c88df79246
- Fix uv pip install syntax to use --python flag instead of incorrect venv activation - Add proper CPU-only PyTorch installation in main and mineru environments - Update entrypoint scripts to check for pre-installed packages first - Ensure proper fallback to runtime installation when needed The previous commit only included documentation files, this commit adds the actual implementation.
317 lines
9.9 KiB
Bash
Executable file
317 lines
9.9 KiB
Bash
Executable file
#!/usr/bin/env bash
|
|
|
|
set -e
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Usage and command-line argument parsing
|
|
# -----------------------------------------------------------------------------
|
|
function usage() {
|
|
echo "Usage: $0 [--disable-webserver] [--disable-taskexecutor] [--disable-datasync] [--consumer-no-beg=<num>] [--consumer-no-end=<num>] [--workers=<num>] [--host-id=<string>]"
|
|
echo
|
|
echo " --disable-webserver Disables the web server (nginx + ragflow_server)."
|
|
echo " --disable-taskexecutor Disables task executor workers."
|
|
echo " --disable-datasync Disables synchronization of datasource workers."
|
|
echo " --enable-mcpserver Enables the MCP server."
|
|
echo " --enable-adminserver Enables the Admin server."
|
|
echo " --consumer-no-beg=<num> Start range for consumers (if using range-based)."
|
|
echo " --consumer-no-end=<num> End range for consumers (if using range-based)."
|
|
echo " --workers=<num> Number of task executors to run (if range is not used)."
|
|
echo " --host-id=<string> Unique ID for the host (defaults to \`hostname\`)."
|
|
echo
|
|
echo "Examples:"
|
|
echo " $0 --disable-taskexecutor"
|
|
echo " $0 --disable-webserver --consumer-no-beg=0 --consumer-no-end=5"
|
|
echo " $0 --disable-webserver --workers=2 --host-id=myhost123"
|
|
echo " $0 --enable-mcpserver"
|
|
echo " $0 --enable-adminserver"
|
|
exit 1
|
|
}
|
|
|
|
ENABLE_WEBSERVER=1 # Default to enable web server
|
|
ENABLE_TASKEXECUTOR=1 # Default to enable task executor
|
|
ENABLE_DATASYNC=1
|
|
ENABLE_MCP_SERVER=0
|
|
ENABLE_ADMIN_SERVER=0 # Default close admin server
|
|
CONSUMER_NO_BEG=0
|
|
CONSUMER_NO_END=0
|
|
WORKERS=1
|
|
|
|
MCP_HOST="127.0.0.1"
|
|
MCP_PORT=9382
|
|
MCP_BASE_URL="http://127.0.0.1:9380"
|
|
MCP_SCRIPT_PATH="/ragflow/mcp/server/server.py"
|
|
MCP_MODE="self-host"
|
|
MCP_HOST_API_KEY=""
|
|
MCP_TRANSPORT_SSE_FLAG="--transport-sse-enabled"
|
|
MCP_TRANSPORT_STREAMABLE_HTTP_FLAG="--transport-streamable-http-enabled"
|
|
MCP_JSON_RESPONSE_FLAG="--json-response"
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Host ID logic:
|
|
# 1. By default, use the system hostname if length <= 32
|
|
# 2. Otherwise, use the full MD5 hash of the hostname (32 hex chars)
|
|
# -----------------------------------------------------------------------------
|
|
CURRENT_HOSTNAME="$(hostname)"
|
|
if [ ${#CURRENT_HOSTNAME} -le 32 ]; then
|
|
DEFAULT_HOST_ID="$CURRENT_HOSTNAME"
|
|
else
|
|
DEFAULT_HOST_ID="$(echo -n "$CURRENT_HOSTNAME" | md5sum | cut -d ' ' -f 1)"
|
|
fi
|
|
|
|
HOST_ID="$DEFAULT_HOST_ID"
|
|
|
|
# Parse arguments
|
|
for arg in "$@"; do
|
|
case $arg in
|
|
--disable-webserver)
|
|
ENABLE_WEBSERVER=0
|
|
shift
|
|
;;
|
|
--disable-taskexecutor)
|
|
ENABLE_TASKEXECUTOR=0
|
|
shift
|
|
;;
|
|
--disable-datasyn)
