Monitoring & Status
Know what broke and why, before a customer tells you
InfraNest checks your sites and servers around the clock. When something fails, it alerts you and shows what changed just before. No more hearing it from customers, guessing why, or reading logs at 2am.
Free plan · No credit card · First monitor live in under a minute
- See what changed just before it failed
- See exactly where it breaks
- Server health beside your uptime
- Suggestions that cut false alarms
Changed 3 min before: Deploy of northwyndr/storefront to production · by jordan-k See it
Show themHide them
- DNS for api.northwyndr.com was changed23 min ago · by Sam Okoye
- A firewall on api-01 was changed36 min ago · by Alex Rivera
Nothing touched it in the 30 minutes before — no deploy, and no change to its server, firewall, DNS or monitor.
See exactly where it breaks
A page depends on many things: your domain, DNS, the certificate, a load balancer, your servers, your database. InfraNest goes through them in order and names the one that breaks. You know where to look before you open a single log. It also splits each response into steps, from looking up the name to the download, and says in one sentence what slows it down.
- Names the layer that breaks
- Shows where the time goes, step by step
- Compares locations on Business
It breaks at: Database
- Domainnorthwyndr.com is registered
- DNSThe name resolves
- CertificateValid
- Load balancerapi-lb · none of its targets answers
- Web serverapi-01 is running, but fails the load balancer’s health check
- Web serverapi-02 is running, but fails the load balancer’s health check
- DatabaseFirewall db-tier blocks api-01, api-02 → db-01 on port 5432
- ApplicationNot reached
Asia-Pacific’s extra time is mostly connect and TLS: distance, not your server. A CDN or a server closer to Asia-Pacific would shorten it.
Comparing locations is on Business.
See the server behind your site, right beside its uptime
A site can look fine while its disk quietly fills up. InfraNest puts your servers next to your monitors: CPU, memory, disk and load, with the last 24 hours behind each figure. A filling disk comes with how long you have left. And when a server goes down, the monitors on it become one problem and one alert. Memory and disk need our small agent on your server.
- CPU, memory, disk and load per server
- Says when a disk will be full
- Monitors on one server fold into one alert
Get told how to watch better, without asking
InfraNest compares your monitors with everything else it knows about your setup. It finds the server nothing checks and the alert that keeps crying wolf. Each suggestion gives its reason and the one step that would fix it. You press the button or you say no. Nothing changes until you do.
- Finds what nothing is watching yet
- Offers one setting for each false alarm
- Spots a page that shrank after a deploy
Gaps and noise, found by comparing your monitors with everything else in InfraNest.
- Search API raised 4 false alarms in the last 30 days3 of them failed from Asia-Pacific only, while other locations saw it up. Waiting until two regions agree that it is down keeps one location’s network trouble out of alerts. An outage everyone sees still alerts.Not neededRequire 2 regions
- Marketing site shrank from 84 KB to 9 KB after a deployIt still answers 200, so it reads as up, but the page is 89% smaller than before the deploy: an error page or an empty shell looks exactly like this. A keyword check alerts when the words that should be on it are missing.Not neededAdd a keyword check
- Watch 15 servers that nothing checks yetThey were imported and nothing checks them yet. Each gets a ping every few minutes. Each uses one of your plan’s monitors.Not neededAdd 15 monitors
One outage, one incident, not ten alerts
When one thing fails, everything behind it fails too. Most tools then alert you for each one. Tell InfraNest what depends on what, and you get one incident that names the cause and lists everything it affected. The timeline shows the deploy, the failure and the alerts in order. When the trouble is on our side or at a provider location, it says “Likely not you”, and the incident still opens, alerts and counts.
- One incident that names the cause
- Every change, check and alert in one timeline
- Says why a name will not resolve
One incident, not several: the effects are grouped under the cause.
- 2:33 PMDeploy of northwyndr/storefront (f19e6c7)Deploys
- 2:36 PMCheckout flow failed: Checkout returning 503Monitoring
- 2:36 PMAPI health failed too, grouped under this incidentMonitoring
- 2:36 PMAlert sent to Ops team — email, On-call — critical only, #infra-alerts, Engineering — Discord, Northwind on-call — PagerDutyNotifications
Find out if your last deploy made things worse
The first question in every outage is “did we just ship something?”. Link a GitHub repository and every deploy appears on your charts. For an hour after a production deploy, the monitors it reaches are checked more often. Then you get one plain answer per location: better, no change or worse. If it got worse, Home tells you and links the deploy, so you can fix it while it is fresh.
- One answer per location, in plain words
- Checked every minute that hour, on Pro and up
- Put the last version back, after you confirm
e61f7d2 to production: p95 up to 30% slower from every location, failed checks unchanged
| Location | p95 before → after | Result |
|---|---|---|
| Asia-Pacific | 391 ms → 502 ms | Worse |
| Europe | 182 ms → 236 ms | Worse |
| North America | 268 ms → 342 ms | Worse |
The hour after the deploy, compared with the hour before it for every location: the slowest 5% of response times (p95) and the share of failed checks. Each side needs at least 10 checks.
