Cloudflare Services

show.fm uses multiple Cloudflare services for hosting, storage, and edge computing.

Account Information

Property	Value
Account ID	`b8eff1b484adf398bda38644efc85bce`
Zone	`podcasterplus.com`
Plan	Workers Paid (required for Queues)

Services Overview

┌─────────────────────────────────────────────────────────────────────────────────┐
│                        show.fm Infrastructure                              │
├─────────────────────────────────────────────────────────────────────────────────┤
│                                                                                  │
│   ┌──────────────────┐    ┌──────────────────────────────────────────────────┐ │
│   │  Cloudflare      │    │             Cloudflare Workers                    │ │
│   │    Pages         │    │                                                   │ │
│   │                  │    │  ┌─────────────┐  ┌─────────────┐               │ │
│   │  Main App        │    │  │  RSS Feed   │  │  Scheduled  │               │ │
│   │  Guest Portal    │    │  │   Worker    │  │  Publisher  │               │ │
│   │  (app. / book.)  │    │  │  (feed.)    │  │   (cron)    │               │ │
│   │                  │    │  └─────────────┘  └─────────────┘               │ │
│   └────────┬─────────┘    │                                                   │ │
│            │              │  ┌─────────────┐  ┌─────────────┐               │ │
│            │              │  │ Automation  │  │ Automation  │               │ │
│            │              │  │  Scheduler  │  │  Executor   │               │ │
│            │              │  │   (cron)    │  │  (queue)    │               │ │
│            │              │  └─────────────┘  └─────────────┘               │ │
│            │              └─────────────────────┬────────────────────────────┘ │
│            │                                    │                               │
│            ▼                                    ▼                               │
│   ┌────────────────────────────────────────────────────────────────────────┐  │
│   │                         Supporting Services                             │  │
│   │                                                                          │  │
│   │  ┌────────┐  ┌─────────────┐  ┌───────────┐  ┌───────┐  ┌───────────┐ │  │
│   │  │   R2   │  │    KV       │  │  Queues   │  │ Hyper │  │   DNS     │ │  │
│   │  │ Media  │  │  RSS Cache  │  │ (4 types) │  │ drive │  │           │ │  │
│   │  └────────┘  └─────────────┘  └───────────┘  └───────┘  └───────────┘ │  │
│   │                                                                          │  │
│   └────────────────────────────────────────────────────────────────────────┘  │
│                                                                                  │
└─────────────────────────────────────────────────────────────────────────────────┘

Cloudflare Pages

Main Application

Property	Value
Project Name	`podcaster-plus-app`
Domain	`app.podcasterplus.com`
Framework	SvelteKit
Build Output	`.svelte-kit/cloudflare`
Configuration	`/wrangler.toml`

Deployment:

bash

cd /path/to/podcaster-plus-app
CLOUDFLARE_ACCOUNT_ID=b8eff1b484adf398bda38644efc85bce \
  npx wrangler pages deploy .svelte-kit/cloudflare --project-name podcaster-plus-app

Bindings:

toml

# wrangler.toml
[[r2_buckets]]
binding = "MEDIA_BUCKET"
bucket_name = "podcasterplus-media"

[[queues.producers]]
queue = "rss-invalidation"
binding = "RSS_INVALIDATION_QUEUE"

[[queues.producers]]
queue = "automation-executions"
binding = "AUTOMATION_EXECUTION_QUEUE"

Guest Booking Portal

Property	Value
Project Name	`podcaster-plus-app`
Domain	`book.podcasterplus.com`
Status	Active
Routes	`/book/[host]/[slug]`

The guest booking portal is served from the main SvelteKit application via subdomain routing. See Google Calendar Integration for the booking system architecture.

Cloudflare Workers

show.fm uses four Cloudflare Workers for edge computing:

RSS Feed Worker

Property	Value
Worker Name	`podcasterplus-rss-feed`
Domain	`feed.podcasterplus.com`
Source	`/workers/rss-feed/`
Handler Types	HTTP (fetch) + Queue (consumer)
Documentation	RSS Feed Worker

Deployment:

bash

cd workers/rss-feed && npx wrangler deploy

Bindings:

KV Namespace: RSS_CACHE
Hyperdrive: HYPERDRIVE
Queue Consumer: rss-invalidation

Scheduled Publisher Worker

Property	Value
Worker Name	`podcasterplus-scheduled-publisher`
Source	`/workers/scheduled-publisher/`
Handler Types	Cron (every minute) + HTTP (health)
Documentation	Scheduled Publisher

Purpose: Automatically publishes episodes at their scheduled time and emits automation events.

