The best website uptime monitoring tools in 2026
UptimeRobot is the best free hosted monitor: 50 monitors checked every 5 minutes, and its pricing page says the free plan suits commercial projects too. Uptime Kuma is the best self-hosted option: on our test server it reported an HTTPS site up, SSH up and a closed port down within 20 seconds. Better Stack is the best all-in-one, with uptime checks, heartbeats, status pages, on-call and logs from one vendor.
For cron jobs and backups, add Healthchecks.io: it alerts you when a job fails to check in on time.
Our top picks
- UptimeRobotBest for free external checks for many websites
- Uptime KumaBest for self-hosted monitoring with a friendly web UI
- Better StackBest for uptime, status pages, on-call and logs from one vendor
Compare at a glance
| Tool | Best for | Pricing | Open source | Deployment | Tested by us |
|---|---|---|---|---|---|
| UptimeRobot | free external checks for many websites | Free: 50 monitors at 5-minute intervals; Solo $9 per month billed annually ($10 monthly) for 10 monitors with 60-second checks; Team $35 per month billed annually ($41 monthly) for 100 monitors; Scale from $65 per month billed annually ($77 monthly) for 200 monitors; Enterprise custom | No | SaaS | Docs only |
| Uptime Kuma | self-hosted monitoring with a friendly web UI | Free | Yes | Self-hosted | Yes |
| Better Stack | uptime, status pages, on-call and logs from one vendor | Free: 10 monitors and heartbeats, 1 status page, 3 GB of logs kept 3 days; Responder seats $29 per month billed yearly ($34 monthly); 50 extra monitors $21 per month yearly; US logs $0.15 per GB ingested plus $0.08 per GB per month retained | No | SaaS | Docs only |
| Healthchecks.io | cron jobs and backups that must run on time | Hobbyist free (20 jobs); Supporter $5 per month; Business $20 per month (100 jobs) or $16 per month yearly; Business Plus $80 per month (1,000 jobs) or $64 per month yearly | Yes | SaaS or self-hosted | Docs only |
| Gatus | health checks and a status page defined as code | Self-hosted free; managed gatus.io Standard $10, Premium $20 and Business $50 per month (20% off yearly, 7-day trial) | Yes | Self-hosted, or managed at gatus.io | Docs only |
| Pingdom | transaction checks plus real user monitoring | Synthetic monitoring from $18 per month ($16.50 billed annually) for 10 uptime checks, 1 advanced check and 50 SMS; Real User Monitoring from $18 per month for 100,000 pageviews; free trial | No | SaaS | Docs only |
| StatusCake | a free plan plus inexpensive 1-minute checks | Free (for individuals and small businesses, per its pricing page): 10 uptime monitors at 5-minute intervals; Superior $20.41 per month billed annually, $24.49 monthly (100 monitors, 1-minute checks); Business $66.66 per month billed annually, $79.99 monthly (300 monitors, 30-second checks); Enterprise custom | No | SaaS | Docs only |
| Checkly | API and browser checks for developers | Hobby free (10 uptime monitors, 10,000 API and 1,000 browser check runs a month); Starter $24 and Team $64 per month billed annually; Enterprise custom | No | SaaS | Docs only |
| Cronitor | hosted cron monitoring and uptime checks on one bill | Hacker free (5 monitors); Business $2 per monitor per month plus $5 per user per month | No | SaaS | Docs only |
| Oh Dear | agencies watching the overall health of many websites | From €15 per month (Solo, 2 sites) to €399 (Scale, 200 sites), every feature on every plan; 10-day free trial | No | SaaS | Docs only |
The full ranking
UptimeRobot
Best for: free external checks for many websites
UptimeRobot has the most generous free plan here: 50 monitors checked every 5 minutes and a status page, and its pricing page says the free plan is good for personal and commercial projects. Paid plans mostly buy speed and scale. Solo is $9 a month billed annually ($10 monthly) with 60-second checks, Team is $35 a month billed annually ($41 monthly) with 30-second checks and 100 monitors, and Scale starts at $65 billed annually ($77 monthly) with 15-second checks. It checks from outside your network, which is the whole point of an uptime monitor. On the free plan, an outage can run a few minutes before the next check notices it.