|
|
ENABLE_DATASYNC=0
|
|
shift
|
|
;;
|
|
--enable-mcpserver)
|
|
ENABLE_MCP_SERVER=1
|
|
shift
|
|
;;
|
|
--enable-adminserver)
|
|
ENABLE_ADMIN_SERVER=1
|
|
shift
|
|
;;
|
|
--mcp-host=*)
|
|
MCP_HOST="${arg#*=}"
|
|
shift
|
|
;;
|
|
--mcp-port=*)
|
|
MCP_PORT="${arg#*=}"
|
|
shift
|
|
;;
|
|
--mcp-base-url=*)
|
|
MCP_BASE_URL="${arg#*=}"
|
|
shift
|
|
;;
|
|
--mcp-mode=*)
|
|
MCP_MODE="${arg#*=}"
|
|
shift
|
|
;;
|
|
--mcp-host-api-key=*)
|
|
MCP_HOST_API_KEY="${arg#*=}"
|
|
shift
|
|
;;
|
|
--mcp-script-path=*)
|
|
MCP_SCRIPT_PATH="${arg#*=}"
|
|
shift
|
|
;;
|
|
--no-transport-sse-enabled)
|
|
MCP_TRANSPORT_SSE_FLAG="--no-transport-sse-enabled"
|
|
shift
|
|
;;
|
|
--no-transport-streamable-http-enabled)
|
|
MCP_TRANSPORT_STREAMABLE_HTTP_FLAG="--no-transport-streamable-http-enabled"
|
|
shift
|
|
;;
|
|
--no-json-response)
|
|
MCP_JSON_RESPONSE_FLAG="--no-json-response"
|
|
shift
|
|
;;
|
|
--consumer-no-beg=*)
|
|
CONSUMER_NO_BEG="${arg#*=}"
|
|
shift
|
|
;;
|
|
--consumer-no-end=*)
|
|
CONSUMER_NO_END="${arg#*=}"
|
|
shift
|
|
;;
|
|
--workers=*)
|
|
WORKERS="${arg#*=}"
|
|
shift
|
|
;;
|
|
--host-id=*)
|
|
HOST_ID="${arg#*=}"
|
|
shift
|
|
;;
|
|
*)
|
|
usage
|
|
;;
|
|
esac
|
|
done
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Replace env variables in the service_conf.yaml file
|
|
# -----------------------------------------------------------------------------
|
|
CONF_DIR="/ragflow/conf"
|
|
TEMPLATE_FILE="${CONF_DIR}/service_conf.yaml.template"
|
|
CONF_FILE="${CONF_DIR}/service_conf.yaml"
|
|
|
|
rm -f "${CONF_FILE}"
|
|
while IFS= read -r line || [[ -n "$line" ]]; do
|
|
eval "echo \"$line\"" >> "${CONF_FILE}"
|
|
done < "${TEMPLATE_FILE}"
|
|
|
|
export LD_LIBRARY_PATH="/usr/lib/x86_64-linux-gnu/"
|
|
PY=python3
|
|
|
|
# -----------------------------------------------------------------------------
|
|
# Function(s)
|
|
# -----------------------------------------------------------------------------
|
|
|
|
function task_exe() {
|
|
local consumer_id="$1"
|
|
local host_id="$2"
|
|
|
|
JEMALLOC_PATH="$(pkg-config --variable=libdir jemalloc)/libjemalloc.so"
|
|
while true; do
|
|
LD_PRELOAD="$JEMALLOC_PATH" \
|
|
"$PY" rag/svr/task_executor.py "${host_id}_${consumer_id}" &
|
|
wait;
|
|
sleep 1;
|
|
done
|
|
}
|
|
|
|
function start_mcp_server() {
|
|
echo "Starting MCP Server on ${MCP_HOST}:${MCP_PORT} with base URL ${MCP_BASE_URL}..."
|
|
"$PY" "${MCP_SCRIPT_PATH}" \
|
|
--host="${MCP_HOST}" \
|
|
--port="${MCP_PORT}" \
|
|
--base-url="${MCP_BASE_URL}" \
|
|
--mode="${MCP_MODE}" \
|
|
--api-key="${MCP_HOST_API_KEY}" \
|
|
"${MCP_TRANSPORT_SSE_FLAG}" \
|
|
"${MCP_TRANSPORT_STREAMABLE_HTTP_FLAG}" \
|
|
"${MCP_JSON_RESPONSE_FLAG}" &
|
|
}
|
|
|
|
function ensure_docling() {
|
|
[[ "${USE_DOCLING}" == "true" ]] || { echo "[docling] disabled by USE_DOCLING"; return 0; }
|
|
|
|
# Check if docling is already available in the virtual environment
|
|
if python3 -c "import importlib.util,sys; sys.exit(0 if importlib.util.find_spec('docling') else 1)" 2>/dev/null; then
|
|
echo "[docling] found in virtual environment"
|
|
return 0
|
|
fi
|
|
|
|
# Fallback to runtime installation if not found (shouldn't happen with optimized Dockerfile)
|
|
echo "[docling] not found, installing at runtime..."
