Spaces:
Sleeping
Sleeping
Add DocAuthoring and Nginx
Browse files- Dockerfile +21 -1
- nginx.conf +1 -1
- start-services.sh +18 -3
Dockerfile
CHANGED
|
@@ -14,14 +14,26 @@ RUN apt-get update && apt-get install -y \
|
|
| 14 |
unzip \
|
| 15 |
clang \
|
| 16 |
cuda-toolkit \
|
|
|
|
| 17 |
&& rm -rf /var/lib/apt/lists/*
|
| 18 |
|
| 19 |
# Create directories and set permissions
|
| 20 |
RUN mkdir -p /tmp/llamafile && \
|
| 21 |
mkdir -p $HOME/models && \
|
| 22 |
mkdir -p $HOME/app && \
|
|
|
|
| 23 |
mkdir -p $HOME/.llamafile && \
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
# Install llamafiler
|
| 27 |
RUN cd /tmp/llamafile && \
|
|
@@ -31,6 +43,9 @@ RUN cd /tmp/llamafile && \
|
|
| 31 |
chmod +x /usr/local/bin/llamafiler && \
|
| 32 |
rm -rf /tmp/llamafile
|
| 33 |
|
|
|
|
|
|
|
|
|
|
| 34 |
# Switch to non-root user
|
| 35 |
USER user
|
| 36 |
WORKDIR $HOME/app
|
|
@@ -39,6 +54,11 @@ WORKDIR $HOME/app
|
|
| 39 |
RUN wget -q https://huggingface.co/bartowski/gemma-2-2b-it-GGUF/resolve/main/gemma-2-2b-it-Q8_0.gguf -O $HOME/models/gemma-2b.gguf && \
|
| 40 |
wget -q https://huggingface.co/leliuga/all-MiniLM-L6-v2-GGUF/resolve/main/all-MiniLM-L6-v2.F16.gguf -O $HOME/models/embeddings.gguf
|
| 41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
# Copy start script
|
| 43 |
COPY --chown=user:user start-services.sh $HOME/app/
|
| 44 |
RUN chmod +x $HOME/app/start-services.sh
|
|
|
|
| 14 |
unzip \
|
| 15 |
clang \
|
| 16 |
cuda-toolkit \
|
| 17 |
+
nginx \
|
| 18 |
&& rm -rf /var/lib/apt/lists/*
|
| 19 |
|
| 20 |
# Create directories and set permissions
|
| 21 |
RUN mkdir -p /tmp/llamafile && \
|
| 22 |
mkdir -p $HOME/models && \
|
| 23 |
mkdir -p $HOME/app && \
|
| 24 |
+
mkdir -p $HOME/app/docauth && \
|
| 25 |
mkdir -p $HOME/.llamafile && \
|
| 26 |
+
mkdir -p /var/cache/nginx && \
|
| 27 |
+
mkdir -p /var/log/nginx && \
|
| 28 |
+
mkdir -p /var/lib/nginx && \
|
| 29 |
+
mkdir -p /run/nginx && \
|
| 30 |
+
touch /var/run/nginx.pid && \
|
| 31 |
+
chown -R user:user $HOME && \
|
| 32 |
+
chown -R user:user /var/cache/nginx && \
|
| 33 |
+
chown -R user:user /var/log/nginx && \
|
| 34 |
+
chown -R user:user /var/lib/nginx && \
|
| 35 |
+
chown -R user:user /var/run/nginx.pid && \
|
| 36 |
+
chown -R user:user /run/nginx
|
| 37 |
|
| 38 |
# Install llamafiler
|
| 39 |
RUN cd /tmp/llamafile && \
|
|
|
|
| 43 |
chmod +x /usr/local/bin/llamafiler && \
|
| 44 |
rm -rf /tmp/llamafile
|
| 45 |
|
| 46 |
+
# Copy nginx configuration
|
| 47 |
+
COPY nginx.conf /etc/nginx/nginx.conf
|
| 48 |
+
|
| 49 |
# Switch to non-root user
|
| 50 |
USER user
|
| 51 |
WORKDIR $HOME/app
|
|
|
|
| 54 |
RUN wget -q https://huggingface.co/bartowski/gemma-2-2b-it-GGUF/resolve/main/gemma-2-2b-it-Q8_0.gguf -O $HOME/models/gemma-2b.gguf && \
|
| 55 |
wget -q https://huggingface.co/leliuga/all-MiniLM-L6-v2-GGUF/resolve/main/all-MiniLM-L6-v2.F16.gguf -O $HOME/models/embeddings.gguf
