If you run your own VPS, you need someone else’s server watching yours. UptimeRobot, Better Uptime and HetrixTools are the three names that come up in every monitoring thread, and they take different approaches: UptimeRobot keeps checks simple, Better Uptime (from Better Stack) adds incident management, and HetrixTools packs the free tier with extras. This UptimeRobot vs Better Uptime vs HetrixTools comparison covers free limits, check intervals, heartbeat monitoring and alerting, with notes written for VPS owners.
The short version
- UptimeRobot is the simplest option and the most generous with free monitor counts, but free checks run every 5 minutes and heartbeat plus SSL monitoring sit behind paid plans.
- Better Uptime is the best pick if you work in a team. You get on-call schedules, incident timelines and polished status pages, but the free tier holds fewer monitors.
- HetrixTools gives you the most free value: 1-minute checks, a server metrics agent, SSL and blacklist monitoring on the free plan. The catch is you must log in periodically to keep a free account active.
Side-by-side comparison
| UptimeRobot | Better Uptime | HetrixTools | |
|---|---|---|---|
| Free monitors | 50 | About 10 monitors + 10 heartbeats | 15 uptime + 15 server |
| Free check interval | 5 minutes | 3 minutes | 1 minute |
| Heartbeat / cron monitoring | Paid plans only | Free (10 heartbeats) | Claimed free on their own site; some third-party comparisons disagree, so verify in their docs |
| SSL and domain expiry alerts | Paid plans | Included | Free |
| Server metrics agent | No | No | Yes (Linux, Windows, macOS) |
| Blacklist / reputation checks | No | No | Yes |
| Automatic diagnostics on failure | Traceroute in incident details | Failure screenshots | Automatic ping and MTR from failing locations |
| Status pages | 1 basic page on free | 1 free page | Multiple on free |
| Paid entry point | Roughly single-digit USD/month (check official pricing) | Higher team pricing (check official pricing) | Roughly around $10/month (check official pricing) |
Free-tier numbers change often. The figures above come from each vendor’s help docs and independent comparisons; confirm the current limits on the official pricing pages before you decide.
UptimeRobot: the focused checker
UptimeRobot does one job: ping your sites and tell you when they stop answering. HTTP, keyword, ping, port and heartbeat monitor types cover the basics, the dashboard takes minutes to learn, and the REST API is mature enough to create monitors from a deploy script. With 50 free monitors it comfortably covers a personal VPS fleet or a small agency’s client sites.
The tradeoffs are clear. Five-minute checks on the free plan mean up to five minutes of blind spot between checks. Heartbeat monitoring and SSL expiry alerts need a paid plan, and the failure diagnostics are thinner than the other two tools here. Pick UptimeRobot when you want the simplest setup and the most monitors for zero cost, and you do not need cron-job monitoring or cert alerts for free.
Better Uptime: uptime checks plus incident workflow
Better Uptime treats an outage as an incident, not just an alert. When a check fails you get an incident timeline, on-call schedules that page the right person, runbooks, and screenshots of the failed page. Status pages look polished out of the box, and Slack integration works on the free plan. Free checks run every 3 minutes, and you get a handful of heartbeat monitors included.
The free tier is the tightest of the three, with roughly ten monitors, so it fills up fast if you watch many sites. The full incident-management value also sits on the paid team plan, which costs more than the entry plans of the other two. Pick Better Uptime when downtime involves a team: client sites, a small SaaS, or any setup where the alert needs to turn into an assigned, tracked incident.
HetrixTools: the feature-packed budget pick
HetrixTools bundles uptime monitoring, server monitoring and blacklist monitoring into one account. The free tier checks every minute, covers 15 uptime monitors plus 15 server monitors, and includes an installable agent for Linux, Windows and macOS that reports CPU, memory and disk. SSL and domain expiry alerts are free, blacklist checks scan your IPs against dozens of reputation lists, and when a monitor fails the tool runs ping and MTR traces from the failing locations automatically.
Two honest caveats. First, free accounts require periodic logins to stay active, so set a calendar reminder or you will lose monitoring silently. Second, heartbeat support is advertised on HetrixTools’ own comparison page, but at least one independent comparison says it is not offered, so check their current docs before relying on it. Pick HetrixTools when you want the most monitoring per free account and you are comfortable doing your own diagnostics with the extra data it gives you.
How heartbeat monitoring actually works on your VPS
Regular checks ask “is the site up?” Heartbeat monitoring asks the reverse: your server tells the monitor “I am still alive.” You create a heartbeat monitor, the tool gives you a unique ping URL, and your cron job curls that URL when it finishes. If no ping arrives within the expected window plus a grace period, you get alerted. This catches the failures normal uptime checks miss: a site can be up while your backups quietly stopped running three months ago.
A typical setup for a nightly backup looks like this:
# Backup runs at 02:30; ping the heartbeat URL only if the backup succeeds
30 2 * * * /usr/local/bin/nightly-backup.sh && curl -fsS --max-time 10 "https://HEARTBEAT-URL-FROM-YOUR-TOOL" >/dev/nullReplace the placeholder with the real ping URL from your monitor, then set the expected interval (24 hours here) and a grace period of 30 to 60 minutes so a slow backup does not page you. Use the same pattern for cert renewals, queue workers, and any cron job whose silence would cost you.
Setting up alerts that actually reach you
Email alerts are free on all three tools and fine as a baseline, but email is also the channel most likely to fail silently when your own mail server has the problem. Add a second channel: Slack or Telegram webhooks are usually available on free or cheap plans, and all three support generic webhooks you can point at your own notification bot. SMS alerts almost always cost extra credits on paid plans, so treat SMS as a paid upgrade, not a default.
Two habits matter more than the channel. First, add at least one contact method that does not depend on the server being monitored. Second, use maintenance windows during deploys and migrations. Every false 2 a.m. page trains you to ignore the real one.
Which one should you pick?
- Many sites, zero budget: UptimeRobot’s 50 free monitors, or HetrixTools if you prefer 1-minute checks and the extra features.
- A team with client sites: Better Uptime. The incident timeline and on-call rotation pay for themselves the first time something breaks at night.
- Cron jobs that must run: pick the tool with free heartbeat monitoring, but verify the current docs first since vendors disagree on this feature.
- You want server metrics too: HetrixTools’ agent, or pair any of the three with Netdata or Prometheus on your own VPS. External checks cannot see a disk filling up.
One more option: Uptime Kuma is a solid self-hosted alternative if you would rather run the monitor on your own infrastructure, but then you own the monitor’s uptime too.
FAQ
Can free uptime monitoring replace server alerting?
No. External checks only see what the network sees: HTTP responses, open ports, ping replies. They cannot warn you about a disk at 95 percent or a runaway process. Pair any of these tools with on-server monitoring for CPU, memory and disk.
How fast will I know about an outage?
Roughly the check interval plus alert delivery time. A 1-minute interval catches outages faster than a 5-minute one, which is why the interval is the main thing you trade off between free tiers.
Do I need heartbeat monitoring?
If you run cron jobs that must run, yes. Backups, certificate renewals and queue workers fail silently by default, and only a heartbeat tells you the job itself stopped.
Start with the free tier of the tool that matches your biggest pain: check intervals, incident workflow, or feature depth. Then verify the current limits on its pricing page before you commit.