- Pricing
- Free: 50 monitors at 5-minute intervals; Solo $9 per month billed annually ($10 monthly) for 10 monitors with 60-second checks; Team $35 per month billed annually ($41 monthly) for 100 monitors; Scale from $65 per month billed annually ($77 monthly) for 200 monitors; Enterprise custom (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; checks public websites and endpoints from outside your network
Pros
- 50 free monitors, for commercial projects too (per its pricing page)
- Cheap paid tiers with 60-, 30- and 15-second checks
- Status pages on every plan
Cons
- Free checks run only every 5 minutes
- Free plan is limited to 5 notification integrations
- Hosted only; no self-hosted option
Not tested by us yet: based on the vendor's documentation.
Uptime Kuma
Best for: self-hosted monitoring with a friendly web UI
Uptime Kuma is the open-source favorite: an MIT-licensed web app with HTTP, TCP, ping, DNS, keyword, JSON-query and push monitors, status pages and more than 90 notification services. On our Debian 12 test server, version 2.5.11 ran on Node.js 24 with SQLite. We added three monitors at 20-second intervals, and it reported an HTTPS site up (200 OK in about 60 ms), SSH up, and a closed port down with the exact connection error. It used about 190 MB of memory. Run it on a different server and network from the ones it watches, or one outage takes out both.
- Pricing
- Free (checked November 15, 2026, source)
- License
- Open source (MIT)
- Deployment
- Self-hosted
- Runs on
- Docker, or Node.js 20.4+ on major Linux distributions and Windows
Pros
- Free and open source (MIT), with no monitor limits
- Clean web UI, status pages and 90+ notification services
- Push monitors also cover cron jobs
- Worked within minutes in our test
Cons
- You host it, ideally on a separate server and network
- Checks only from where you install it
- About 190 MB of memory and a 278 MB install in our test
Tested by us: Uptime Kuma 2.5.11 on Debian 12.15, 14 Nov 2026 (UTC): official 2.5.11 source with Node.js 24.22.3 (checksum-verified), SQLite, bound to localhost; HTTPS, TCP and failing-port monitors at 20-second intervals. Uninstalled afterwards.
Better Stack
Best for: uptime, status pages, on-call and logs from one vendor
Better Stack bundles uptime checks, heartbeats, status pages, on-call scheduling and log storage. The free plan, meant for personal projects, includes 10 monitors and heartbeats, one status page and 3 GB of logs kept for 3 days, with Slack and email alerts. Paid use is priced per responder at $29 a month billed yearly ($34 monthly), with extra monitors in packs of 50 for $21 a month yearly, and the product advertises checks as often as every 30 seconds. It's the strongest pick if you want alerting, on-call and status pages without stitching tools together. For a few personal sites, UptimeRobot's free plan goes further.
- Pricing
- Free: 10 monitors and heartbeats, 1 status page, 3 GB of logs kept 3 days; Responder seats $29 per month billed yearly ($34 monthly); 50 extra monitors $21 per month yearly; US logs $0.15 per GB ingested plus $0.08 per GB per month retained (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; uptime checks, heartbeats, status pages, on-call and log storage
Pros
- Uptime, heartbeats, status pages, on-call and logs in one product
- Checks as often as every 30 seconds
- Free plan to try it
Cons
- Free plan is for personal projects only
- Per-responder pricing adds up for large on-call teams
- Hosted only
Not tested by us yet: based on the vendor's documentation.
Healthchecks.io
Best for: cron jobs and backups that must run on time
Healthchecks.io watches for things that should happen, not things that should answer. Each cron job, backup or timer pings a unique URL when it finishes; if a ping is later than the grace time you set, you get an alert. Schedules can be a simple period or a cron expression. The hosted Hobbyist plan is free for 20 jobs, and Business is $20 a month for 100 jobs, or $16 a month billed yearly. The software is open source (BSD-3-Clause) if you'd rather self-host it with Python, Django and PostgreSQL or MySQL. It doesn't check websites, so pair it with an uptime monitor.
- Pricing
- Hobbyist free (20 jobs); Supporter $5 per month; Business $20 per month (100 jobs) or $16 per month yearly; Business Plus $80 per month (1,000 jobs) or $64 per month yearly (checked November 15, 2026, source)
- License
- Open source (BSD-3-Clause); the hosted service is paid above the free tier
- Deployment
- SaaS or self-hosted
- Runs on
- Any job that can make an HTTP request or send an email; self-hosting needs Python 3.12+, Django and PostgreSQL, MySQL or MariaDB
Pros
- Catches silent failures that uptime checks miss
- Free for 20 jobs; open source and self-hostable
- Cron-expression schedules with grace periods
Cons
- Not a website or port monitor
- Self-hosting needs Python, Django, a database and SMTP
Not tested by us yet: based on the vendor's documentation.
