example](env.example) to `.env` and update its `OPENAI_API_KEY`.
 
 An OTLP compatible endpoint should be listening for traces, metrics and logs on
-`http://localhost:4317`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
-
-For example, if Elastic APM server is running locally, edit `.env` like this:
-```
-OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-```
+`http://localhost:4318`. If not, update `OTEL_EXPORTER_OTLP_ENDPOINT` as well.
 
 ## Run with Docker
 

genai-function-calling/vercel-ai/env.example

@@ -23,10 +23,11 @@ OPENAI_API_KEY=
 ## "Name" from https://oai.azure.com/resource/deployments
 # CHAT_MODEL=YOUR_DEPLOYMENT_NAME
 
-# OTEL_EXPORTER_* variables are not required. If you would like to change your
-# OTLP endpoint to Elastic APM server using HTTP, uncomment the following:
-# OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200
-# OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
+# Default to send logs, traces and metrics to an OpenTelemetry collector,
+# accessible via localhost. For example, Elastic Distribution of OpenTelemetry
+# (EDOT) Collector.
+OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
 
 OTEL_SERVICE_NAME=genai-function-calling
 # Don't print status message on startup