|
| 56 |
|
| 57 |
+
# Copy web files
|
| 58 |
+
COPY --chown=user:user index.html $HOME/app/docauth/
|
| 59 |
+
COPY --chown=user:user document-authoring.js $HOME/app/docauth/
|
| 60 |
+
COPY --chown=user:user Sample.docx $HOME/app/docauth/
|
| 61 |
+
|
| 62 |
# Copy start script
|
| 63 |
COPY --chown=user:user start-services.sh $HOME/app/
|
| 64 |
RUN chmod +x $HOME/app/start-services.sh
|
nginx.conf
CHANGED
|
@@ -8,7 +8,7 @@ http {
|
|
| 8 |
listen 7861;
|
| 9 |
|
| 10 |
location / {
|
| 11 |
-
root /docauth;
|
| 12 |
}
|
| 13 |
|
| 14 |
location /healthcheck {
|
|
|
|
| 8 |
listen 7861;
|
| 9 |
|
| 10 |
location / {
|
| 11 |
+
root /home/user/app/docauth;
|
| 12 |
}
|
| 13 |
|
| 14 |
location /healthcheck {
|
start-services.sh
CHANGED
|
@@ -10,9 +10,24 @@ nvidia-smi --query-gpu=compute_cap --format=csv,noheader || echo "Warning: Could
|
|
| 10 |
# Create temporary directory for llamafiler
|
| 11 |
mkdir -p /tmp/llamafiler
|
| 12 |
|
| 13 |
-
# Start
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
echo "Starting chat model..."
|
| 15 |
-
TMPDIR=/tmp/llamafiler /usr/local/bin/llamafiler --model $HOME/models/gemma-2b.gguf -ngl 2 --listen 0.0.0.0:
|
| 16 |
GEMMA_PID=$!
|
| 17 |
|
| 18 |
# Wait for the model to be ready with a timeout
|
|
@@ -21,7 +36,7 @@ TIMEOUT=600 # 10 minutes timeout
|
|
| 21 |
START_TIME=$SECONDS
|
| 22 |
|
| 23 |
wait_for_services() {
|
| 24 |
-
curl -s --fail -X POST http://127.0.0.1:
|
| 25 |
-H "Content-Type: application/json" \
|
| 26 |
-d '{"model": "gemma-2b", "messages":[{"role":"user","content":"hi"}]}' >/dev/null 2>&1
|
| 27 |
}
|
|
|
|
| 10 |
# Create temporary directory for llamafiler
|
| 11 |
mkdir -p /tmp/llamafiler
|
| 12 |
|
| 13 |
+
# Start nginx
|
| 14 |
+
echo "Starting nginx..."
|
| 15 |
+
/usr/sbin/nginx -c /etc/nginx/nginx.conf
|
| 16 |
+
if [ $? -ne 0 ]; then
|
| 17 |
+
echo "Failed to start nginx"
|
| 18 |
+
exit 1
|
| 19 |
+
fi
|
| 20 |
+
|
| 21 |
+
# Verify nginx is running
|
| 22 |
+
if ! ps aux | grep nginx | grep -v grep > /dev/null; then
|
| 23 |
+
echo "Nginx failed to start"
|
| 24 |
+
exit 1
|
| 25 |
+
fi
|
| 26 |
+
echo "Nginx started successfully"
|
| 27 |
+
|
| 28 |
+
# Start the model in server mode (now on port 8081)
|
| 29 |
echo "Starting chat model..."
|
| 30 |
+
TMPDIR=/tmp/llamafiler /usr/local/bin/llamafiler --model $HOME/models/gemma-2b.gguf -ngl 2 --listen 0.0.0.0:8081 --verbose &
|
| 31 |
GEMMA_PID=$!
|
| 32 |
|
| 33 |
# Wait for the model to be ready with a timeout
|
|
|
|
| 36 |
START_TIME=$SECONDS
|
| 37 |
|
| 38 |
wait_for_services() {
|
| 39 |
+
curl -s --fail -X POST http://127.0.0.1:8081/v1/chat/completions \
|
| 40 |
-H "Content-Type: application/json" \
|
| 41 |
-d '{"model": "gemma-2b", "messages":[{"role":"user","content":"hi"}]}' >/dev/null 2>&1
|
| 42 |
}
|