|
|
python3 -c 'import pip' >/dev/null 2>&1 || python3 -m ensurepip --upgrade || true
|
|
DOCLING_PIN="${DOCLING_VERSION:-==2.58.0}"
|
|
python3 -m pip install -i https://pypi.tuna.tsinghua.edu.cn/simple --extra-index-url https://pypi.org/simple --no-cache-dir "docling${DOCLING_PIN}"
|
|
}
|
|
|
|
function ensure_mineru() {
|
|
[[ "${USE_MINERU}" == "true" ]] || { echo "[mineru] disabled by USE_MINERU"; return 0; }
|
|
|
|
export HUGGINGFACE_HUB_ENDPOINT="${HF_ENDPOINT:-https://hf-mirror.com}"
|
|
|
|
local default_prefix="/ragflow/uv_tools"
|
|
local venv_dir="${default_prefix}/.venv"
|
|
local exe="${MINERU_EXECUTABLE:-${venv_dir}/bin/mineru}"
|
|
|
|
# Check if the pre-installed mineru is available
|
|
if [[ -x "${exe}" ]]; then
|
|
echo "[mineru] found pre-installed: ${exe}"
|
|
export MINERU_EXECUTABLE="${exe}"
|
|
|
|
# Verify it works
|
|
if "${MINERU_EXECUTABLE}" --help >/dev/null 2>&1; then
|
|
echo "[mineru] pre-installed version is working"
|
|
return 0
|
|
else
|
|
echo "[mineru] pre-installed version not working, will reinstall"
|
|
fi
|
|
fi
|
|
|
|
# Check if mineru was excluded during build
|
|
if [[ ! -d "${venv_dir}" ]]; then
|
|
echo "[mineru] not included in build (BUILD_MINERU=0), installing at runtime..."
|
|
else
|
|
echo "[mineru] not found or not working, bootstrapping with uv ..."
|
|
fi
|
|
|
|
(
|
|
set -e
|
|
mkdir -p "${default_prefix}"
|
|
cd "${default_prefix}"
|
|
[[ -d "${venv_dir}" ]] || uv venv "${venv_dir}"
|
|
|
|
# Install CPU-only PyTorch first to avoid CUDA dependencies
|
|
echo "[mineru] installing CPU-only PyTorch to avoid CUDA packages..."
|
|
uv pip install --python "${venv_dir}/bin/python" torch torchvision --index-url https://download.pytorch.org/whl/cpu
|
|
|
|
# Then install mineru
|
|
uv pip install --python "${venv_dir}/bin/python" -U "mineru[core]" -i https://mirrors.aliyun.com/pypi/simple --extra-index-url https://pypi.org/simple
|
|
)
|
|
export MINERU_EXECUTABLE="${exe}"
|
|
if ! "${MINERU_EXECUTABLE}" --help >/dev/null 2>&1; then
|
|
echo "[mineru] installation failed: ${MINERU_EXECUTABLE} not working" >&2
|
|
return 1
|
|
fi
|
|
echo "[mineru] installed: ${MINERU_EXECUTABLE}"
|
|
echo "[mineru] installation failed: ${MINERU_EXECUTABLE} not working" >&2
|
|
return 1
|
|
fi
|
|
echo "[mineru] installed: ${MINERU_EXECUTABLE}"
|
|
}
|
|
# -----------------------------------------------------------------------------
|
|
# Start components based on flags
|
|
# -----------------------------------------------------------------------------
|
|
ensure_docling
|
|
ensure_mineru
|
|
|
|
if [[ "${ENABLE_WEBSERVER}" -eq 1 ]]; then
|
|
echo "Starting nginx..."
|
|
/usr/sbin/nginx
|
|
|
|
echo "Starting ragflow_server..."
|
|
while true; do
|
|
"$PY" api/ragflow_server.py &
|
|
wait;
|
|
sleep 1;
|
|
done &
|
|
fi
|
|
|
|
if [[ "${ENABLE_DATASYNC}" -eq 1 ]]; then
|
|
echo "Starting data sync..."
|
|
while true; do
|
|
"$PY" rag/svr/sync_data_source.py &
|
|
wait;
|
|
sleep 1;
|
|
done &
|
|
fi
|
|
|
|
if [[ "${ENABLE_ADMIN_SERVER}" -eq 1 ]]; then
|
|
echo "Starting admin_server..."
|
|
while true; do
|
|
"$PY" admin/server/admin_server.py &
|
|
wait;
|
|
sleep 1;
|
|
done &
|
|
fi
|
|
|
|
if [[ "${ENABLE_MCP_SERVER}" -eq 1 ]]; then
|
|
start_mcp_server
|
|
fi
|
|
|
|
|
|
if [[ "${ENABLE_TASKEXECUTOR}" -eq 1 ]]; then
|
|
if [[ "${CONSUMER_NO_END}" -gt "${CONSUMER_NO_BEG}" ]]; then
|
|
echo "Starting task executors on host '${HOST_ID}' for IDs in [${CONSUMER_NO_BEG}, ${CONSUMER_NO_END})..."
|
|
for (( i=CONSUMER_NO_BEG; i<CONSUMER_NO_END; i++ ))
|
|
do
|
|
task_exe "${i}" "${HOST_ID}" &
|
|
done
|
|
else
|
|
# Otherwise, start a fixed number of workers
|
|
echo "Starting ${WORKERS} task executor(s) on host '${HOST_ID}'..."
|
|
for (( i=0; i<WORKERS; i++ ))
|
|
do
|
|
task_exe "${i}" "${HOST_ID}" &
|
|
done
|
|
fi
|
|
fi
|
|
|
|
wait
|