Add debug logging

document-authoring.js

@@ -178,7 +178,7 @@ let retryCount = 0;
 const MAX_RETRIES = 10; // Will try for 20 seconds (10 attempts * 2 second interval)
 
 function checkServicesStatus() {
-  fetch('/v1/chat/completions', {
+  fetch('/inference/api/v1/chat/completions', {
     method: 'POST',
     headers: {
       'Content-Type': 'application/json'
@@ -287,7 +287,7 @@ script.onload = async () => {
 
   async function translate(content, targetLang, sourceLang = 'English') {
     try {
-      const response = await fetch('/v1/chat/completions', {
+      const response = await fetch('/inference/api/v1/chat/completions', {
         method: 'POST',
         headers: {
           'Content-Type': 'application/json',
@@ -421,7 +421,7 @@ ${targetLang}: "Las empresas usan Nutrient para..."`
   // Add language detection function
   async function detectLanguage(text) {
     try {
-      const response = await fetch('/client/inference/api/v1/chat/completions', {
+      const response = await fetch('/inference/api/v1/chat/completions', {
         method: 'POST',
         headers: {
           'Content-Type': 'application/json',

nginx.conf

@@ -15,7 +15,7 @@ http {
             proxy_pass http://127.0.0.1:4000/healthcheck;
         }
 
-        location /client/inference/api/v1/ {
+        location /inference/api/v1/ {
              proxy_pass http://127.0.0.1:4000;
         }
 

service-config.yml

@@ -11,7 +11,7 @@ aiServices:
       name: 'openai-compat'
       baseUrl: http://127.0.0.1:7861/v1
     model: 'all-MiniLM-L6-v2'
-  headless:
+  inference:
     - provider:
         name: 'openai-compat'
         baseUrl: http://127.0.0.1:7861/v1

start-services.sh

@@ -25,61 +25,50 @@ if ! ps aux | grep nginx | grep -v grep > /dev/null; then
 fi
 echo "Nginx started successfully"
 
-# Start AI Assistant
-echo "Starting AI Assistant..."
-cd $HOME/app/aia
-PORT=4000 node app/main.bundle.js &
-AIA_PID=$!
-
 # Start the models
-echo "Starting chat model..."
+echo "Starting models..."
 TMPDIR=/tmp/llamafiler /usr/local/bin/llamafiler --model $HOME/models/gemma-2b.gguf -ngl 999 --listen 0.0.0.0:8082 &
 GEMMA_PID=$!
 
-echo "Starting embedding model..."
-TMPDIR=/tmp/llamafiler /usr/local/bin/llamafiler --model $HOME/models/embeddings.gguf  -ngl 999 --listen 0.0.0.0:8081 &
+TMPDIR=/tmp/llamafiler /usr/local/bin/llamafiler --model $HOME/models/embeddings.gguf -ngl 999 --listen 0.0.0.0:8081 &
 EMBEDDINGS_PID=$!
 
-# Wait for the models to be ready with a timeout
+# Wait for models to be ready
 echo "Waiting for models to be ready..."
-TIMEOUT=600  # 10 minutes timeout
 START_TIME=$SECONDS
+TIMEOUT=600  # 10 minutes
 
-wait_for_services() {
-    curl -s --fail -X POST http://127.0.0.1:7861/v1/chat/completions \
+wait_for_models() {
+    CHAT_TEST=$(curl -s -X POST http://127.0.0.1:8082/v1/chat/completions \
         -H "Content-Type: application/json" \
-        -d '{"model": "gemma-2b", "messages":[{"role":"user","content":"hi"}]}' >/dev/null 2>&1 && \
-    curl -s --fail -X POST http://127.0.0.1:7861/v1/embeddings \
+        -d '{"model":"gemma-2b","messages":[{"role":"user","content":"test"}]}')
+    
+    EMBED_TEST=$(curl -s -X POST http://127.0.0.1:8081/v1/embeddings \
         -H "Content-Type: application/json" \
-        -d '{"input":"test"}' >/dev/null 2>&1
+        -d '{"input":"test"}')
+    
+    [[ "$CHAT_TEST" == *"content"* ]] && [[ "$EMBED_TEST" == *"embedding"* ]]
 }
 
-until wait_for_services; do
+until wait_for_models; do
     ELAPSED=$((SECONDS - START_TIME))
     if [ $ELAPSED -gt $TIMEOUT ]; then
-        echo "Timeout waiting for services after ${TIMEOUT} seconds"
+        echo "Timeout after ${TIMEOUT} seconds"
         exit 1
     fi
-    
-    # Check if processes are still running
-    if ! kill -0 $GEMMA_PID 2>/dev/null; then
-        echo "Gemma model process died"
+    if ! kill -0 $GEMMA_PID 2>/dev/null || ! kill -0 $EMBEDDINGS_PID 2>/dev/null; then
+        echo "Model process died"
         exit 1
     fi
-    if ! kill -0 $EMBEDDINGS_PID 2>/dev/null; then
-        echo "Embeddings model process died"
-        exit 1
-    fi
-    if ! kill -0 $AIA_PID 2>/dev/null; then
-        echo "AI Assistant process died"
-        exit 1
-    fi
-    
-    echo "Waiting for services... (${ELAPSED}s elapsed, PIDs: Gemma=$GEMMA_PID, Embeddings=$EMBEDDINGS_PID, AIA=$AIA_PID)"
+    echo "Waiting for models... (${ELAPSED}s elapsed)"
     sleep 2
 done
 
-echo "All services are ready!"
+# Start AI Assistant
+echo "Models ready after ${ELAPSED}s. Starting AI Assistant..."
+cd $HOME/app/aia
+PORT=4000 node app/main.bundle.js &
+AIA_PID=$!
 
-# Keep the container running
+# Keep container running
 wait $GEMMA_PID