Update .env.example with APISIX gateway documentation

Signed-off-by: gopal-raj-suresh <gopal.raj.dummugudupu@cloud2labs.com>

Author: gopal-raj-suresh

sample_solutions/RAGChatbot/TROUBLESHOOTING.md

@@ -28,6 +28,29 @@ decoding failed due to the following error(s):
 2. If using a local domain (e.g., `inference.example.com`), replace `not-needed` with your domain name (without `https://`)
 3. Restart Docker Compose: `docker compose down && docker compose up`
 
+### Error: "Network is unreachable" or "Connection refused" with custom domain
+
+**Problem**:
+```
+httpcore.ConnectError: [Errno 101] Network is unreachable
+httpcore.ConnectError: [Errno 111] Connection refused
+```
+
+**Cause**: Docker's `host-gateway` cannot resolve to your API server IP address.
+
+**Solution**:
+
+Update `docker-compose.yml` line 17 to use your API server's actual IP:
+```yaml
+# Change from:
+- "${LOCAL_URL_ENDPOINT}:host-gateway"
+
+# To:
+- "${LOCAL_URL_ENDPOINT}:YOUR_API_SERVER_IP"
+```
+
+Then restart: `docker compose down && docker compose up`
+
 ### Error: "404 Not Found" when uploading PDF
 
 **Problem**:

sample_solutions/RAGChatbot/api/.env.example

@@ -24,8 +24,10 @@ INFERENCE_MODEL_NAME=meta-llama/Llama-3.1-8B-Instruct
 
 # APISIX Gateway Endpoints
 # Uncomment and set these when using APISIX Gateway:
-# EMBEDDING_API_ENDPOINT=https://api.example.com/bge-base-en-v1.5
-# INFERENCE_MODEL_ENDPOINT=https://api.example.com/Llama-3.1-8B-Instruct
+# IMPORTANT: Use exact APISIX route paths:
+# Example routes: /bge-base-en-v1.5-vllmcpu/* and /Llama-3.1-8B-Instruct-vllmcpu/*
+# EMBEDDING_API_ENDPOINT=https://api.example.com/bge-base-en-v1.5-vllmcpu
+# INFERENCE_MODEL_ENDPOINT=https://api.example.com/Llama-3.1-8B-Instruct-vllmcpu
 
 # Local URL Endpoint (only needed for non-public domains)
 # If using a local domain like api.example.com mapped to localhost: