Skip to content

System monitor ​

A live health check of everything the exchange depends on: web server, database, queues, websockets, price feeds, blockchain nodes and integrations. It can email you when something goes offline.

ReferenceOperationsSuperadminVerified against the platform code
yourdomain.com/exchange-control-panel/system-monitor
Admin → Overview → System Monitor
Admin → Overview → System Monitor

Reading the page ​

The summary at the top counts services online, in maintenance, degraded and offline, and the Dashboard shows the overall percentage online. Services that need attention are listed first, with their error message. Filter the list by category.

Checks run every 5 minutes in the background. While a check is running, its card shows Checking.

What's checked ​

AreaServices
CoreWeb Server, Database Connection, Redis Server, Website Access, Encryption Module, Cache Service, API Gateway, Artisan Commands
BackgroundJob Queue Processor, Task Scheduler, Websockets
Market dataPrice Feed Service, Market Data Service, Liquidity Provider (every market with liquidity on had a book update in the last 2 minutes), Background Workers (no service stopped or stuck; see Market services)
BlockchainsBitcoin Node, Ethereum Node (ERC-20), Binance Smart Chain (BEP-20), TRON Network (TRC-20), Solana Network, Ripple (XRP) Network, TON Network, Polygon Network
IntegrationsFireblocks Custody, Stripe Payment Gateway, KYC Verification
Security & deliveryTwo-Factor Authentication, Rate Limiter, Email Service, Push Notifications, Webhook Delivery

A blockchain or integration you don't use may show as offline. That's expected.

Email Service connects to the mail server and logs in, without sending anything. It shows degraded when MAIL_MAILER=log (emails are written to the log, not delivered). API mailers (SES, Mailgun, Postmark) count as online when configured; a failed email shows up in the failed jobs.

Alert emails that wait in the queue for more than 15 minutes (for example while the queue was down) are dropped instead of being sent late, so a recovered queue never sends a burst of old alerts.

Email alerts ​

Click Set up alerts:

FieldMeaning
Send alert emailsEmail the team when a service goes offline after a health check. Alerts go to the admin email from Settings → Notifications.
Wait between emails5 to 1440 minutes, so an outage doesn't flood the inbox

The dialog also shows when the last alert was sent.

When a service is offline ​

ServiceFirst things to check
Job Queue Processorpm2w status: is horizon online? See Market services
Task SchedulerThe cron entry for www-data (see Installation)
Websocketspm2w status: is soket online? Nginx proxy for the socket subdomain
A blockchain nodeThe bridge process for that network (pm2w logs bridge-…)
Email ServiceEmail settings

Error messages are shortened and masked on this page, so passwords and keys never appear here.

Owner documentation for the CryptoTech exchange platform.