Gatus
Best for: health checks and a status page defined as code
Gatus is a developer-oriented health dashboard you configure in YAML. Each endpoint gets conditions, such as status code, response body, response time or time left before the certificate expires, and it checks HTTP, ICMP, TCP, DNS, SSH, TLS and STARTTLS targets, with SSH tunnels for internal services. Because the checks are a config file, you review them in Git like any other change. It's Apache-2.0 and the project ships a Docker image. By default it stores results in memory, which a restart wipes, so set SQLite or PostgreSQL storage. A managed version at gatus.io starts at $10 a month.
- Pricing
- Self-hosted free; managed gatus.io Standard $10, Premium $20 and Business $50 per month (20% off yearly, 7-day trial) (checked November 15, 2026, source)
- License
- Open source (Apache-2.0)
- Deployment
- Self-hosted, or managed at gatus.io
- Runs on
- Self-hosted with Docker or from source; checks HTTP, ICMP, TCP, DNS, SSH, TLS and STARTTLS endpoints
Pros
- Checks defined as code with precise conditions
- Many protocols, including DNS, SSH and certificate expiry
- Free and open source, with an optional managed version
Cons
- Checks are added in YAML, not through a web form
- Default in-memory storage loses history on restart
Not tested by us yet: based on the vendor's documentation.
Pingdom
Best for: transaction checks plus real user monitoring
Pingdom, from SolarWinds, pairs uptime checks with advanced checks for multi-step transactions and with real user monitoring of page performance. Synthetic monitoring starts at $18 a month ($16.50 billed annually) for 10 uptime checks, 1 advanced check and 50 SMS, and real user monitoring starts at $18 a month for 100,000 pageviews. There's a free trial but no free plan. It fits when you need to watch a login or checkout flow and see how real visitors experience page speed. For plain up/down checks on many sites, it costs more per check than UptimeRobot or StatusCake.
- Pricing
- Synthetic monitoring from $18 per month ($16.50 billed annually) for 10 uptime checks, 1 advanced check and 50 SMS; Real User Monitoring from $18 per month for 100,000 pageviews; free trial (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; checks public websites and endpoints
Pros
- Transaction checks for multi-step flows
- Real user monitoring alongside synthetic checks
- SMS alerts included
Cons
- No free plan
- Only 10 uptime checks at the entry tier
- Hosted only
Not tested by us yet: based on the vendor's documentation.
StatusCake
Best for: a free plan plus inexpensive 1-minute checks
StatusCake's free plan covers 10 uptime monitors at 5-minute intervals. The Superior plan raises that to 100 monitors checked every minute for $20.41 a month billed annually ($24.49 monthly), and Business to 300 monitors at 30-second intervals for $66.66 a month billed annually ($79.99 monthly). It's a straightforward hosted uptime monitor with room to grow. The free plan has fewer monitors than UptimeRobot's; StatusCake's pricing page describes it as for individuals and small businesses.
- Pricing
- Free (for individuals and small businesses, per its pricing page): 10 uptime monitors at 5-minute intervals; Superior $20.41 per month billed annually, $24.49 monthly (100 monitors, 1-minute checks); Business $66.66 per month billed annually, $79.99 monthly (300 monitors, 30-second checks); Enterprise custom (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; checks public websites and endpoints
Pros
- Free plan with 10 monitors
- 100 monitors at 1-minute checks on the first paid tier
- Enterprise plan with constant checking
Cons
- Fewer free monitors than UptimeRobot
- Free checks run every 5 minutes
Not tested by us yet: based on the vendor's documentation.
Checkly
Best for: API and browser checks for developers
Checkly goes beyond up or down. Alongside uptime monitors it runs API checks and scripted browser checks, so you can test that a login works or an API returns the right data. The free Hobby plan includes 10 uptime monitors checked every 2 minutes, 10,000 API check runs and 1,000 browser check runs a month, with a hard cap. Starter is $24 a month and Team $64 a month, billed annually, with uptime checks every minute or every 30 seconds. Choose it when uptime depends on application behavior. For simple website pings, it's more than you need.
