Alerting

cronbase sends notifications when jobs complete — via webhooks (Slack, Discord, any URL) or email (built-in SMTP client). Alerts are configured per-job and fire on success, failure, or timeout events.

Quick setup

The simplest way to add alerting is in your config file:

jobs:
  - name: backup-db
    schedule: "0 2 * * *"
    command: pg_dump mydb > /backups/db.sql
    on_failure: https://hooks.slack.com/services/T.../B.../xxx
    on_failure_email: ops@example.com

Config file shortcuts

Webhook shortcuts

Property	Events triggered
on_failure	failed, timeout
on_success	success
on_complete	success, failed, timeout

Email shortcuts

Property	Events triggered
on_failure_email	failed, timeout
on_success_email	success
on_complete_email	success, failed, timeout

Email fields accept a single address or a comma-separated list (e.g. ops@example.com, oncall@example.com).

You can combine webhook and email shortcuts on the same job — both fire independently.

Platform support

cronbase auto-detects the webhook platform from the URL and sends appropriately formatted messages.

Slack

URLs containing hooks.slack.com or slack.com/api receive Block Kit formatted messages with:

• Color-coded attachments (green for success, red for failure)
• Structured fields: schedule, duration, exit code, attempt number
• stderr output block on failure

Discord

URLs containing discord.com/api/webhooks receive embed formatted messages with:

• Color-coded embeds
• Inline fields: schedule, duration, exit code
• stderr output on failure

Generic webhook

All other URLs receive the raw JSON payload:

{
  "event": "failed",
  "job": {
    "id": 1,
    "name": "backup-db",
    "schedule": "0 2 * * *",
    "command": "pg_dump mydb > /backups/db.sql"
  },
  "execution": {
    "id": 42,
    "status": "failed",
    "exitCode": 1,
    "durationMs": 5230,
    "startedAt": "2025-01-15T02:00:00.000Z",
    "finishedAt": "2025-01-15T02:00:05.230Z",
    "stdoutTail": "...",
    "stderrTail": "pg_dump: error: connection refused",
    "attempt": 2
  },
  "timestamp": "2025-01-15T02:00:05.235Z"
}

Email (SMTP)

cronbase includes a built-in SMTP client for email alerts — no external mail libraries or services required.

Setup

Set the SMTP environment variables before starting cronbase:

export CRONBASE_SMTP_HOST="smtp.gmail.com"
export CRONBASE_SMTP_PORT=465
export CRONBASE_SMTP_SECURE=true
export CRONBASE_SMTP_FROM="alerts@example.com"
export CRONBASE_SMTP_USERNAME="alerts@example.com"
export CRONBASE_SMTP_PASSWORD="app-password-here"

cronbase start

With Docker:

docker run -d --name cronbase \
  -p 7433:7433 \
  -e CRONBASE_SMTP_HOST="smtp.gmail.com" \
  -e CRONBASE_SMTP_PORT=465 \
  -e CRONBASE_SMTP_SECURE=true \
  -e CRONBASE_SMTP_FROM="alerts@example.com" \
  -e CRONBASE_SMTP_USERNAME="alerts@example.com" \
  -e CRONBASE_SMTP_PASSWORD="app-password-here" \
  -v cronbase-data:/data \
  ghcr.io/paperkite-hq/cronbase

Environment variables

Variable	Default	Description
CRONBASE_SMTP_HOST	(none)	SMTP server hostname (required to enable email alerts)
CRONBASE_SMTP_PORT	587	SMTP server port
CRONBASE_SMTP_SECURE	false	Set to true for TLS/SMTPS on connect (typically port 465)
CRONBASE_SMTP_FROM	cronbase@localhost	Sender address
CRONBASE_SMTP_USERNAME	(none)	SMTP AUTH username
CRONBASE_SMTP_PASSWORD	(none)	SMTP AUTH password

Email content

Alert emails include:

• Subject line with status icon: [cronbase] ✓ backup-db succeeded or [cronbase] ✗ backup-db failed
• Job name, schedule, start time, duration, exit code, and attempt number
• Last 500 characters of stderr (on failure) or stdout (on success)

Common SMTP providers

Provider	Host	Port	Secure	Notes
Gmail	smtp.gmail.com	465	true	Use App Passwords
Outlook/Office 365	smtp.office365.com	587	false	Standard auth or OAuth
Amazon SES	email-smtp.us-east-1.amazonaws.com	465	true	SMTP credentials from SES console
Postmark	smtp.postmarkapp.com	587	false	Server API token as password
Local relay	localhost	25	false	No auth needed for local relay

REST API configuration

You can also configure alerts via the REST API:

# Set webhook + email alerts for job ID 1
curl -X PUT http://localhost:7433/api/jobs/1/alerts \
  -H "Content-Type: application/json" \
  -d '{
    "webhooks": [
      {
        "url": "https://hooks.slack.com/services/T.../B.../xxx",
        "events": ["failed", "timeout"]
      }
    ],
    "emails": [
      {
        "to": ["ops@example.com", "oncall@example.com"],
        "events": ["failed", "timeout"]
      }
    ]
  }'

# View current alert config
curl http://localhost:7433/api/jobs/1/alerts

# Remove alerts
curl -X DELETE http://localhost:7433/api/jobs/1/alerts

Email alerts require SMTP environment variables to be set (see Email setup above). If CRONBASE_SMTP_HOST is not set, email alerts are silently skipped with a warning in the logs.

Behavior

• Alerts are sent asynchronously after job completion — they don't block the scheduler
• Each webhook has a 10-second timeout to prevent hanging
• Failed webhook deliveries are logged but don't affect job status
• The stdoutTail and stderrTail fields contain the last 500 characters of output