Always watching.
Never surprised.
Continuous HTTP, TCP, and DNS probes run across multiple AWS regions on 1- or 5-minute intervals. Catch regressions before they become user complaints.
Three check types
Status code validation, optional response body assertion, configurable auth header via Secrets Manager.
Confirms a port is reachable at the network layer — useful for non-HTTP services and firewall validation.
Verifies that a hostname resolves from each probe region — catches split-horizon and propagation failures early.
Under the hood
Multi-Region Probes
An EventBridge schedule fires a dispatcher Lambda that fans out check batches to per-region probe Lambdas. Results represent real-world reachability from each AWS region — not a single vantage point.
1-Minute & 5-Minute Intervals
Configure check frequency per check. High-priority endpoints run every minute; background services can run every five. Intervals are configurable at the check level.
Secrets Manager Integration
Auth credentials for protected endpoints — Bearer tokens, API keys — are stored in AWS Secrets Manager and referenced by check ID. No secrets in the database or admin UI.
Incident History
When checks fail, incidents are opened and tracked in the admin UI. Resolved automatically when the check recovers. 90-day retention in DynamoDB.
CloudWatch Metrics
Every probe emits structured EMF metrics: ProbeSuccess, ProbeLatencyMs, ProbeBatchJobs, and ProbeBatchDurationMs. Plug directly into existing CloudWatch dashboards and alarms.
Uptime & Latency Reports
Per-check report endpoints expose uptime percentage and latency percentiles over any time window. Surface these in your own dashboards or include them in compliance exports.
Serverless from end to end
No persistent polling servers. EventBridge triggers the dispatcher on schedule. The dispatcher fans out to regional probe Lambdas. Results land in DynamoDB with a 90-day TTL and emit CloudWatch EMF metrics for alarming.