Uptime · Ping · SSL
GeeksUptime watches your sites, APIs, and certificates from eight regions, then tells the people who own the fix — and stays quiet when there is nothing to say.
Example GeeksUptime console: five monitors across eight regions. Checkout is down in two regions with an open incident routed to Slack, email, SMS, and a webhook.
Quorum met at 09:41 UTC. Frankfurt and London return 503; the other six regions still answer. Response times climbed for six minutes before the first failure.
Three jobs, one workspace
Monitoring is only useful if it ends in a decision. Every screen here is built to move you from “something changed” to “someone is on it”.
Schedule HTTP, HTTPS, ping, DNS-record, and TCP-port checks. Pick compatible regions, the interval, and how many regions must fail before it counts. Response times are kept, so you can see the slowdown that came before the outage. DNS and TCP are included on Complete and Ultimate.
Notification groups decide who hears what. Attach a group to the checks and domains it covers, and open a maintenance window when work is planned — results keep recording while the paging stops.
A certificate nobody renewed takes a service down just as hard as a bad deploy. Track expiry dates and fingerprint changes per domain, and get told weeks ahead rather than on the morning it breaks.
Android app
The GeeksUptime Android app carries the full operating workflow with you. Create and manage checks, respond to incidents, review domains and certificates, schedule maintenance, configure notifications, switch workspaces, manage team access, inspect activity, and handle billing from your phone.
One account, the same workspaces and permissions, on web and Android.
Example GeeksUptime Android overview showing workspace totals, an open checkout incident, and access to uptime checks, domains, maintenance, and notification groups.
2 of 8 regions return 503. Failure threshold reached.
Production workspace · Android overview
The part nobody advertises
Most alert fatigue is not caused by real outages. It is caused by monitoring that cannot tell the difference between “your service is broken” and “we could not reach it”.
All the regions you selected returned a definitive result, and failures stayed below your threshold. This is the only state that counts as healthy.
Enough regions got a real answer you rejected — a bad status code, a failed content match, a refused connection. The incident opens on that cycle, not the next one.
A probe timed out, a runner could not be reached, or a result came back unreadable. That is our failure, not yours, so it stays out of your incident history and your numbers.
Regional agreement
Every region probes your endpoint once per cycle and reports independently. You decide how many have to agree, so a single unhappy network path never wakes anyone up.
Checkout · threshold 2 of 8 · quorum met 09:41 UTC
Alert routing
Pick the channels per notification group. The same event can post to Slack or Telegram for visibility, text the on-call engineer, and hit your automation endpoint with structured JSON.
See every integrationPricing
Every account begins with 30 days of Growth features and no card. Pick a paid plan in Billing only when you want monitoring to keep running.
Great for small projects
$10 / month
Plus everything below
Start 30-day trialFor growing operations
$25 / month
Plus everything below
Start 30-day trialFor high-traffic production
$60 / month
Plus everything below
Start 30-day trialFor enterprise scale
$250 / month
Plus everything below
Start 30-day trialMonthly billing, no setup fee. Compare the plans in detail.
Questions
Still unsure whether it fits how your team works? Ask us directly — a person answers.
Yes. Run HTTP, HTTPS, and ICMP ping checks, add DNS-record or TCP-port checks on Complete and Ultimate, and track SSL expiry dates and certificate fingerprint changes from the same workspace.
Each check runs from every region you selected, once per cycle. An incident opens as soon as the number of failing regions meets the failure threshold you set, and it resolves automatically when the service comes back.
That run is recorded as unknown rather than down. Unknown runs never open an incident, never send an alert, and never count against your availability figure. Only a genuine response from your service can move a check to down.
Notification groups route to email, SMS, Slack, Telegram, a generic HTTPS webhook, an AMQP queue, and mobile push. You attach a group to the checks and domains it should cover, so each team only hears about what it owns.
Open a maintenance window for a workspace or specific checks in your own timezone. Monitoring keeps recording results the whole time, but no new incidents open and no notifications go out until the window ends.
Thirty days of Growth features with no card. You will see the exact end date in your account as soon as you sign up, and you choose a paid plan in Billing only if you want monitoring to continue.
Get set up today
Add your first check in about two minutes, point one notification group at your team, and let the quiet days stay quiet.
No card · cancel by doing nothing · 30 days of Growth features