feat: Endpoint Persistence using Network Volume (phase 2)

.github/workflows/ci.yml

@@ -108,7 +108,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
       - name: Build CPU Docker image
@@ -183,7 +183,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
       - name: Build and push GPU Docker image (main)
@@ -236,7 +236,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
       - name: Build and push CPU Docker image (main)
@@ -299,7 +299,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
       - name: Build and push GPU Docker image (prod)
@@ -362,7 +362,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
       - name: Build and push CPU Docker image (prod)

.github/workflows/deploy.yml

@@ -84,7 +84,7 @@ jobs:
       - name: Setup dependencies
         run: |
           uv sync
-          git submodule update --remote --rebase
+          git submodule update
           cp tetra-rp/src/tetra_rp/protos/remote_execution.py .
 
       - name: Build and push GPU Docker image

CLAUDE.md

@@ -146,6 +146,7 @@ git submodule update --remote --rebase    # Update tetra-rp to latest
 - `RUNPOD_ENDPOINT_ID`: Used for workspace isolation (automatically set by RunPod)
 - `HF_HUB_ENABLE_HF_TRANSFER`: Set to "1" in Dockerfile to enable accelerated HuggingFace downloads
 - `HF_TOKEN`: Optional authentication token for private/gated HuggingFace models
+- `HF_HOME=/hf-cache`: HuggingFace cache location, set outside `/root/.cache` to exclude from volume sync
 - `DEBIAN_FRONTEND=noninteractive`: Set during system package installation
 - `UV_CACHE_DIR`: Package cache configuration
 - `VIRTUAL_ENV`: Virtual environment path configuration

Dockerfile

@@ -4,6 +4,8 @@ WORKDIR /app
 
 # Enable HuggingFace transfer acceleration
 ENV HF_HUB_ENABLE_HF_TRANSFER=1
+# Relocate HuggingFace cache outside /root/.cache to exclude from volume sync
+ENV HF_HOME=/hf-cache
 
 # Install system dependencies and uv
 RUN apt-get update && apt-get install -y --no-install-recommends \

Makefile

@@ -31,8 +31,7 @@ clean: # Remove build artifacts and cache files
 	find . -type f -name "*.pkl" -delete
 
 setup: dev # Initialize project, sync deps, update submodules
-	git submodule init
-	git submodule update --remote --rebase
+	git submodule update --init --recursive
 	cp tetra-rp/src/tetra_rp/protos/remote_execution.py src/
 
 build: # Build both GPU and CPU Docker images

docs/System_Python_Runtime_Architecture.md

@@ -71,7 +71,7 @@ graph TB
 ### System Installation Strategy
 ```python
 # Docker environment
-command = ["uv", "pip", "install", "--system", "--no-cache"] + packages
+command = ["uv", "pip", "install", "--system"] + packages
 
 # Local environment
 command = ["uv", "pip", "install", "--python-preference=managed"] + packages