Monitor websites, APIs, servers and cron jobs on one screen
Most outages are not a site going dark. A cron job stops, an API returns the wrong answer, a certificate runs out, a port closes. InfraNest watches for all of these. Each row shows the last 30 days, the response time and who gets the alert. Warnings can go to Slack and real outages to PagerDuty, with email, Teams, Discord or Telegram alongside.
- Websites, APIs, ports, DNS, certificates and cron jobs
- Group by domain or by server
- Filter to what is down or alerts nobody
No monitors match this filter.
No monitors match this filter.
No monitors match this filter.
Everything else it handles
The parts that are only interesting when you need them.
Checked from three continents
On Business, every check runs from Europe, North America and Asia-Pacific, and a status only changes when enough of them agree. On Free and Pro, you pick one region.
Alerts that reach the right person
Email, Slack, Teams, Discord, Telegram and PagerDuty. Set a destination to critical-only and it stays quiet for the rest.
Status pages
Turn what you monitor into a branded public page, with your incidents and maintenance on it.
Your own probes
Run a probe on one of your own servers to watch something only reachable from inside your network.
Maintenance windows
On Pro and up, schedule the work. The alerts, the status page and the uptime figure all know it was planned.
Automations on monitors
When a monitor fails, an automation can open a Jira or Linear issue, post to your status page or reboot the server. You decide when it runs.
First monitor live in under a minute
Paste a URL. No agent to install, nothing to configure, and the free plan keeps twenty of them.
Rolling your own vs InfraNest
The difference between knowing it is down and knowing what to do next.
By hand
- A customer tells you the site is down
- Asking around to find out what changed
- Reading logs at 2am to find the broken part
- Every dependent service alerting separately
- A cron job that stopped, noticed a week later
With InfraNest
- An alert from your own monitor, before anyone writes in
- The change just before it, named when one stands out
- The layer that breaks, named on the monitor page
- One incident naming the cause, with the rest listed under it
- A heartbeat monitor that notices the silence within minutes
Comparing tools? See Better Stack vs InfraNestCronitor vs InfraNestFreshping vs InfraNestOh Dear vs InfraNestPingdom vs InfraNestStatusCake vs InfraNestUptime Kuma vs InfraNestUptimeRobot vs InfraNest
One login, ten modules
Nine modules are on every plan, Free included, and firewalls start on Pro. What changes is how much of each you get, plus a few advanced features from Pro.
Build on it
The cost of doing it separately
Buy each piece from a different tool and it adds up fast:
- Uptime monitoring
- ~€26
- SSL tracking
- ~€15
- Status page
- ~€26
- Across 3 separate tools
- ~€67/mo
Also in Pro, not counted above: 100 domains with DNS, 10 servers across your cloud providers, 25 automations, Dynamic IP and Drop Catch.
Each line is that tool’s cheapest paid plan. Pro covers all of it: 50 monitors, 100 certificates and 3 status pages on your own domain.
Prices checked October 2026 · dollar prices at the ECB rate
Built for how you work
Try it with a free tool
Frequently asked
How do I find out why my website went down?
InfraNest looks at what changed in the 30 minutes before the failure: deploys, a provider outage, and edits to the server, firewall, DNS or monitor. If one kind of change stands out, it names it, with a link. If several do, it lists them and picks none. If nothing changed, it says so. It answers by rule, so it never guesses.
Where do the checks run from?
On Business, from Europe, North America and Asia-Pacific at once, and a status only changes when enough of them agree. That stops one flaky location from alerting you. On Free and Pro, each monitor checks from one region you choose. You can also run a probe on your own server to watch something only reachable from inside your network.
What can it monitor?
Websites and APIs, keyword and redirect checks, TCP ports, ping, DNS records, SSL certificates, heartbeat monitors for cron jobs, your servers’ CPU, memory, disk and services, and other providers’ status pages.
How do I reduce false alarms from monitoring?
On Business, a status only changes when enough locations agree. InfraNest also looks at your own history. If a monitor keeps failing briefly, or people keep marking its incidents as false alarms, it suggests the one setting that would help and says why. Nothing changes until you confirm.
How do I stop one outage sending ten alerts?
Tell InfraNest which monitors depend on which. When a parent goes down, its children are listed under one incident instead of each alerting separately. A service that keeps bouncing up and down is handled separately, so it does not alert you each time.
Can I monitor server CPU, memory and disk?
Yes. Your servers sit beside your monitors with CPU, memory, disk and load, and a forecast of when a disk will be full. Memory and disk come from our small agent. Without it, you see what your cloud provider reports.
Where can alerts go?
Email, Slack, Microsoft Teams, Discord, Telegram and PagerDuty. Each destination can be set to a severity, so a critical-only channel stays quiet for warnings. An automation can go further: open a Jira, Linear, Notion or GitHub issue, post to your status page or call your own webhook.
Does monitoring a cron job need an agent?
No. A heartbeat monitor gives you a URL for your script to call when it finishes. If the call does not arrive in time, you are alerted. That catches the job that silently stopped, which a URL check can never see.
What happens during planned maintenance?
On Pro and up, open a maintenance window. Checks inside it stop alerting, the status page says the work was planned, and your uptime figure is not punished for it.
Twenty monitors, on the free plan
Paste a URL. Your first check runs in under a minute, and the next outage comes with a reason.
Free plan · No credit card required · Set up in minutes