StatusPulse vs Uptime Kuma
Uptime Kuma is the free, open-source monitor most engineers try first, and it's genuinely good. StatusPulse is the managed option. Checks run outside your infrastructure, customers can subscribe to your status page, and on-call is built in. Nobody on your team has to keep the monitor itself alive. Here's where each one wins, including where Kuma beats us.
Uptime Kuma facts last verified 2026-10-27 against its GitHub repository and wiki (release 2.5.8). It moves fast — check its changelog before deciding.
Self-hosted vs managed, in three cards
$0
Kuma's licence (MIT)
No monitor limits and no subscription. The cost is a server and the hours to install, upgrade, back up and watch it. StatusPulse is free for 10 probes, 3 status pages; Pro is $19/mo.
Outside
Where the checks run
Kuma runs wherever you install it, often next to the app it watches. One outage can take down both. StatusPulse checks from Azure West Europe (plus a US location on Business), outside your infrastructure.
0
Servers for you to run
No upgrades, backups or TLS renewals. Kuma 2.x removed its JSON backup. Copying the data directory is now the supported way to back it up.
What "free" costs
The licence is free; running it isn't. These are our assumptions for a small team. Put in your own server price and hourly cost.
| StatusPulse | Uptime Kuma, self-hosted | |
|---|---|---|
| Software | Free plan, or Pro $19 / mo | Free (MIT) |
| Server | Included | A small VPS — about $5–10 / mo |
| Setup | Minutes, with the probe wizard | 2–4 hours: Docker, a reverse proxy with TLS, notifications, a custom domain for the status page |
| Keeping it running | Our job | 1–2 hours a month: upgrades (several releases a month), backups, disk, certificates |
| Who watches the monitor? | We do — three independent layers, public at status.statuspulse.ai | You need a second monitor somewhere else |
At a loaded engineering cost of $75 an hour, an hour and a half a month is about $110. That's before the server. If your team already runs servers, or your time is genuinely free, Kuma wins on cost. If not, $19 is usually the cheaper line.
What a Kuma install involves
The quick start is one container. Kuma's
README
gives it as docker run -d --restart=always -p 3001:3001 -v uptime-kuma:/app/data
--name uptime-kuma louislam/uptime-kuma:2. That gets you a dashboard on port
3001 on every network interface. Production needs a few more pieces.
- Storage on a local disk. The README warns that network file systems such as NFS aren't supported. The data volume has to live on the host or a local volume.
- A reverse proxy with TLS if the status page is public. The wiki's install guide covers it. You also own the certificate renewals.
- An update routine. Releases land several times a month. You pull the new image and restart, following the wiki's update page.
- A second monitor for the monitor. If the host or its network goes down, Kuma's alerts go down with it.
None of this is hard for an SRE. It is recurring work, and it lands on the same people who get paged when production breaks.
Kuma concepts and their StatusPulse names
- HTTP(s) and keyword monitors become HTTP probes. The keyword is the "body contains" assertion. Kuma's JSON query has no direct twin; match a substring of the response instead.
- Ping, TCP and DNS monitors map one to one. They need Starter or above.
- Push monitors become Heartbeat probes. Your job calls a unique URL; silence past the grace period marks it Down.
- Docker container monitors have no equivalent. StatusPulse can't see inside your network.
- Intervals: Kuma allows 20 seconds. StatusPulse's floor is 5 minutes on Free, 2 on Starter, 1 on Pro and 30 seconds on Business.
- A status page on its own domain is a custom domain on Pro and up. The certificate is issued and renewed for you.
- Maintenance windows become scheduled maintenance (Starter and up). The status page lists the window as upcoming, then opens and closes it on time.
What each one does
| Feature | StatusPulse | Uptime Kuma |
|---|---|---|
| Monitoring | ||
| HTTP(S), keyword, ping, TCP, DNS, heartbeat | Yes | Yes (plus HTTP JSON query) |
| Monitor types in total | 11 | About 30 |
| Databases | Postgres, MySQL, SQL Server, Redis — with your own query (Business) | Postgres, MySQL / MariaDB, SQL Server, Oracle, MongoDB, Redis |
| gRPC | Health Check protocol: Check + Watch, mTLS (Pro+) | Keyword match on a gRPC call |
| WebSocket | Connect, subprotocol and round-trip latency (Pro+) | Upgrade check |
| SMTP → IMAP round trip, with SPF / DKIM / DMARC results (Pro+) | SMTP check | |
| Things inside your network (Docker, PM2, system services, MQTT, Kafka, RabbitMQ, SNMP) | No — we can't reach them | Yes |
| Fastest check interval | 30 seconds (Business) | 20 seconds recommended, lower possible |
| Checks from more than one region | Business — EU + US | Not built in — run more instances |
| Status pages | ||
| Several status pages, custom domain | Yes; custom domain with automatic TLS on Pro+ | Yes; you set up DNS and a reverse proxy |
| Visitors subscribe to updates | Yes — subscribe button, you approve | No (requested since 2021) |
| Incident timeline with updates, post-mortems | Yes | Pinned incidents and a basic history |
| Scheduled maintenance | Yes | Yes, including recurring windows |
| Branding, white-label | Branding Pro+, white-label Business | Basic |
| Alerting and on-call | ||
| Notification channels | Email, SMS, Slack, Teams, signed webhooks | 110 integrations, 20+ SMS gateways |
| On-call schedules and escalation | Pro+ | No — hands off to PagerDuty or Opsgenie |
| AI-drafted incident updates | Pro+ | No |
| Scheduled PDF uptime / SLA reports | Starter+ | No |
| Team and operations | ||
| Users | Up to 50, four roles; SSO on Business | One user (multi-user is in an unreleased version) |
| Backups and upgrades | Ours | Yours — copy the data directory; upgrade by hand |
| Where your data lives | EU (Azure West Europe) | Wherever you host it |
| Hosted version | It is the hosted version | None official |
Where each one fits
Pick Uptime Kuma if…
- Most of what you watch is inside your network. Containers, queues and internal services a SaaS can't reach.
- You already run servers and one more container is no extra work for you.
- Nobody outside your team reads the status page, so subscribers and incident timelines don't matter.
- It's a homelab or side project and the budget is zero.
Pick StatusPulse if…
- Customers read your status page and want to subscribe to updates instead of refreshing it.
- The monitor must survive your outage. Checks run outside your cloud: from the EU, plus the US on Business.
- Someone has to be woken up. On-call rotations and escalation page by email and SMS.
- Your team has better things to run than the monitoring server.
Or use both
Plenty of teams keep Kuma inside the network for the services only it can reach. StatusPulse handles the outside view: external checks of public endpoints, the customer-facing status page with subscribers, and on-call. The free plan is enough to try that split.
Switching from Uptime Kuma
- Sign up for the free plan and add your public-facing monitors, with the same URLs and intervals.
- On a paid plan, I'll move them for you. Kuma 2.x has no export. Email support@statuspulse.ai a list of your monitors (name, URL, interval). They're recreated within one business day.
- Run both for a week. Then publish the StatusPulse status page and point your status domain at it.
Keep Kuma inside, add the outside view
Free plan: 10 probes, 3 status pages. No card, no server.