Further fixes for Docker + Windows

.dockerignore

@@ -14,4 +14,5 @@ logo.*
 *.sample
 .env*
 Dockerfile
-docker-compose.yml
+docker-compose.yml
+*.log

Dockerfile

@@ -33,6 +33,12 @@ USER appuser
 # Install application into container
 COPY . .
 
+# Expose the Streamlit port
+EXPOSE 8501
+
+# Setup a health check against Streamlit
+HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
+
 # Run the application
-ENTRYPOINT ["python", "-m", "streamlit"]
-CMD ["run", "main.py"]
+ENTRYPOINT [ "python", "-m", "streamlit" ]
+CMD ["run", "main.py", "--server.port=8501", "--server.address=0.0.0.0"]

components/page_state.py

@@ -21,15 +21,16 @@ def set_initial_state():
         try:
             models = get_models()
             st.session_state["ollama_models"] = models
-        except Exception as err:
-            logs.log.warn(
-                f"Warning: Initial loading of Ollama models failed. You might be hosting Ollama somewhere other than localhost. -- {err}"
-            )
+        except Exception:
             st.session_state["ollama_models"] = []
             pass
 
     if "selected_model" not in st.session_state:
-        st.session_state["selected_model"] = st.session_state["ollama_models"][0]
+        try:
+            st.session_state["selected_model"] = st.session_state["ollama_models"][0]
+        except Exception:
+            st.session_state["selected_model"] = None
+            pass
 
     if "messages" not in st.session_state:
         st.session_state["messages"] = [

docker-compose.yml

@@ -3,18 +3,20 @@ services:
   local-rag:
     container_name: local-rag
     image: jonfairbanks/local-rag
-    network_mode: host
     restart: unless-stopped
     environment:
       - TZ=America/Los_Angeles
     ports:
       - '8501:8501/tcp'
     volumes:
-      - .:/home/appuser:rw
+      - ./data:/home/appuser/data:rw
     deploy:
         resources:
           reservations:
             devices:
               - driver: nvidia
                 device_ids: ['0']
-                capabilities: [gpu]
+                capabilities: [gpu]
+
+volumes:
+ local-rag: {}

docs/todo.md

@@ -17,6 +17,7 @@ Although not final, items are generally sorted from highest to lowest priority.
     - [ ] Websites
 - [x] Export Data (Chat History, ...)
 - [x] Docker Support
+    - [x] Windows Support
 - [ ] Extract Metadata and Load into Index
 - [ ] Parallelize Document Embeddings
 - [ ] Swap to OpenAI compatible endpoints
@@ -31,6 +32,7 @@ Although not final, items are generally sorted from highest to lowest priority.
 - [x] Show Loaders in UI (File Uploads, Conversions, ...)
 - [x] View and Manage Imported Files
 - [x] About Tab in Sidebar w/ Resources
+- [x] Enable Caching
 - [ ] Allow Users to Set LLM Settings
     - [x] System Prompt
     - [ ] Chat Mode
@@ -56,7 +58,7 @@ Although not final, items are generally sorted from highest to lowest priority.
 
 ### Known Issues & Bugs
 
-- [ ] **HIGH PRIORITY:** Upon sending a Chat message, the File Processing expander appears to re-run itself (seems something is not using state correctly)
+- [x] Upon sending a Chat message, the File Processing expander appears to re-run itself (seems something is not using state correctly)
 - [ ] Refreshing the page loses all state (expected Streamlit behavior; need to implement local-storage)
 - [x] Files can be uploaded before Ollama config is set, leading to embedding errors
 - [x] Assuming Ollama is hosted on localhost, Models are automatically loaded and selected, but the dropdown does not render the selected option