| 
 | 1 | +# Install with Docker Compose  | 
 | 2 | + | 
 | 3 | +Run Semantic Router + Envoy locally using Docker Compose v2.  | 
 | 4 | + | 
 | 5 | +## Prerequisites  | 
 | 6 | + | 
 | 7 | +- Docker Engine and Docker Compose v2 (use the `docker compose` command, not the legacy `docker-compose`)  | 
 | 8 | + | 
 | 9 | +   ```bash  | 
 | 10 | +   # Verify  | 
 | 11 | +   docker compose version  | 
 | 12 | +   ```  | 
 | 13 | + | 
 | 14 | +   Install Docker Compose v2 for Ubuntu(if missing), see more in [Docker Compose Plugin Installation](https://docs.docker.com/compose/install/linux/#install-using-the-repository)  | 
 | 15 | + | 
 | 16 | +   ```bash  | 
 | 17 | +   # Remove legacy v1 if present (optional)  | 
 | 18 | +   sudo apt-get remove -y docker-compose || true  | 
 | 19 | + | 
 | 20 | +   sudo apt-get update  | 
 | 21 | +   sudo apt-get install -y ca-certificates curl gnupg  | 
 | 22 | +   sudo install -m 0755 -d /etc/apt/keyrings  | 
 | 23 | +   curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --yes --dearmor -o /etc/apt/keyrings/docker.gpg  | 
 | 24 | +   echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu $(. /etc/os-release && echo $VERSION_CODENAME) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null  | 
 | 25 | +   sudo apt-get update  | 
 | 26 | +   sudo apt-get install -y docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin  | 
 | 27 | + | 
 | 28 | +   docker compose version  | 
 | 29 | +   ```  | 
 | 30 | + | 
 | 31 | +- Ensure ports 8801, 50051, 19000 are free  | 
 | 32 | + | 
 | 33 | +## Install and Run with Docker Compose v2  | 
 | 34 | + | 
 | 35 | +1) Clone the repo and move into it (from your workspace root):  | 
 | 36 | + | 
 | 37 | +```bash  | 
 | 38 | +git clone https://github.com/vllm-project/semantic-router.git  | 
 | 39 | +cd semantic-router  | 
 | 40 | +```  | 
 | 41 | + | 
 | 42 | +2) Download required models (classification models):  | 
 | 43 | + | 
 | 44 | +```bash  | 
 | 45 | +make download-models  | 
 | 46 | +```  | 
 | 47 | + | 
 | 48 | +This downloads the classification models used by the router:  | 
 | 49 | + | 
 | 50 | +- Category classifier (ModernBERT-base)  | 
 | 51 | +- PII classifier (ModernBERT-base)  | 
 | 52 | +- Jailbreak classifier (ModernBERT-base)  | 
 | 53 | + | 
 | 54 | +Note: The BERT similarity model defaults to a remote Hugging Face model. See Troubleshooting for offline/local usage.  | 
 | 55 | + | 
 | 56 | +3) Start the services with Docker Compose v2:  | 
 | 57 | + | 
 | 58 | +```bash  | 
 | 59 | +# Start core services (semantic-router + envoy)  | 
 | 60 | +docker compose up --build  | 
 | 61 | + | 
 | 62 | +# Or run in background (recommended)  | 
 | 63 | +docker compose up --build -d  | 
 | 64 | + | 
 | 65 | +# With testing profile (includes mock vLLM)  | 
 | 66 | +docker compose --profile testing up --build  | 
 | 67 | +```  | 
 | 68 | + | 
 | 69 | +4) Verify  | 
 | 70 | + | 
 | 71 | +- Semantic Router (gRPC): localhost:50051  | 
 | 72 | +- Envoy Proxy: http://localhost:8801  | 
 | 73 | +- Envoy Admin: http://localhost:19000  | 
 | 74 | + | 
 | 75 | +## Common Operations  | 
 | 76 | + | 
 | 77 | +```bash  | 
 | 78 | +# View service status  | 
 | 79 | +docker compose ps  | 
 | 80 | + | 
 | 81 | +# Follow logs for the router service  | 
 | 82 | +docker compose logs -f semantic-router  | 
 | 83 | + | 
 | 84 | +# Exec into the router container  | 
 | 85 | +docker compose exec semantic-router bash  | 
 | 86 | + | 
 | 87 | +# Stop and clean up containers  | 
 | 88 | +docker compose down  | 
 | 89 | +```  | 
 | 90 | + | 
 | 91 | +## Troubleshooting  | 
 | 92 | + | 
 | 93 | +### 1) Router exits immediately with a Hugging Face DNS/download error  | 
 | 94 | + | 
 | 95 | +Symptoms (from `docker compose logs -f semantic-router`):  | 
 | 96 | + | 
 | 97 | +```  | 
 | 98 | +Failed to initialize BERT: request error: https://huggingface.co/... Dns Failed: resolve dns name 'huggingface.co:443'  | 
 | 99 | +```  | 
 | 100 | + | 
 | 101 | +Why: `bert_model.model_id` in `config/config.yaml` points to a remote model (`sentence-transformers/all-MiniLM-L12-v2`). If the container cannot resolve or reach the internet, startup fails.  | 
 | 102 | + | 
 | 103 | +Fix options:  | 
 | 104 | + | 
 | 105 | +- Allow network access in the container (online):  | 
 | 106 | +  - Ensure your host can resolve DNS, or add DNS servers to the `semantic-router` service in `docker-compose.yml`:  | 
 | 107 | + | 
 | 108 | +      ```yaml  | 
 | 109 | +      services:  | 
 | 110 | +         semantic-router:  | 
 | 111 | +            # ...  | 
 | 112 | +            dns:  | 
 | 113 | +               - 1.1.1.1  | 
 | 114 | +               - 8.8.8.8  | 
 | 115 | +      ```  | 
 | 116 | +        | 
 | 117 | +  - If behind a proxy, set `http_proxy/https_proxy/no_proxy` env vars for the service.  | 
 | 118 | + | 
 | 119 | +- Use a local copy of the model (offline):  | 
 | 120 | +   1. Download `sentence-transformers/all-MiniLM-L12-v2` to `./models/sentence-transformers/all-MiniLM-L12-v2/` on the host.  | 
 | 121 | +   2. Update `config/config.yaml` to use the local path (mounted into the container at `/app/models`):  | 
 | 122 | + | 
 | 123 | +       ```yaml  | 
 | 124 | +       bert_model:  | 
 | 125 | +          model_id: "models/sentence-transformers/all-MiniLM-L12-v2"  | 
 | 126 | +          threshold: 0.6  | 
 | 127 | +          use_cpu: true  | 
 | 128 | +       ```  | 
 | 129 | + | 
 | 130 | +   3. Recreate services: `docker compose up -d --build`  | 
 | 131 | + | 
 | 132 | +### 2) Port already in use  | 
 | 133 | + | 
 | 134 | +Make sure 8801, 50051, 19000 are not bound by other processes. Adjust ports in `docker-compose.yml` if needed.  | 
0 commit comments