Deployment:

bash

cd workers/scheduled-publisher && npx wrangler deploy

Bindings:

Hyperdrive: HYPERDRIVE
Queue Producer: RSS_INVALIDATION_QUEUE

Automation Scheduler Worker

Property	Value
Worker Name	`podcasterplus-automation-scheduler`
Source	`/workers/automation-scheduler/`
Handler Types	Cron (every minute) + Queue (consumer) + HTTP
Documentation	Automation Scheduler

Purpose: Processes time-based automation triggers and event-based triggers from the queue.

Deployment:

bash

cd workers/automation-scheduler && npx wrangler deploy

Bindings:

Hyperdrive: HYPERDRIVE
Queue Producer: AUTOMATION_QUEUE (to automation-executions)
Queue Consumer: automation-events

Automation Executor Worker

Property	Value
Worker Name	`podcasterplus-automation-executor`
Source	`/workers/automation-executor/`
Handler Types	Queue (consumer) + HTTP (health)
Documentation	Automation Executor

Purpose: Executes automation actions (email, webhook, field update) with delay handling.

Deployment:

bash

cd workers/automation-executor && npx wrangler deploy

Bindings:

Hyperdrive: HYPERDRIVE
Queue Consumer: automation-executions

R2 Object Storage

Media Bucket

Property	Value
Bucket Name	`podcasterplus-media`
Binding	`MEDIA_BUCKET`
Public URL	`https://media.podcasterplus.com`
Purpose	Audio files, podcast artwork, transcripts

Directory Structure:

podcasterplus-media/
├── podcasts/
│   └── {podcast-id}/
│       ├── images/
│       │   ├── cover/
│       │   │   └── {uuid}.jpg          # 1400x1400 cover art
│       │   ├── header/
│       │   │   └── {uuid}.jpg          # Header/banner image
│       │   ├── profile/
│       │   │   └── {uuid}.jpg          # Profile image
│       │   └── banner/
│       │       └── {uuid}.jpg          # Banner image
│       └── episodes/
│           └── {episode-id}/
│               └── audio/
│                   └── {uuid}.mp3      # Episode audio (mp3/m4a/wav/ogg)
├── imports/
│   └── {user-id}/
│       └── {uuid}.jpg                  # Temporary artwork imports from RSS
└── users/
    └── {user-id}/
        └── avatar.jpg

Supported Audio Formats: MP3, M4A, WAV (max 500MB)

Image Types:

Type	Purpose	Typical Size
`cover`	Podcast cover art	1400x1400 - 3000x3000
`header`	Page header/banner	Variable
`profile`	Circular profile image	400x400
`banner`	Wide banner image	Variable

CORS Configuration:

json

{
	"AllowedOrigins": ["https://app.podcasterplus.com", "https://book.podcasterplus.com"],
	"AllowedMethods": ["GET", "PUT", "HEAD"],
	"AllowedHeaders": ["*"],
	"MaxAgeSeconds": 3600
}

Cloudflare Image Transformations

show.fm uses Cloudflare Image Transformations for on-demand image resizing and optimization.

Transform URL Pattern:

https://app.podcasterplus.com/cdn-cgi/image/{options}/{source-url}

Preset Variants:

Variant	Width	Height	Quality	Use Case
`thumbnail`	150	150	80	List views, small cards
`avatar`	100	100	85	Profile images
`card`	300	300	85	Card displays
`preview`	600	600	85	Preview images
`hero`	1200	1200	90	Hero sections
`social`	1200	630	90	Social media shares

Usage (from src/lib/utils/image-url.ts):

typescript

import { getVariantUrl, generateSrcSet, getOriginalUrl } from '$lib/utils/image-url';

// Get optimized URL for a preset variant
const thumbnailUrl = getVariantUrl(originalUrl, 'thumbnail');

// Generate responsive srcset
const srcSet = generateSrcSet(originalUrl, [150, 300, 600]);

// Get original R2 URL (for RSS feeds - must be JPEG/PNG)
const rssImageUrl = getOriginalUrl(originalUrl);

Note: RSS feeds require original URLs (not transformed) because podcast apps expect JPEG/PNG formats.

