Know what needs your attention
Monitor websites, APIs and background jobs. Catch expiring certificates and domains. Bring the evidence, alerts and incident updates together in Uplapse.
Start with one important endpoint. Find your plan.
Different failures need different checks. Build coverage around the requests and jobs your customers depend on.
Check availability, response time and page content. Investigate results by location to understand where a failure is observed.
Explore uptime checksConfigure requests and assert the status and content you expect. Catch a broken response even when the server answers.
Explore API checksRun a sequence of requests, extract values and pass them to later steps. Test the connections between your endpoints.
Explore journeysLet scheduled jobs report completion. Detect a missing backup or import even when the server itself stays healthy.
Explore heartbeat checksCombine these with focused checks at the network, content and delivery layers. Browse all features.
Use a narrow check to answer a specific diagnostic question. Pair it with a customer-facing check when you need to know whether the whole service works.
Require a stable phrase in the response, even when the page returns HTTP 200.
Check the record answer your site, mail or certificate policy depends on.
Observe host reachability; the method may be ICMP or a TCP fallback.
Check that a chosen TCP listener accepts a connection.
Check a small protocol reply, with an optional request payload.
Watch the administrative access path without treating it as a login test.
Confirm that a marked enquiry reaches the configured mailbox.
A port can be open while a website serves an error. A form can accept a POST while the enquiry never arrives. Check the behavior that matters to the person using your service.
Availability checks tell you what works now. Dependency checks help uncover the renewal deadlines and configuration changes that could cause the next failure.
Watch certificate expiry and check certificate details so a broken renewal can be investigated before access is affected.
How it worksTrack registry expiry information through RDAP. Give domain renewal its own signal, independent of server health.
How it worksCheck published email authentication records and the DMARC policy strength you expect, with an explicit selector for DKIM.
How it worksWatch public certificate records for your domain, including certificates that are never presented by your own server.
How it worksValues shown on the cards are examples.
The SSL check reads the certificate a site presents: who issued it, the names it covers, whether browsers trust the chain and how long is left. Uplapse runs it on a schedule and alerts you at the threshold you choose, so a renewal that quietly stopped is caught while there is still time.
12 days leftshop.example.com, issued by Let's Encrypt (R12)
Send alerts to your team's tools. Escalate unacknowledged incidents, stop escalation when someone takes ownership, and suppress notifications during planned maintenance.
Explore incident alertingPublish service status, incident updates and resolutions. Give customers a place to follow progress and subscribe to changes while your team investigates.
Explore status pagesConnect monitors to service-level objectives. Review availability, latency and remaining error budget, then use burn rate to understand how quickly that budget is being spent.
Explore SLO monitoringBring together security-header and DNS changes, CSP violations and DMARC reports. These observations help investigate configuration and delivery problems; they do not establish that a site is secure.
Read about the reporting approachRead response timing alongside check outcomes. A request that passes eventually may still miss your latency objective. Compare the same operation across time and locations, then use application evidence to investigate the cause.
Interpret response timeDocument which services a maintenance window covers and when normal alerting should resume. Keep observations available, verify fresh results after the work and review how planned downtime is treated in reliability reports.
Use the maintenance checklistThese suggested setups connect a business need to a small set of meaningful checks. Adapt the endpoints, frequency and response process to your own services.
A public health endpoint can stay green while authenticated requests fail. Combine a basic availability check with an HTTP journey that signs in and reads a protected resource. Add a heartbeat for the jobs that keep customer data current.
The page loads, but an enquiry never arrives or a renewal deadline has no owner. Check page content and contact-form delivery, then add certificate and domain expiry monitoring. Give each alert a named recipient and keep a record of planned changes.
A reachable homepage does not prove product data is current. Check a dedicated catalog item, track response timing and have scheduled imports report completion. Use test data for synthetic requests and account for any records or messages they create.
Set useful certificate expiry warnings, choose identity assertions and verify that a renewed certificate is actually served to your users.
Read guideTest contact forms from submission to mailbox receipt. Use unique message markers, delivery windows and controlled test data to detect lost enquiries.
Read guideWrite useful incident updates with affected services, customer impact, investigation progress and a clear next update time.
Read guideBuild a useful website monitoring setup with endpoint checks, content assertions, expiry alerts and a clear response plan.
Read guideLearn how status codes, response content, authentication and timeouts work together to make API monitoring reflect real service health.
Read guideUse heartbeat monitoring for cron jobs, backups and imports. Decide when to send a ping and how to distinguish completion from useful work.
Read guideUnderstand what your checks can establish, then choose the coverage you need.
More questions about UplapseThat depends on the agreement and its measurement rules. A maintenance label or muted notification does not automatically exclude downtime. Record the scope and timing, apply the stated exclusion policy and retain evidence of what customers experienced.
Understand maintenance exclusionsScheduled checks repeat the same request and expectations over time, including when nobody is looking. They create a history and can trigger an alert. Their coverage still depends on what you configure: a homepage check does not establish that every feature works.
Choose your first checksYes, when the monitor includes an assertion that the broken response fails. Set a stable keyword or content expectation in addition to the expected status. Choose a value that distinguishes the correct response from a login page, maintenance document or application error.
Learn about response assertionsJourneys execute HTTP requests and can pass extracted values between steps. They do not render the page or run its JavaScript. They are useful for authentication and API workflows; browser testing remains useful for client-side behavior and interaction.
Understand HTTP journeysMaintenance settings let you suppress expected notification noise while retaining observations. Scope the work carefully, set an end time and verify recovery afterwards. Review SLO exclusions separately: muting an alert and excluding downtime from a report are different decisions.
Plan a maintenance windowIt means the observed check met its configured expectations. Region, freshness, sampling interval and the chosen endpoint all affect what that proves. Combine relevant checks and inspect missing or conflicting observations before relying on a single summary.
Interpret regional resultsYour main API, the nightly import, or the certificate due to renew. Make that failure visible, then build from there.