- Pricing
- Hobby free (10 uptime monitors, 10,000 API and 1,000 browser check runs a month); Starter $24 and Team $64 per month billed annually; Enterprise custom (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; checks websites and APIs
Pros
- API and browser checks, not just pings
- Free Hobby plan
- 30-second uptime checks on Team
Cons
- Free plan is hard-capped
- Hobby uptime checks run every 2 minutes
- More setup than a simple ping monitor
Not tested by us yet: based on the vendor's documentation.
Cronitor
Best for: hosted cron monitoring and uptime checks on one bill
Cronitor covers cron jobs, heartbeats, websites and APIs on every plan, and includes a status page. The free Hacker plan gives you 5 monitors with email and Slack alerts and a basic status page. Business is priced per item: $2 per monitor and $5 per user each month, with SMS and on-call alerts and 12 months of data retention. That pricing suits a handful of important jobs and endpoints watched from one place. With dozens of monitors, Healthchecks.io for jobs and UptimeRobot for sites will cost less.
- Pricing
- Hacker free (5 monitors); Business $2 per monitor per month plus $5 per user per month (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; monitors cron jobs, heartbeats, websites and APIs
Pros
- Cron, heartbeat, website and API monitoring together
- Simple per-monitor pricing
- Status page on the free plan
Cons
- Only 5 free monitors
- Per-monitor pricing grows quickly with large fleets
Not tested by us yet: based on the vendor's documentation.
Oh Dear
Best for: agencies watching the overall health of many websites
Oh Dear monitors whole websites, not just whether they answer: uptime, ping and TCP ports, SSL certificate health, broken links and mixed content, scheduled tasks, DNS records and blocklists, Lighthouse performance and domain expiry. Every plan includes every feature, and plans differ only in the number of sites, from €15 a month for 2 sites to €399 for 200. There's a 10-day free trial and no free plan. It suits agencies and teams responsible for many client websites. If you only need up/down alerts for a few servers, it's more than you need.
- Pricing
- From €15 per month (Solo, 2 sites) to €399 (Scale, 200 sites), every feature on every plan; 10-day free trial (checked November 15, 2026, source)
- License
- Proprietary
- Deployment
- SaaS
- Runs on
- Hosted; monitors websites and servers from outside
Pros
- Broad website checks in one place
- Every feature on every plan
- Cron monitoring included
Cons
- No free plan
- Priced per site, in euros
Not tested by us yet: based on the vendor's documentation.
How we picked
An uptime monitor has one job: tell you quickly, from outside your network, when a website, API, port or scheduled job stops working. We ranked the tools on:
- The free or entry plan: how many monitors, how often, and whether commercial use is allowed.
- Check types: HTTP and keyword checks, TCP ports, ping, DNS, certificates, and heartbeats for cron jobs.
- Alerting: channels, escalation and status pages.
- Price as you grow, from each vendor's own pricing page, checked on November 15, 2026.
- Self-hosting: whether you can run it yourself, and what that takes.
We ran Uptime Kuma on a Debian 12.15 test server on November 14, 2026 (UTC), then removed it. Everything else is ranked from its documentation and pricing pages, listed under Sources. Nobody paid to be listed.
What we saw running Uptime Kuma
We installed Uptime Kuma 2.5.11 the non-Docker way, on Node.js 24.22.3 with SQLite, bound to localhost on a 2 GB server. The dependency install took about 7 seconds and the app plus its modules came to 278 MB. We then added three monitors at 20-second intervals:
| Monitor | Type | Result |
|---|---|---|
| A public HTTPS website | HTTP | Up: 200 OK in about 53 to 66 ms |
| SSH on the server itself | TCP port 22 | Up |
| A closed local port | HTTP | Down, with the error connect ECONNREFUSED |
It used about 190 MB of memory while running. The lesson for any self-hosted monitor: this one was watching the server it ran on, which is fine for a test and useless in an outage. Put it somewhere else.
What should you look for in an uptime monitor?
The check interval decides how long an outage can run before you hear about it. Free plans usually check every 5 minutes; paid plans go to 1 minute, 30 seconds or faster. Next, look at what a check can test. A plain HTTP 200 can hide a broken page, so keyword or response-body checks catch more, and TCP, ping and DNS checks cover services that aren't websites. Certificate and domain expiry checks prevent an entirely avoidable kind of outage. Scheduled jobs need the reverse: a heartbeat monitor that alerts when a job fails to report in. Then check alerting: which channels you get on the free plan (UptimeRobot's allows 5 integrations), whether alerts escalate when nobody answers, and whether you get a public status page. Finally, read the license terms of free plans. UptimeRobot's free plan allows commercial use, while Better Stack's is meant for personal projects.
