mirror of
https://github.com/jokob-sk/NetAlertX.git
synced 2025-12-07 09:36:05 -08:00
Final touches on devcontainer
This commit is contained in:
Adam Outler
@@ -1,9 +1,43 @@
|
||||
#!/bin/sh
|
||||
|
||||
# entrypoint.sh - Main container entrypoint script for NetAlertX
|
||||
|
||||
#make this red
|
||||
################################################################################
|
||||
# NetAlertX Container Entrypoint
|
||||
################################################################################
|
||||
#
|
||||
# Purpose: Main entrypoint script for NetAlertX Docker containers
|
||||
#
|
||||
# Responsibilities:
|
||||
# 1. Display NetAlertX banner and container startup info
|
||||
# 2. Run pre-startup health checks
|
||||
# 3. Initialize required directories and log files
|
||||
# 4. Start and monitor core services (crond, php-fpm, nginx, Python backend)
|
||||
# 5. Handle service failures and graceful shutdown
|
||||
# 6. Manage process signals (INT, TERM) for clean container termination
|
||||
#
|
||||
# Environment Variables:
|
||||
# - ENVIRONMENT: Container environment type (debian or alpine). If not "debian",
|
||||
# crond scheduler service will be started.
|
||||
# - NETALERTX_DEBUG: If set to 1, services won't auto-shutdown on failure;
|
||||
# container will wait for all to exit naturally (development mode).
|
||||
# - NETALERTX_PLUGINS_LOG: Directory path for plugin logs
|
||||
# - SYSTEM_SERVICES_RUN_LOG: Directory path for service runtime logs
|
||||
# - SYSTEM_SERVICES_RUN_TMP: Directory path for service temporary files
|
||||
# - LOG_DB_IS_LOCKED: File path for database lock status
|
||||
# - LOG_EXECUTION_QUEUE: File path for execution queue log
|
||||
#
|
||||
# Exit Codes:
|
||||
# - 0: Graceful shutdown (unlikely in production)
|
||||
# - 143: Caught signal (INT/TERM)
|
||||
# - Non-zero: Service failure status code
|
||||
#
|
||||
# Service Monitoring:
|
||||
# In production mode (NETALERTX_DEBUG != 1), if any service exits, all services
|
||||
# are terminated and the container exits with the failed service's status code.
|
||||
# This ensures container restart policies can properly reinitialize the stack.
|
||||
#
|
||||
################################################################################
|
||||
|
||||
# Banner display
|
||||
printf '
|
||||
\033[1;31m
|
||||
_ _ _ ___ _ _ __ __
|
||||
@@ -19,17 +53,31 @@ printf '
|
||||
|
||||
set -u
|
||||
|
||||
# Run all checks at container startup.
|
||||
for script in /services/check-*.sh; do
|
||||
# Run all pre-startup checks to validate container environment and dependencies
|
||||
for script in ${SYSTEM_SERVICES_SCRIPTS}/check-*.sh; do
|
||||
sh "$script"
|
||||
done
|
||||
|
||||
# Update vendor data (MAC address OUI database) in the background
|
||||
# This happens concurrently with service startup to avoid blocking container readiness
|
||||
${SYSTEM_SERVICES_SCRIPTS}/update_vendors.sh &
|
||||
|
||||
|
||||
SERVICES=""
|
||||
FAILED_NAME=""
|
||||
FAILED_STATUS=0
|
||||
|
||||
# Service management state variables
|
||||
SERVICES="" # Space-separated list of active services in format "pid:name"
|
||||
FAILED_NAME="" # Name of service that failed (used for error reporting)
|
||||
FAILED_STATUS=0 # Exit status code from failed service or signal
|
||||
|
||||
################################################################################
|
||||
# is_pid_active() - Check if a process is alive and not in zombie/dead state
|
||||
################################################################################
|
||||
# Arguments:
|
||||
# $1: Process ID to check
|
||||
# Returns:
|
