DDoS or a connection fault? What an outage actually tells you
A slow website is evidence of a problem, not proof of an attack. Separate service availability, access-link congestion and local network faults.

A website that stops responding gives you one fact: the service is unavailable from your position. It does not tell you why. A distributed denial-of-service attack is one possible explanation, alongside failed software, an overloaded database, DNS problems or a broken route.
Unavailable does not mean breached. Keep the availability diagnosis separate from any claim of data theft.
Availability is a different question from data theft
A DDoS attack attempts to exhaust a service or the infrastructure around it using traffic from many sources. The resource under pressure might be network capacity, connection handling or application work. A successful attack can prevent legitimate users from receiving a response.
That does not, by itself, prove that customer records were stolen or that the server was taken over. An availability incident and a confidentiality incident need different evidence. They can happen together, but one should not be inferred from the other.
Start by locating the failure
If only one website is unavailable while other services work, investigate that website and the route to it. If every device at home loses access, check the router and access connection. A comparison over a second, independent connection can help establish the scope, although it still does not identify the cause conclusively.
| Observation | What it supports | What it does not prove |
|---|---|---|
| One service fails on several networks | A service-side or shared-path problem is plausible | That the outage is malicious |
| Only one home device fails | A device or local configuration fault deserves attention | That the ISP is attacking or blocking it |
| Traffic and errors rise together | Load may be contributing to failure | That all extra traffic is hostile |
| A provider confirms mitigation | The provider has identified an incident requiring filtering | That every remaining error has the same cause |
Mitigation has to reach the bottleneck
A firewall inside a home network cannot restore capacity already consumed on the ISP's link. The filtering needs to happen upstream of the exhausted resource. For an online service, that may involve the hosting provider, an edge network, traffic scrubbing or application-level controls.
The layers are not interchangeable. A web application firewall may help with some HTTP request floods, while a large network flood calls for capacity and filtering elsewhere. Rate limits can also reject legitimate users if they are applied without understanding shared connections and normal demand.
Record evidence before making claims
For a household connection, keep timestamps, affected services, router symptoms and results from a wired device. Ask the ISP to investigate persistent widespread failures. Do not repeatedly reset or reconfigure everything before recording what is happening.
For a service you operate, preserve request and error logs, resource metrics and provider incident information. Compare traffic with the normal pattern for that time of day. An increase after a popular announcement is a different hypothesis from coordinated abusive requests, even if both overload the same server.
The useful question is where legitimate traffic stops being served, and which evidence explains that point. Calling every outage a DDoS attack makes the diagnosis less precise.
How this article was prepared
He checks metric definitions, compares equivalent samples and examines geographic and time coverage before drawing conclusions. His analytical writing follows causal relationships, compares possible explanations and states where the evidence ends.
Read our methodology →Reviewed by the Internet Analysis Editorial Team
Reviewed by the Internet Analysis Editorial Team · Updated October 28, 2026
Meet the editorial team →Article context, review and related questions
A slow website is evidence of a problem, not proof of an attack. Separate service availability, access-link congestion and local network faults.
| Measure | Value | Context |
|---|---|---|
| Article type | Network analysis | Editorial classification |
| Reading time | 3 minutes | Estimated at approximately 220 words per minute |
| Editorial review | Internet Analysis Editorial Team | Updated October 28, 2026 |
| Review date | October 28, 2026 | Latest stored article update |
Methodology
He checks metric definitions, compares equivalent samples and examines geographic and time coverage before drawing conclusions. His analytical writing follows causal relationships, compares possible explanations and states where the evidence ends.
Full methodology →Data freshness
- Page updated
- Data period
- October 28, 2026
- Responsible editor
- TomaszFounder and network data analyst
Primary sources
- Internet Analysis editorial methodologyReview and limitation rules
Limitations
- The article is informational and may simplify technical details for readability.
- Products, standards, prices and service availability can change after the review date.
- The latest review date does not guarantee that every external product or service remains unchanged.
Related questions
What is the main point of “DDoS or a connection fault? What an outage actually tells you”?
A slow website is evidence of a problem, not proof of an attack. Separate service availability, access-link congestion and local network faults.
How was this article prepared?
He checks metric definitions, compares equivalent samples and examines geographic and time coverage before drawing conclusions. His analytical writing follows causal relationships, compares possible explanations and states where the evidence ends.
When was this information last reviewed?
The latest stored review or update date is October 28, 2026.
