Solution · Uptime

The simplest answer to is it up.

Engager probes every host you operate every thirty seconds, from six regions, with the same care a senior engineer would. When a host stops responding, an incident opens before the page even loads in a browser somewhere.

30s

Default probe interval

Tunable per host

6

Regions

Cross continent triangulation

15s

Cold start guard

Stages 10s · 25s · 55s reverify

99.9%

SLA target

Rolling 7 day, 30 day windows

Continuous probing

One probe is luck. Engager makes a thousand.

Every host you add starts collecting one probe every thirty seconds, the moment it lands in your workspace. The default cadence stays low so a free tier render or a cold cache never gets paged for a fluke.

For paid tiers the floor drops to fifteen seconds, then five seconds, then one second for the surfaces that absolutely cannot blink.

engager.rookhq.com / @realm / trackers / domain / rookhq.com

Domain tracker

rookhq.com

Probed every 30s · 99.95% over 30d

Healthy

15 sub probes

SSL certificateCert valid 47d
Domain expiry (WHOIS)Expires in 412d
DNS resolutionA 76.76.21.21 · DNSSEC
Response timeP95 142ms · P99 311ms
Security headers4/4 · HSTS preload
Mixed contentClean
Keyword presenceOK
Open port drift443✓ 80✓

Recovery without waiting

The moment your service is back, the report is already on its way.

Most uptime tools wait until the next scheduled probe to confirm recovery. Engager spins a fast recovery loop when an incident opens, polling every ten seconds until a healthy response lands.

The recovery email goes out the second the response code clears · not on the cadence, on the truth.

Fast recovery probe

Ten second cadence on incident

Inside the probe

Six things every probe checks at once.

The legacy monitoring playbook fires one probe per concern. Engager fires one probe and answers six questions in the same round trip.

Reachability

TCP open, TLS handshake completed, body returned within the timeout you set per host.

Latency budget

Round trip recorded, p95 and p99 trended over rolling windows. Anomalies flagged when current run exceeds 1.5× p99.

DNS truth

A and CNAME records compared to last known. Unexpected drift opens a security event, not a quiet log line.

Cert chain

TLS protocol, cipher strength, full chain expiry. Warnings begin at thirty days, escalations begin at seven.

Body integrity

Page size compared to a rolling baseline. Sudden fifty percent jumps surface a content drift event.

Keyword presence

A list of must have phrases for the host. Page renders fine but the marketing footer is gone? Engager notices.

How a check lands

From the probe to the inbox in under a second.

  • Probe runs

    Six regional workers each fire an independent request. Latencies stored locally.

  • Roll up

    Snapshot pushed to your workspace every fifteen seconds. The dashboard reads from this snapshot, never the worker.

  • Routing decision

    For every channel that subscribes to the matching notification kind, the dispatcher honours timezone, active days, quiet hours, and mute windows.

  • Send

    Telegram, email, and Slack go in parallel. Failures of one branch never stall the others.

  • Audit

    Every fanout result goes into the report run table for replay later if you need to.

Six regions, one truth

No region speaks for the rest

Add your first host. Probes start in under a minute.

Free tier covers five hosts at three minute cadence. Paid tiers drop to thirty seconds and below.