||||
# 0 (success): Process is alive and healthy
|
||||
# 1 (failure): Process is dead, zombie, or PID is empty
|
||||
################################################################################
|
||||
is_pid_active() {
|
||||
pid="$1"
|
||||
[ -z "${pid}" ] && return 1
|
||||
@@ -51,6 +99,10 @@ is_pid_active() {
|
||||
}
|
||||
|
||||
add_service() {
|
||||
# Start a new service script and track it for monitoring
|
||||
# Arguments:
|
||||
# $1: Path to service startup script (e.g., /services/start-backend.sh)
|
||||
# $2: Human-readable service name (for logging and error reporting)
|
||||
script="$1"
|
||||
name="$2"
|
||||
"$script" &
|
||||
@@ -58,6 +110,13 @@ add_service() {
|
||||
SERVICES="${SERVICES} ${pid}:${name}"
|
||||
}
|
||||
|
||||
################################################################################
|
||||
# remove_service() - Remove a service from the active services list
|
||||
################################################################################
|
||||
# Arguments:
|
||||
# $1: Process ID to remove
|
||||
# Updates: SERVICES variable to exclude the specified PID
|
||||
################################################################################
|
||||
remove_service() {
|
||||
target_pid="$1"
|
||||
updated=""
|
||||
@@ -70,6 +129,16 @@ remove_service() {
|
||||
SERVICES="${updated}"
|
||||
}
|
||||
|
||||
################################################################################
|
||||
# shutdown_services() - Gracefully stop all active services
|
||||
################################################################################
|
||||
# Process:
|
||||
# 1. Send SIGTERM to all active services (time to clean up)
|
||||
# 2. Wait for all services to fully terminate
|
||||
# Notes:
|
||||
# - Tolerates services that are already dead
|
||||
# - Uses 'wait' to reap zombie processes
|
||||
################################################################################
|
||||
shutdown_services() {
|
||||
for entry in ${SERVICES}; do
|
||||
pid="${entry%%:*}"
|
||||
@@ -86,6 +155,15 @@ shutdown_services() {
|
||||
echo "All services stopped."
|
||||
}
|
||||
|
||||
################################################################################
|
||||
# handle_exit() - Terminate all services and exit container
|
||||
################################################################################
|
||||
# Process:
|
||||
# 1. Log failure information if a service exited abnormally
|
||||
# 2. Shut down all remaining services gracefully
|
||||
# 3. Exit container with recorded status code
|
||||
# Note: Used when a monitored service fails or signal is caught
|
||||
################################################################################
|
||||
handle_exit() {
|
||||
if [ -n "${FAILED_NAME}" ]; then
|
||||
echo "Service ${FAILED_NAME} exited with status ${FAILED_STATUS}."
|
||||
@@ -94,6 +172,15 @@ handle_exit() {
|
||||
exit "${FAILED_STATUS}"
|
||||
}
|
||||
|
||||
################################################################################
|
||||
# on_signal() - Handle container signals (INT, TERM) for graceful shutdown
|
||||
################################################################################
|
||||
# Signals handled: SIGINT (Ctrl+C), SIGTERM (docker stop)
|
||||
# Process:
|
||||
# 1. Set exit status to 143 (128 + 15, standard SIGTERM code)
|
||||
# 2. Trigger full shutdown sequence
|
||||
# Note: Registered via 'trap' command below
|
||||
################################################################################
|
||||
on_signal() {
|
||||
echo "Caught signal, shutting down services..."