KV Namespaces

RSS Cache

Property	Value
Namespace ID	`c22b685ea2bb4099a58a4788b58e1007`
Binding	`RSS_CACHE`
Used By	RSS Feed Worker

Key Patterns:

feed:{slug}           # XML content (TTL: 3600s)
feed:{slug}:etag      # ETag hash (TTL: 3600s)
feed:{slug}:meta      # Metadata JSON (TTL: 3600s)

Management:

bash

# List keys
npx wrangler kv:key list --namespace-id=c22b685ea2bb4099a58a4788b58e1007

# Get a value
npx wrangler kv:key get --namespace-id=c22b685ea2bb4099a58a4788b58e1007 "feed:wp-tavern"

# Delete a key
npx wrangler kv:key delete --namespace-id=c22b685ea2bb4099a58a4788b58e1007 "feed:wp-tavern"

Hyperdrive

Supabase Connection Pool

Property	Value
Config ID	`a81d477ff9264805989f5a72f0354ee8`
Binding	`HYPERDRIVE`
Target	Supabase PostgreSQL
Used By	All 4 Workers

Benefits:

10-100x faster cold starts vs direct connections
Automatic connection pooling
Prepared statement caching (disabled for compatibility)

Configuration:

toml

# All worker wrangler.toml files
[[hyperdrive]]
binding = "HYPERDRIVE"
id = "a81d477ff9264805989f5a72f0354ee8"

Usage:

typescript

import postgres from 'postgres';

const sql = postgres(env.HYPERDRIVE.connectionString, {
	max: 1, // Hyperdrive handles pooling
	prepare: false // Required for Hyperdrive
});

Management:

bash

# List configs
npx wrangler hyperdrive list

# Get config details
npx wrangler hyperdrive get a81d477ff9264805989f5a72f0354ee8

# Update connection string
npx wrangler hyperdrive update a81d477ff9264805989f5a72f0354ee8 \
  --origin-host=xxx.supabase.co \
  --origin-port=5432 \
  --database=postgres \
  --origin-user=postgres \
  --origin-password=xxx

Queues

show.fm uses four Cloudflare Queues for async messaging:

RSS Invalidation Queue

Property	Value
Queue Name	`rss-invalidation`
Producer	Main App, Scheduled Publisher
Consumer	RSS Feed Worker
Max Batch Size	10
Max Retries	3

Message Schema:

typescript

type RssInvalidationMessage = {
	type: string; // Event type (e.g., "episode.published")
	podcast_id: string; // UUID
	podcast_slug: string; // Cache key identifier
	episode_id?: string; // Optional
	timestamp: string; // ISO 8601
};

Automation Events Queue

Property	Value
Queue Name	`automation-events`
Producer	Reserved for future use
Consumer	Automation Scheduler Worker
Max Batch Size	50
Max Retries	3
Dead Letter Queue	`automation-dlq`

Database-Backed Pattern

Current Implementation: The main app does NOT send messages to this queue directly. Instead, event producers write to the automation_executions database table. The Scheduler Worker polls this table and sends messages to the automation-executions queue.

This queue exists for potential future use cases requiring real-time queue-based event dispatch.

Message Schema (if queue is used):

typescript

type AutomationEventMessage = {
	trigger_type: string; // e.g., "booking.confirmed"
	podcast_id: string;
	episode_id?: string;
	booking_id?: string;
	guest_id?: string;
	timestamp: string; // ISO 8601
};

Automation Executions Queue

Property	Value
Queue Name	`automation-executions`
Producer	Automation Scheduler Worker
Consumer	Automation Executor Worker
Max Batch Size	25
Max Retries	5
Dead Letter Queue	`automation-dlq`

Message Schema:

typescript

type AutomationExecutionMessage = {
	type: 'execute_rule';
	rule_id: string;
	execution_id: string;
	payload: {
		trigger_type: string;
		podcast_id: string;
		episode_id?: string;
		booking_id?: string;
		guest_id?: string;
		context: Record<string, unknown>;
		timestamp: string;
	};
	attempt: number;
	scheduled_at: string;
};

Automation Dead Letter Queue

Property	Value
Queue Name	`automation-dlq`
Purpose	Failed automation messages after max retries
Producers	`automation-events`, `automation-executions`

Queue Flow Diagram:

