InfraNestInfraNest

Monitoring & Status

Uptime & SSL monitoring from multiple regions

Watch websites, APIs, servers, DNS and certificates from probes around the world — and only hear about it when it is real. Consensus across regions kills the false alarms, dependency-aware incidents kill the alert storms, and everything lands in the channel your team actually reads.

Free plan · No credit card · First monitor live in under a minute

Monitors18 monitors · 5 regions
MonitorStatusResponse
api.acme.ioUp · 5 regions agree182 ms
acme.ioUp · 5 regions agree241 ms
checkout.acme.ioDegraded · FRA slow1,340 ms
status.acme.ioDown · 4 of 5 regions
db-worker heartbeatUp · last ping 41s agoon time

One region flaked on checkout — the others disagreed, so nobody got paged.

Checked from everywhere, confirmed by consensus

Every check runs from multiple regions at once, so you see the outage your users in Frankfurt hit even when it looks fine from Ashburn. A status only flips when enough probes agree, which means one flaky location never pages you for nothing.

  • Probes across Europe, North America and Asia-Pacific
  • A status changes only when your quorum of regions agrees
  • Per-region response times, so you can see where it is slow
checkout.acme.io · this checkQuorum: 4 of 5 must agree
Frankfurttimeout · 1,340 ms
Amsterdam200 OK · 196 ms
Ashburn200 OK · 214 ms
London200 OK · 188 ms
Singapore200 OK · 402 ms

One region disagreed — held as “degraded”, not paged as “down”.

One outage, one incident — not ten alerts

When a failure is confirmed, InfraNest opens a tracked incident instead of firing a page per check. If the root cause is a dependency — a database, a load balancer, an upstream API — the checks that only failed because of it are suppressed and folded into the same incident. Planned work goes into a maintenance window, so it never reads as an outage at all.

  • Dependency-aware suppression collapses the storm to one root cause
  • A full incident timeline: detected, confirmed, updated, resolved
  • Maintenance windows pause checks and mark the work as planned
Incident #482 · database unreachable1 root cause · 6 effects suppressed
ROOTpg-primary heartbeat missed for 90sRoot cause →
EFFECTapi.acme.io returning 500sSuppressed →
EFFECTcheckout.acme.io failingSuppressed →
EFFECTacme.io keyword check missing “In stock”Suppressed →
WINDOWNo maintenance window open — this is realPaged

The right people, once, on the channel they read

Alerts route by policy, not by brute force: warnings go to a Slack channel, hard-down pages an on-call phone, and everything can also hit Telegram, email or a webhook. Delivery waits for probe consensus first, so the message you get is a confirmed problem — not a probe having a bad second.

  • Email, Slack, Telegram and webhooks, chosen per severity
  • Policies escalate warning vs critical to different people
  • Recovery notices close the loop, so “is it back?” is answered
status.acme.io · from probe to page
First region reports a failed request0s
Re-checked across all 5 regions8s
4 of 5 regions agree it is down — confirmed22s
Critical policy pages on-call + posts to Slack24s

No alert left the building until the regions agreed.

Everything else it handles

The parts that are only interesting when you need them.

Ten check types

HTTP and API, TCP port, ICMP ping, keyword-in-body, and inbound heartbeat checks for cron and background jobs that fail silently.

SSL & expiry checks

Watch certificate and domain expiry alongside uptime, with alerts well before anything lapses and takes a site offline.

Response-time charts

Latency history per monitor and per region, so you can see a slow creep long before it becomes a hard outage.

On-call & escalation

Escalation policies that move an unacknowledged critical alert up the chain instead of dying in one muted inbox.

Flapping protection

A check that bounces up and down is held and confirmed across probes before it is allowed to alert, so noisy endpoints stay quiet.

Per-monitor thresholds

Set the timeout, retry count, check interval and region quorum per monitor — a marketing page and a payments API are not judged the same.

Alerting integrations

Send confirmed events to Slack, Telegram, email or any webhook, and drive downstream automations from the same signal.

Full audit trail

Every status change, incident update and acknowledgement recorded with who and when, exportable for post-mortems and compliance.

See a real check run in about a minute

Add one URL and watch it come up green from five regions. No agent to install, no credit card.

Basic uptime pings vs InfraNest

The difference between a noisy single-location ping and monitoring your team actually trusts.

By hand

  • One location — a flaky probe pages you at 3am for nothing
  • Pings a URL but ignores SSL, DNS and the dependencies behind it
  • One database outage becomes ten separate alerts
  • Monitoring, incidents and the status page in three tools

With InfraNest

  • Multi-region probes that confirm by consensus before alerting
  • Ten check types incl. SSL, DNS, keyword and heartbeat
  • Dependency-aware incidents collapse the storm to one root cause
  • Monitoring, incidents and alerting in one place

Pairs well with

Frequently asked

What can InfraNest monitor?

Websites, APIs, servers, DNS and TLS certificates — via HTTP, keyword, TCP port, ping, SSL, DNS and inbound heartbeat checks. Each runs from multiple regions with flapping protection built in.

How do you avoid false alarms at 3am?

A status only changes when a quorum of regions agrees, and flapping checks are held and re-confirmed before they can alert. A single probe having a bad second never reaches your phone.

What is dependency-aware root-cause suppression?

When one thing fails and takes others down with it — a database, a load balancer, an upstream API — InfraNest identifies the root cause and folds the knock-on failures into the same incident, so you get one page, not ten.

Which alert channels are supported?

Email, Slack, Telegram and webhooks today, routed by policy so warnings and critical pages can go to different people and places. Webhooks let you drive any downstream tool or automation.

Can I pause alerts for planned work?

Yes — open a maintenance window and the affected checks are paused and shown as planned work rather than an outage, on your alerts and on any status page fed by them.

Know before your users do

Add one check and watch it come up green from five regions.

Start for free