Detect Hanging Cron Job: Timeout & Execution Time Monitoring

The Problem: Scripts That Hang Without Failing

A cron job can "run" but never complete. The process is alive, consuming resources, but stuck on a network call, a locked file, or a dead database connection. Exit code is never returned. No ping, no alert—until you notice something is wrong.

• Script blocks on I/O (API timeout, DB lock, NFS hang)
• Script runs 5 hours instead of 5 minutes and nobody notices
• Next scheduled run is skipped because the previous one is still "running"
• Resource exhaustion: too many stuck processes

A simple "ping when done" only tells you the job finished. It doesn't tell you it took 6 hours. You need to measure time between start and stop.

Solution: Start/Stop Tracking and Timeout Detection

Send a start signal when the job begins and a stop signal (or payload ping) when it ends. The monitor measures elapsed time. If no stop arrives within your configured timeout, you get an alert—hanging job detected.

• Alert if the job doesn't complete within expected duration (e.g. 30 minutes)
• See actual execution time in the dashboard for every run
• Detect gradual slowdown (job used to take 2 min, now 45 min)

Bash: Start at Beginning, Ping at End

#!/bin/bash
# 1. Signal "job started" (get run_id from response if you need it)
curl -s -X POST "https://deadmanping.com/api/ping/backup-daily/start" -o /tmp/run.json

# 2. Do the actual work
/path/to/backup.sh

# 3. Signal "job finished" - optional: send run_id so monitor can match start/stop
curl -X POST "https://deadmanping.com/api/ping/backup-daily"
# If backup.sh hangs, step 3 never runs -> timeout alert

Python: Start/Stop with run_id

import requests
import subprocess

PING_SLUG = "data-sync-job"
BASE = "https://deadmanping.com/api/ping"

# Start tracking
r = requests.post(f"{BASE}/{PING_SLUG}/start")
run_id = r.json().get("run_id")  # optional, for linking start/stop

# Your long-running work
subprocess.run(["/path/to/sync.sh"], check=True)

# Stop: job completed (optionally pass run_id)
requests.post(f"{BASE}/{PING_SLUG}", json={"run_id": run_id})
# If sync hangs, this never runs -> you get a timeout alert

Node.js: Async Job with Start/Stop

const slug = 'report-generation';
const base = 'https://deadmanping.com/api/ping';

// Start
const startRes = await fetch(`${base}/${slug}/start`, { method: 'POST' });
const { run_id } = await startRes.json();

try {
  await runReportJob(); // your async work
  await fetch(`${base}/${slug}`, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ run_id })
  });
} catch (e) {
  // Optionally ping with failure status so monitor knows it didn't hang, it failed
  await fetch(`${base}/${slug}`, {
    method: 'POST',
    body: JSON.stringify({ run_id, status: 'failure' })
  });
  throw e;
}

Configure Timeout in DeadManPing

When you use start/stop tracking, set an expected max duration (e.g. 30 minutes). If no completion ping arrives within that window, DeadManPing marks the run as timed out and sends an alert—so you know the job is stuck, not just late.

You keep your script and cron. Add two HTTP calls: one at start, one at end. No SDK, no agent—just curl or your language's HTTP client.

Related Articles

Learn more about cron job monitoring and troubleshooting:

• Monitor Cron Jobs: One Curl Line, No Migration

  Monitor cron jobs with one curl line. No SDK, no migration—get alerts when jobs fail.

• Cron Job Failed? How to Detect & Fix in 5 Min

  Step-by-step guide to detect why your cron failed, fix it, and get alerts so it never happens again.

• Verify Cron Job Completed

  How to verify that cron jobs completed successfully.

• Cron Job Not Executing

  How to detect when cron jobs are not executing.