01 Dependencies
Is it them, or is it you? We probe the real endpoints of the providers you rely on and reconcile that with their status page and crowd reports. One verdict, even while their page still reads operational.
Real endpoint probes, not a reprinted status page A handful of reports never cries wolf Alerts on your dependencies like on your own monitors How reconciliation works → S Stripe API dependency
Official status page
Our probes
Crowd reports
Verdict
Early warning: Stripe API is degrading
It's them, not you. Their page still reads operational.
02 Catalog
500+ services, already watched Clouds, CDNs, payment and AI APIs. Pick the ones your stack leans on and you are covered in a click.
AWS, Stripe, GitHub, Cloudflare, OpenAI and more No setup: we already run the checks See what broke first when several fail together Check a service now →
03 Monitors
Watch from everywhere HTTP, TCP, TLS, and more — every 15 seconds. A failure is re-checked from other regions before anyone is paged.
Multi-region confirmation kills false alarms Checks as fast as every 15 seconds Certificate expiry and response content rules Monitor docs → iad timeout fra timeout sin timeout syd 231ms lax 61ms api.acme.com Down, confirmed from 3 regions. Incident opened.
04 Alerts
The alarm and the all-clear One alert when an incident opens, one when it recovers. Email, SMS, Slack, Discord or a webhook.
Recovery is a first-class event No alert storm while something stays down Per-contact channels and priorities 14:02:11 Incident opened
checkout-api · HTTP 503 from 3 regions
No repeat pings while it stays down
14:06:23 Recovered
Down for 4m 12s · all-clear sent everywhere
05 Latency SRE
A diagnosis, not just down DNS, TCP, TLS, HTTP, hosting and third-party dependencies checked in one pass, with a plain-language verdict.
Tells you which layer broke Names the provider when it is not your fault Free on any domain, no account Try it on your domain → ⌕ latency.app/inspect/shop.acme.com
✓ DNS resolves, 2 A records
✓ TCP connect 18ms
✓ TLS valid, expires in 61 days
✕ HTTP 503 from the edge
✕ Deps Cloudflare: degraded
Latency SRE
Your origin is healthy. The CDN in front of it is returning 503s, and Cloudflare is degraded right now. Nothing to fix on your side.
06 Timings and rules
See what's actually slow Every check is broken into DNS, connect, TLS and first byte. Alert on status codes, percentiles or a slow phase.
Per-phase timings on every check Rules on p50 to p99, jitter and failures Rolling windows instead of single blips Anatomy of a check → GET /checkout 200 412ms
DNS 14ms
TCP 22ms
TLS 41ms
TTFB 318ms
Body 17ms
rule p95 TTFB > 300ms for 3 checks → alert
07 Telemetry Beta
Logs and traces in one place Send OpenTelemetry logs, traces and events, then search them in the same console that told you something broke.
OpenTelemetry native ingest One search across logs, spans and events Dashboards and monitors on your own data Telemetry docs → ⌕ level:error stream:checkout Live
14:02:09 info POST /checkout 200 182ms
14:02:10 warn stripe: retry 1/3 after timeout
14:02:11 error stripe: upstream 503 charge.create
14:02:11 error POST /checkout 502 3012ms
14:02:12 error span checkout.pay failed trace=8f2c…
08 iOS app
In your pocket A native app with ping, traceroute, a live map and trending incidents. Free, no account needed.
Ping and traceroute from your phone Trending incidents at a glance Your monitors when you sign in About the app → San Francisco, US My Location ✓ 1.1.1.1 AS13335 Cloudflare, Inc.
••• 15.4ms MIN
104ms MAX
8.2ms JITTER
LOSS 0% STD DEV 13.9ms VIA Wi-Fi
2 4 6 ✓ yahoo.com AS26101 Yahoo Holdings Inc.
••• HOPS 6 DEST Reached LAST RTT 16.5ms
Hop 1 10.0.0.1 5.9ms
Hop 2 10.28.66.226 16.9ms
Hop 3 68.85.178.1 17.1ms
Hop 4 69.139.202.230 13.9ms
Hop 5 24.153.88.193 13.6ms
Hop 6 24.153.88.189 16.5ms
My Location Trending Incidents
Elastic Major Incident
No user reports yet SendGrid Degraded
No user reports yet Popular
The difference
Accurate, not just aggregated Most status trackers echo whatever a provider publishes. We reconcile the official page against our own live probes and the crowd, so a green page hiding a real outage doesn’t fool you, and a noisy status indicator with nothing behind it doesn’t either.
See how we compare → 3 signals, one verdict
Official status page Multi-region confirmation Crowd reports Your first monitor in two minutes. Start on the free plan. Upgrade when you need more monitors, SMS or teammates.