healthcheck checks for cpu overload

compose.yml

@@ -57,16 +57,16 @@ services:
     depends_on:
       - db
     healthcheck:
-      test: node /build/healthcheck.js # no curl for healthcheck in wekan container
+      test: bash /build/healthcheck.sh
       interval: 30s
       timeout: 10s
       retries: 10
-      start_period: 1m
+      start_period: 3m
     secrets:
       - oauth2_secret
     configs:
-      - source: healthcheck_js
-        target: /build/healthcheck.js
+      - source: healthcheck_sh
+        target: /build/healthcheck.sh
         mode: 0555
       - source: entrypoint
         target: /home/wekan/app/docker-entrypoint.sh
@@ -95,9 +95,9 @@ networks:
   internal:
 
 configs:
-  healthcheck_js:
-    name: ${STACK_NAME}_healthcheck_js
-    file: healthcheck.js
+  healthcheck_sh:
+    name: ${STACK_NAME}_healthcheck_sh
+    file: healthcheck.sh
   entrypoint:
     name: ${STACK_NAME}_entrypoint_${ENTRYPOINT_VERSION}
     file: entrypoint.sh.tmpl

healthcheck.js

@@ -1,23 +0,0 @@
-const http = require('http');
-const options = {
-    host: '0.0.0.0',
-    port: 8080,
-    timeout: 2000
-};
-
-const healthCheck = http.request(options, (res) => {
-    console.log(`HEALTHCHECK STATUS: ${res.statusCode}`);
-    if (res.statusCode == 200) {
-        process.exit(0);
-    }
-    else {
-        process.exit(1);
-    }
-});
-
-healthCheck.on('error', function (err) {
-    console.error('ERROR');
-    process.exit(1);
-});
-
-healthCheck.end();

healthcheck.sh

@@ -0,0 +1,87 @@
+#!/bin/sh
+
+printf "GET / HTTP/1.1\n\n" > /dev/tcp/127.0.0.1/8080
+if [ $? != 0 ]; then
+    echo "Webservice not accessible"
+    exit 1
+fi
+
+CPU_USAGE_THRESHOLD=100.0
+
+PID=$(pidof node)
+if [ -z "$PID" ]; then
+    echo "Node.js process not running."
+    exit 1
+fi
+STATE_FILE="/tmp/cpu_usage_state_$PID"
+
+# Function to get total CPU time
+get_total_cpu_time() {
+    cat /proc/stat | grep '^cpu ' | awk '{sum=0; for(i=2;i<=NF;i++) sum += $i; print sum;}'
+}
+
+# Function to get process CPU time
+get_process_cpu_time() {
+    cat /proc/$PID/stat | awk '{print $14+$15}'  # Only user and system time
+}
+
+# Function to get the number of CPUs
+get_cpu_count() {
+    grep -c '^processor' /proc/cpuinfo
+}
+
+# Main script
+if [ -z "$PID" ]; then
+    echo "Usage: $0 <pid>"
+    exit 1
+fi
+
+# Check if the process directory exists
+if [ ! -d "/proc/$PID" ]; then
+    echo "Process with PID $PID does not exist."
+    exit 1
+fi
+
+# Get current CPU times
+current_total=$(get_total_cpu_time)
+current_process=$(get_process_cpu_time $PID)
+
+# Check for previous state
+if [ -f "$STATE_FILE" ]; then
+    # Read previous state
+    read prev_total prev_process < "$STATE_FILE"
+    
+    # Calculate the CPU usage
+    delta_total=$((current_total - prev_total))
+    delta_process=$((current_process - prev_process))
+    CPU_COUNT=$(get_cpu_count)
+
+    # Normalize CPU usage across all CPUs and calculate percentage
+    if [ $delta_total -gt 0 ]; then  # To avoid division by zero
+        CPU_USAGE=$(awk -v delta_process="$delta_process" -v delta_total="$delta_total" -v cpu_count="$CPU_COUNT" \
+            'BEGIN {printf "%.2f", (delta_process / delta_total) * 100 * cpu_count}')
+        awk -v usage="$CPU_USAGE" -v threshold="$CPU_USAGE_THRESHOLD" '
+            BEGIN {
+                if (usage > threshold) {
+                    print "High CPU usage: " usage "% (Threshold: " threshold "%)";
+                    exit 1;  # Exit with status 1 for high usage
+                } else {
+                    print "CPU usage within limits: " usage "%";
+                    exit 0;  # Normal exit status
+                }
+            }'
+        EXIT_STATUS=$?
+    else
+        echo "No enough data to calculate CPU usage. Please try again."
+    fi
+else
+    echo "No previous data found. Run the script later to get CPU usage."
+fi
+
+# Save current state
+echo "$current_total $current_process" > "$STATE_FILE"
+
+if [ $EXIT_STATUS -ne 0 ]; then
+    echo "Exiting due to high CPU usage."
+    exit $EXIT_STATUS
+fi