Main App (Producers)
   │
   ├──[Queue]──────────────────────────────┐
   │                                        │
   │  ┌──[Database]───────────────────┐    │
   │  │                                │    │
   ▼  ▼                                ▼    ▼
[rss-invalidation]              automation_executions (DB)
   │                            automation_scheduled_jobs (DB)
   ▼                                       │
RSS Feed Worker                            │
   │                                       ▼
   ▼                            Automation Scheduler (Cron)
KV Cache                                   │
                                           ▼
                                [automation-executions]
                                           │
                                           ▼
                                Automation Executor
                                           │
                               ┌───────────┼───────────┐
                               ▼           ▼           ▼
                             Email      Webhook    Field Update
                            (Resend)   (HTTP)      (Database)

                    ┌─────────────────────────────────────┐
                    │           automation-dlq            │
                    │  (Failed messages after retries)    │
                    └─────────────────────────────────────┘

Note: Automation uses database-backed queue pattern.
Main app writes to DB tables, Scheduler polls and dispatches.

Create Queues (if not exists):

bash

npx wrangler queues create rss-invalidation
npx wrangler queues create automation-events
npx wrangler queues create automation-executions
npx wrangler queues create automation-dlq

DNS Configuration

Subdomains

Subdomain	Type	Target	Purpose
`app`	CNAME	`podcaster-plus-app.pages.dev`	Main application
`feed`	Route	Workers route binding	RSS feeds
`media`	CNAME	R2 public bucket	Media CDN
`book`	CNAME	`podcaster-plus-app.pages.dev`	Guest booking portal

Workers Routes

toml

# RSS Feed Worker route
routes = [
  { pattern = "feed.podcasterplus.com/*", zone_name = "podcasterplus.com" }
]

Secrets Management

Main App Secrets

Set via Cloudflare Dashboard or wrangler secret:

Secret	Description
`STRIPE_SECRET_KEY`	Stripe API key
`STRIPE_WEBHOOK_SECRET`	Stripe webhook signing
`SUPABASE_SECRET_KEY`	Database access
`RSS_INVALIDATION_SECRET`	RSS cache invalidation auth
`RESEND_API_KEY`	Email service API key
`GOOGLE_CLIENT_SECRET`	Google OAuth client secret

Worker Secrets

Each worker requires specific secrets:

RSS Feed Worker:

bash

cd workers/rss-feed
npx wrangler secret put SUPABASE_SECRET_KEY
npx wrangler secret put RSS_INVALIDATION_SECRET

Scheduled Publisher:

bash

cd workers/scheduled-publisher
npx wrangler secret put SUPABASE_SECRET_KEY
npx wrangler secret put RSS_INVALIDATION_SECRET

Automation Scheduler:

bash

cd workers/automation-scheduler
npx wrangler secret put SUPABASE_SECRET_KEY

Automation Executor:

bash

cd workers/automation-executor
npx wrangler secret put SUPABASE_SECRET_KEY
npx wrangler secret put RESEND_API_KEY
npx wrangler secret put RESEND_FROM_EMAIL

Environment Variables

Main App (`wrangler.toml`)

toml

[vars]
PUBLIC_APP_URL = "https://app.podcasterplus.com"
PUBLIC_SITE_URL = "https://app.podcasterplus.com"
PUBLIC_BOOK_URL = "https://book.podcasterplus.com"
PUBLIC_SUPABASE_URL = "https://cmhnfgvbfrkgayrrgrmf.supabase.co"
PUBLIC_SUPABASE_PUBLISHABLE_KEY = "sb_publishable_xxx"
PUBLIC_STRIPE_PUBLISHABLE_KEY = "pk_test_xxx"
R2_PUBLIC_URL = "https://media.podcasterplus.com"

Workers (`workers/*/wrangler.toml`)

toml

[vars]
PUBLIC_MEDIA_URL = "https://media.podcasterplus.com"
PUBLIC_FEED_URL = "https://feed.podcasterplus.com"
PUBLIC_SUPABASE_URL = "https://cmhnfgvbfrkgayrrgrmf.supabase.co"
PUBLIC_APP_URL = "https://app.podcasterplus.com"
PUBLIC_BOOK_URL = "https://book.podcasterplus.com"

Deployment Commands

Full Deployment

bash

# 1. Build main app
pnpm build

# 2. Deploy main app
CLOUDFLARE_ACCOUNT_ID=b8eff1b484adf398bda38644efc85bce \
  npx wrangler pages deploy .svelte-kit/cloudflare --project-name podcaster-plus-app

# 3. Deploy all Workers
cd workers/rss-feed && npx wrangler deploy
cd ../scheduled-publisher && npx wrangler deploy
cd ../automation-scheduler && npx wrangler deploy
cd ../automation-executor && npx wrangler deploy

