How We Monitor Website Status

At IsItDown, we use a multi-layered verification system to ensure our status reports are as accurate as possible. We don't just rely on a single "ping"; we perform deep analysis to distinguish between a global outage and a local connection issue.

Step 1: DNS Resolution Verification

Before any HTTP request is made, we verify that the domain's DNS records are resolving correctly. If DNS fails, it often indicates a registry-level issue or a major configuration error.

Step 2: HTTP HEAD Request

We initiate an HTTP HEAD request. This allows us to check the server's response headers (and the HTTP status code) without downloading the entire page content, making our checks faster and less intrusive.

Step 3: Fallback GET Request

Some servers or CDNs block HEAD requests. If the HEAD request fails or returns an ambiguous status code, we fall back to a standard GET request, specifically fetching only the first few bytes (Range: 0-500) to confirm if the server is serving content.

Step 4: Automatic Retries

Network hiccups happen. If a check fails, we wait for 5–10 seconds and retry. We never mark a site as "Down" based on a single failed connection.

Step 5: Status Confidence Scoring

We combine our technical checks with real-time user reports.

  • Operational: Our technical check passed.
  • Degraded: Slow response times or mixed technical signals.
  • Possible Outage: Technical checks failing, but low user reports.
  • Confirmed Outage: Technical checks failing AND multiple user reports received.

What We Inspect During a Check

Our status workflow covers the most common layers that affect availability: DNS resolution, HTTP response code, latency, redirect behavior, and retry stability. This helps us separate temporary blips from genuine outages.

Common Error Examples

  • HTTP 503 Service Unavailable: The server is reachable but overloaded or under maintenance.
  • HTTP 404 Not Found: The server is online, but the specific URL is missing or broken.
  • Timeout: The server took too long to respond, often due to overload or network instability.
  • SSL error: The site may be reachable, but certificate issues can prevent a secure connection.

Frequently Asked Questions

How do you decide whether a site is down?
We combine DNS resolution, HTTP checks, fallback requests, and retry logic before confirming an outage.
Why can a site be up for you but not for me?
That usually indicates a local DNS, ISP, VPN, firewall, or regional routing issue rather than a global outage.
What do status codes like 503 and 404 mean?
503 usually means the server is temporarily unavailable, while 404 means the server responded but the requested page was not found.
Do you test from multiple regions?
We describe our checks as globally aware and account for regional access problems when interpreting results and reports.