Getting Started

One curl command. Start simple and unlock more intelligence as you go.

Start with a heartbeat

Add one line to your job. When the ping doesn't arrive on time, CronDoctor knows something is wrong.

curl -s https://crondoctor.com/api/v1/ping/YOUR_ID

nightly-backupDOWN

Check-in missed. Expected at 03:00 UTC.

You get: missed and late job detection, grace periods, recovery alerts.

Add duration

Signal 1

Add one query parameter. Now CronDoctor learns what "normal" looks like for this job and catches performance drift — even when the job "succeeds."

curl -s "https://crondoctor.com/api/v1/ping/YOUR_ID?duration=$SECONDS"

nightly-backupSLOW

Ran in 247s — 4x your P95 baseline of 62s. Something is degrading.

You unlock: adaptive thresholds (P95 + 20% buffer by default, configurable on Pro), duration trend tracking, "slow" alerts.

Add error context

Signal 2

When your job fails, POST the exit code and stderr. CronDoctor's AI analyzes the failure, tells you what likely went wrong, and suggests a fix to investigate.

curl -X POST "https://crondoctor.com/api/v1/ping/YOUR_ID/fail" \
  -H "Content-Type: application/json" \
  -d '{ "exit_code": 1,
        "stderr": "$(tail -100 /tmp/err.log)" }'

nightly-backupCRITICALSLOWAI diagnosed

Failed after 247s — 4x your P95 baseline of 62s.

Disk full on /var/data — PostgreSQL could not write WAL segments. The backup stalled waiting for disk space, then timed out.

Suggested fix: df -h /var/data && sudo journalctl --vacuum-size=500M

87% confidence

You unlock: AI root cause diagnosis, suggested fix commands, confidence scoring.

Full integration examples

Copy-paste examples in your language. Signal 1 (heartbeat with duration) and Signal 2 (error context on failure) — use one or both.

Signal 1Heartbeat with duration

# Bash

SECONDS=0
/path/to/backup.sh
curl -s "https://crondoctor.com/api/v1/ping/YOUR_ID\\
  ?duration=$SECONDS"

# Python

import time, requests
start = time.time()
run_job()
requests.get(
  f"https://crondoctor.com/api/v1/ping/{monitor_id}",
  params={"duration": time.time() - start}
)

# Node.js

const start = Date.now();
await runJob();
await fetch(
  `https://crondoctor.com/api/v1/ping/${monitorId}`
  + `?duration=${(Date.now() - start) / 1000}`
);

Signal 2Error context on failure

# Bash

/path/to/backup.sh 2>/tmp/err.log
EXIT_CODE=$?
if [ $EXIT_CODE -ne 0 ]; then
  curl -X POST \\
    "https://crondoctor.com/api/v1/ping/YOUR_ID/fail" \\
    -H "Content-Type: application/json" \\
    -d "{
      \"exit_code\": $EXIT_CODE,
      \"stderr\": \"$(cat /tmp/err.log)\"
    }"
fi

# Python

try:
    run_job()
except Exception as e:
    requests.post(
      f"https://crondoctor.com/api/v1/ping/{monitor_id}/fail",
      json={
        "exit_code": 1,
        "stderr": str(e)
      }
    )

# Node.js

try {
  await runJob();
} catch (err) {
  await fetch(
    `https://crondoctor.com/api/v1/ping/${monitorId}/fail`,
    {
      method: "POST",
      headers: {
        "Content-Type": "application/json"
      },
      body: JSON.stringify({
        exit_code: 1,
        stderr: err.message
      })
    }
  );
}

See the difference yourself

Set up in under a minute. Start with a bare ping and add more when you're ready.

or create a free account →

What each data field unlocks

ping aloneStandard monitoring — missed and late job detection
+ durationAdaptive thresholds — CronDoctor learns your job's normal runtime and flags performance anomalies automatically
+ exit_codeFailure detection — non-zero exit codes trigger alerts, signal codes (like 137 = OOM killed) provide context
+ stderrAI root cause diagnosis — error output lets the AI pinpoint the exact problem and suggest a fix
+ stdoutAdditional context for the AI when stderr alone isn't enough

Start with a simple ping. Add more data as you need deeper insights. All capabilities are included on every plan.

Learn how adaptive thresholds work →

Set up in 60 seconds. Free forever for up to 5 jobs.

or create a free account →

See plans and pricing →