Individual Services

bash

# Pages only
pnpm run deploy

# Individual workers
cd workers/rss-feed && npx wrangler deploy
cd workers/scheduled-publisher && npx wrangler deploy
cd workers/automation-scheduler && npx wrangler deploy
cd workers/automation-executor && npx wrangler deploy

# View logs
npx wrangler pages deployment tail --project-name podcaster-plus-app
npx wrangler tail podcasterplus-rss-feed
npx wrangler tail podcasterplus-scheduled-publisher
npx wrangler tail podcasterplus-automation-scheduler
npx wrangler tail podcasterplus-automation-executor

Monitoring & Debugging

Dashboard Locations

Service	Dashboard Path
Pages	Workers & Pages > podcaster-plus-app
RSS Worker	Workers & Pages > podcasterplus-rss-feed
Publisher	Workers & Pages > podcasterplus-scheduled-publisher
Scheduler	Workers & Pages > podcasterplus-automation-scheduler
Executor	Workers & Pages > podcasterplus-automation-executor
R2	R2 > podcasterplus-media
KV	Workers & Pages > KV > RSS_CACHE
Queues	Workers & Pages > Queues
Hyperdrive	Workers & Pages > Hyperdrive

CLI Commands

bash

# Real-time logs
npx wrangler tail [worker-name]
npx wrangler pages deployment tail --project-name podcaster-plus-app

# KV inspection
npx wrangler kv:key list --namespace-id=c22b685ea2bb4099a58a4788b58e1007
npx wrangler kv:key get --namespace-id=c22b685ea2bb4099a58a4788b58e1007 "feed:podcast-slug"

# Queue inspection (via dashboard)
# Cloudflare Dashboard > Queues > [queue-name]

# Hyperdrive status
npx wrangler hyperdrive list
npx wrangler hyperdrive get a81d477ff9264805989f5a72f0354ee8

Cost Considerations

Service	Free Tier	Paid Tier
Pages	Unlimited sites	N/A
Workers	100K req/day	$5/mo + usage
R2	10GB storage, 1M Class A, 10M Class B	Usage-based
KV	100K reads/day, 1K writes/day	Usage-based
Queues	Requires Workers Paid	Included
Hyperdrive	Requires Workers Paid	Included
Image Transforms	5000 transforms/mo	Usage-based

Cloudflare Services ​

Account Information ​

Services Overview ​

Cloudflare Pages ​

Main Application ​

Guest Booking Portal ​

Cloudflare Workers ​

RSS Feed Worker ​

Scheduled Publisher Worker ​

Automation Scheduler Worker ​

Automation Executor Worker ​

R2 Object Storage ​

Media Bucket ​

Cloudflare Image Transformations ​

KV Namespaces ​

RSS Cache ​

Hyperdrive ​

Supabase Connection Pool ​

Queues ​

RSS Invalidation Queue ​

Automation Events Queue ​

Automation Executions Queue ​

Automation Dead Letter Queue ​

DNS Configuration ​

Subdomains ​

Workers Routes ​

Secrets Management ​

Main App Secrets ​

Worker Secrets ​

Environment Variables ​

Main App (wrangler.toml) ​

Workers (workers/*/wrangler.toml) ​

Deployment Commands ​

Full Deployment ​

Individual Services ​

Monitoring & Debugging ​

Dashboard Locations ​

CLI Commands ​

Cost Considerations ​

Related Documentation ​

Cloudflare Services

Account Information

Services Overview

Cloudflare Pages

Main Application

Guest Booking Portal

Cloudflare Workers

RSS Feed Worker

Scheduled Publisher Worker

Automation Scheduler Worker

Automation Executor Worker

R2 Object Storage

Media Bucket

Cloudflare Image Transformations

KV Namespaces

RSS Cache

Hyperdrive

Supabase Connection Pool

Queues

RSS Invalidation Queue

Automation Events Queue

Automation Executions Queue

Automation Dead Letter Queue

DNS Configuration

Subdomains

Workers Routes

Secrets Management

Main App Secrets

Worker Secrets

Environment Variables

Main App (`wrangler.toml`)

Workers (`workers/*/wrangler.toml`)

Deployment Commands

Full Deployment

Individual Services

Monitoring & Debugging

Dashboard Locations

CLI Commands

Cost Considerations

Related Documentation