Building a Task Notification System for maintenance.queenofsandiego.com: Real-Time Alerts and Persistence

What Was Done

We implemented a comprehensive task notification and persistence system for the JADA maintenance tracking tool. When crew members add new maintenance tasks, the system now:

• Persists task data durably to Google Cloud Storage via Lambda
• Routes notifications intelligently based on task criticality
• Sends real-time alerts to assigned crew members (Sergio) and admins
• Maintains a calendar integration for scheduled maintenance work
• Operates in staging environment with email delivery to jadasailing@gmail.com

The implementation spans three primary components: Google Apps Script handlers, Lambda persistence layer, and frontend task notification UI.

Technical Details

Google Apps Script Architecture

We created two new GAS files in the /Users/cb/Documents/repos/sites/queenofsandiego.com/ directory:

• MaintenancePersistence.gs – Handles task data persistence and validation
• MaintenanceCalendar.gs – Manages calendar integration for scheduled maintenance

Both files are deployed to the same Apps Script project as BookingAutomation.gs, which routes incoming maintenance requests. We added the following route handler to BookingAutomation.gs in the doPost function:

case 'log_maintenance':
  return MaintenancePersistence.handleTaskSubmission(params);

This routing pattern maintains consistency with existing action handlers in the codebase and ensures maintenance requests flow through the same authentication and validation pipeline.

Persistence Strategy: Lambda + S3

We use AWS Lambda as the durable backend for task persistence. The Lambda function receives POST requests from GAS and stores task data in S3 with the following structure:

s3://jada-maintenance-tasks/
├── tasks/
│   ├── YYYY-MM-DD/
│   │   ├── task-{uuid}.json
│   └── active/
│       ├── task-{uuid}.json
└── logs/
    └── persistence-{date}.log

This dual-path storage allows quick queries of "active" tasks while maintaining full historical audit trails by date. Each task JSON includes:

• Task ID (UUID)
• Title and description
• Criticality level (critical, high, medium, low)
• Assigned crew member
• Created timestamp and creator
• Status (open, in_progress, completed)

Notification Logic

The MaintenancePersistence.gs file contains a notifyOnNewTask() function that implements criticality-based notification rules:

CRITICAL tasks → Immediate Slack + Email to Sergio + SMS fallback
HIGH tasks → Immediate Email + 2-hour digest consideration
MEDIUM tasks → Daily digest at 6 PM
LOW tasks → Weekly digest on Mondays

This decision is backed by research from high-performing maintenance teams (SEALs logistics, aviation maintenance, nuclear operations) which shows that alert fatigue is a primary cause of missed critical issues. By batching lower-priority tasks, we preserve attention for genuinely urgent work while ensuring nothing is lost.

For staging, all notifications route to jadasailing@gmail.com using the email service configured in GAS's MailApp API with a "JADA Maintenance" sender alias.

Calendar Integration

The MaintenanceCalendar.gs` file creates Google Calendar events for scheduled maintenance. When a task includes a scheduled date, it:

aws s3 cp staging-index.html s3://jada-maintenance-staging-us-west-2/tools/maintenance/staging-index.html --cache-control "max-age=0, no-cache"

aws cloudfront create-invalidation --distribution-id E1234... --paths "/tools/maintenance/*"

clasp push