Free plans compared
| Tool | Free monitors | Free check interval | Notes |
|---|---|---|---|
| UptimeRobot | 50 | 5 minutes | Personal and commercial use, 1 status page |
| Better Stack | 10 monitors and heartbeats | Not stated | Personal projects; includes 3 GB of logs for 3 days |
| StatusCake | 10 | 5 minutes | For individuals and small businesses (its wording) |
| Checkly | 10 uptime monitors | 2 minutes | Plus 10,000 API and 1,000 browser check runs, hard-capped |
| Healthchecks.io | 20 jobs | Your schedule | Cron and heartbeat monitoring only |
| Cronitor | 5 | 5 minutes | Jobs, heartbeats and website checks |
| Uptime Kuma, Gatus | Unlimited | You choose | Self-hosted; you run the server |
| Pingdom, Oh Dear | None | None | Free trial only |
Self-hosted or hosted?
Hosted monitors check from the vendor's network, so they see what your visitors see and keep working when your whole provider is down. They cost little or nothing for a few sites. Self-hosted monitors (Uptime Kuma, Gatus, Healthchecks.io) are free, private and unlimited, and good for internal services a hosted checker can't reach. Their weakness is the vantage point: they check from wherever you install them. If you self-host, put the monitor on a small server with a different provider, and consider one hosted check on the monitor itself so you hear about it if it stops.
Do you need uptime monitoring at all?
If anyone depends on your site or API, yes: without it, you learn about outages from users. Server monitoring doesn't replace it, because an agent on a server can't report that the server is unreachable. Pair it with server monitoring for the inside view, and add a heartbeat monitor for backups and other cron jobs, because a job that never runs produces no error for anyone to see.
How we checked this: Debian 12.15 test server (2 vCPU, 2 GB RAM) on 14 Nov 2026 (UTC): Uptime Kuma 2.5.11 from its official 2.5.11 source tag, on Node.js 24.22.3 from checksum-verified nodejs.org binaries, with SQLite and bound to localhost; we added HTTPS, TCP and failing-port monitors and read their results, then uninstalled everything. The hosted services, Gatus and Healthchecks.io are researched from vendor documentation and pricing pages checked on 15 Nov 2026, not run on our servers. Anything we didn't run ourselves comes from the official documentation listed under Sources.
Check which CVEs are really open on your server
HostCrust Patch Proof compares every installed package with your distribution's own security data. It counts fixes your distro backported and checks the kernel that is actually running, not just the one on disk. Read-only, no root, free.
FAQ
What is the best free uptime monitor?
For a hosted service, UptimeRobot: its free plan has 50 monitors checked every 5 minutes, and its pricing page says it's good for personal and commercial projects. To self-host, Uptime Kuma is free and open source with no monitor limits.
How often should an uptime monitor check my site?
Every minute is a good default for a business site; every 5 minutes is fine for personal projects. Faster checks cost more: UptimeRobot's Scale plan checks every 15 seconds, and Checkly's Enterprise plan as often as every second.
Can I self-host uptime monitoring?
Yes. Uptime Kuma, Gatus and Healthchecks.io are open source. Run the monitor on a different server, ideally with a different provider, from the ones it watches. Otherwise one outage takes down both the site and the monitor that should warn you.
How do I monitor cron jobs?
Use a heartbeat (dead man's switch) monitor: the job pings a URL when it finishes, and you're alerted if the ping is late. Add && curl -fsS -m 10 <your ping URL> to the end of the cron line. Healthchecks.io, Cronitor, Better Stack heartbeats and Uptime Kuma's push monitors all work this way.
Is UptimeRobot free for commercial use?
Its pricing page describes the free plan as good for personal and commercial projects. The free plan checks every 5 minutes and is limited to 5 notification integrations. Better Stack's free plan, by contrast, is labeled for personal projects.
Sources
- UptimeRobot pricing
- Better Stack pricing
- Uptime Kuma on GitHub (features, MIT license, install requirements)
- Pingdom pricing
- StatusCake pricing
- Healthchecks.io pricing
- Healthchecks on GitHub (BSD-3-Clause, self-hosting)
- Gatus README (endpoint types, conditions, storage)
- Gatus managed pricing
- Cronitor pricing
- Checkly pricing
- Oh Dear pricing