UptimeFlare vs UptimeRobot

Compare operating responsibility first

Decision	UptimeFlare at the reviewed commit	UptimeRobot current public plans
Who operates it?	You maintain the repository, Cloudflare resources and secrets	A managed monitoring service
Check cadence	Terraform configures a one-minute Cron trigger	Free five minutes; Solo one minute; faster paid intervals have an official-source conflict
Basic overlap	HTTP/HTTPS/TCP checks, status codes and keyword conditions	Basic uptime checks plus plan-specific advanced types
Status page	Source-owned page with charts, incidents and maintenance notices	Hosted pages with limits and controls by plan
Alerts	Configure a webhook; optional external gateways have their own terms	Managed destinations and plan-specific integrations; SMS/voice and seats need budgeting
Cost evidence	Small-workload Free eligibility; capacity and total bill unmeasured	Dated public pricing; currency and interval caveats in the pricing guide

The UptimeFlare project page includes the actual README, Apache-2.0 license, upstream application screenshot and configuration evidence. Our review covers source declarations and documented workflows. It is not a completed deployment or independent alert-reliability test.

Cadence is only part of detection time

UptimeFlare's production Terraform uses a one-minute Cron, while its example monitor config has timeouts and an optional notification grace period. A check schedule alone does not determine when a human receives an alert. Verify failed-check handling, repeated incidents and recovery messages using your intended notification provider.

UptimeRobot's Free and Solo intervals are consistent across the checked sources. Team and Scale are not: the pricing page advertises 30 and 15 seconds, but the help article says 60 and 30 seconds. Read the pricing review and confirm the account entitlement when faster cadence matters.

Do not assume every monitor type transfers

At the pinned revision, UptimeFlare supports HTTP request methods, headers and bodies, accepted status codes, keyword conditions and TCP ports. The README lists SSL-certificate checks and ICMP through a proxy as unfinished items. Those limitations matter if your existing setup depends on expiry notices or ping semantics. A successful HTTP request cannot stand in for a tested certificate-expiry workflow.

Regional checks also need scrutiny. Terraform provisions a SQLite RemoteChecker Durable Object, used for configured worker:// checks. Globalping and HTTP proxy modes depend on separate services. Do not treat an advertised number of cities as proof that every region is configured, available or included without cost in your deployment.

Free infrastructure has measurable conditions

UptimeFlare writes compacted monitor state to D1, and its Pages API reads that state for the public dashboard. Monitor count, scheduled CPU, writes and page traffic all affect the budget. The upstream claim of 50 one-minute checks has not been established as a free-tier capacity benchmark here.

Choose UptimeFlare when owning a small configurable monitoring system is useful and someone will maintain it. Choose a managed service when delegated operations, collaboration, documented paid controls or contractual support matter more. For a broader shortlist, see UptimeRobot alternatives.

Validate a parallel pilot before switching

Follow the self-hosting guide with one controlled endpoint. Compare healthy, unavailable, wrong-keyword and recovery results. Confirm that an alert reaches an authorized destination, verify the public page exposes no credentials, and measure usage. Keep the existing monitors until the replacement passes your own acceptance checks. Historical UptimeRobot data has not been shown to import into UptimeFlare automatically.