|
||||
FAILED_NAME="signal"
|
||||
@@ -101,34 +188,53 @@ on_signal() {
|
||||
handle_exit
|
||||
}
|
||||
|
||||
/services/update_vendors.sh &
|
||||
|
||||
# Register signal handlers for graceful shutdown
|
||||
trap on_signal INT TERM
|
||||
|
||||
[ ! -d "${NETALERTX_PLUGINS_LOG}" ] && mkdir -p "${NETALERTX_PLUGINS_LOG}"
|
||||
[ ! -d "${SYSTEM_SERVICES_RUN_LOG}" ] && mkdir -p "${SYSTEM_SERVICES_RUN_LOG}"
|
||||
[ ! -d "${SYSTEM_SERVICES_RUN_TMP}" ] && mkdir -p "${SYSTEM_SERVICES_RUN_TMP}"
|
||||
[ ! -f "${LOG_DB_IS_LOCKED}" ] && touch "${LOG_DB_IS_LOCKED}"
|
||||
[ ! -f "${LOG_EXECUTION_QUEUE}" ] && touch "${LOG_EXECUTION_QUEUE}"
|
||||
|
||||
|
||||
################################################################################
|
||||
# Service Startup Section
|
||||
################################################################################
|
||||
# Start services based on environment configuration
|
||||
|
||||
# Only start crond scheduler on Alpine (non-Debian) environments
|
||||
# Debian typically uses systemd or other schedulers
|
||||
if [ "${ENVIRONMENT:-}" ] && [ "${ENVIRONMENT:-}" != "debian" ]; then
|
||||
add_service "/services/start-crond.sh" "crond"
|
||||
fi
|
||||
add_service "/services/start-php-fpm.sh" "php-fpm83"
|
||||
add_service "/services/start-nginx.sh" "nginx"
|
||||
add_service "/services/start-backend.sh" "python3"
|
||||
|
||||
# Start core frontend and backend services
|
||||
# Order: web server, application server, then Python backend
|
||||
add_service "${SYSTEM_SERVICES}/start-php-fpm.sh" "php-fpm83"
|
||||
add_service "${SYSTEM_SERVICES}/start-nginx.sh" "nginx"
|
||||
add_service "${SYSTEM_SERVICES}/start-backend.sh" "python3"
|
||||
|
||||
# if NETALERTX_DEBUG=1 then we will not kill any services if one fails. We will just wait for all to exit.
|
||||
################################################################################
|
||||
# Development Mode Debug Switch
|
||||
################################################################################
|
||||
# If NETALERTX_DEBUG=1, skip automatic service restart on failure
|
||||
# Useful for devcontainer debugging where individual services need to be debugged
|
||||
if [ "${NETALERTX_DEBUG:-0}" -eq 1 ]; then
|
||||
echo "NETALERTX_DEBUG is set to 1, will not shut down other services if one fails."
|
||||
wait
|
||||
exit $?
|
||||
fi
|
||||
|
||||
|
||||
# If any service fails, we will shut down all others and exit with the same status.
|
||||
# This improves reliability in production environments by reinitializing the entire stack if one service fails.
|
||||
################################################################################
|
||||
# Service Monitoring Loop (Production Mode)
|
||||
################################################################################
|
||||
# Behavior depends on NETALERTX_DEBUG setting:
|
||||
# - DEBUG OFF (production): Any service failure triggers full container restart
|
||||
# - DEBUG ON: Services can fail individually; container waits for natural exit
|
||||
#
|
||||
# Loop Process:
|
||||
# 1. Check each active service every 10 seconds
|
||||
# 2. If service is not active, wait for it and capture exit status
|
||||
# 3. Log failure and terminate all other services
|
||||
# 4. Exit container with failed service's status code
|
||||
# 5. This enables Docker restart policies to reinitialize the stack
|
||||
################################################################################
|
||||
while [ -n "${SERVICES}" ]; do
|
||||
for entry in ${SERVICES}; do
|
||||
pid="${entry%%:*}"
|
||||
@@ -147,6 +253,8 @@ while [ -n "${SERVICES}" ]; do
|
||||
sleep 10
|
||||
done
|
||||
|
||||
# If we exit the loop with no service failures, set status to 1 (error)
|
||||
# This should not happen in normal operation
|
||||
if [ "${FAILED_STATUS}" -eq 0 ] && [ "${FAILED_NAME}" != "signal" ]; then
|
||||
FAILED_STATUS=1
|
||||
fi
|
||||
Reference in New Issue
Block a user