denial of service
An attack that makes a service unavailable to its real users by overloading it or making it crash.
An outage stops every user, and algorithmic variants need very little attacker bandwidth.
In more detail
Some attacks flood a service with traffic; algorithmic attacks such as hash flooding instead send small inputs that trigger worst-case behaviour. Defences include input limits, rate limiting and algorithms whose worst case is acceptable. When the attack comes from many machines at once it is called distributed denial of service.
Examples
The course shows that an attacker can spend your resources without reading any data, for example with unlimited login guesses or a request for a million rows, which is why rate limits and size bounds belong at the API boundary.Amplified DDoS abuses open services such as DNS or time servers that send large replies to small spoofed requests, aiming the replies at the victim.Related terms
- hash flooding — An attack that sends a server many keys crafted to collide in its hash tables, turning each O(1) operation into O(n) and exhausting its CPU.
- adversarial input — Input chosen deliberately by someone who wants your code to behave badly, often by hitting its worst case.
- worst case — The input of a given size that makes an algorithm do the most work, and the cost on that input.
- rate limiting — Capping how many requests a client may make in a period of time, and rejecting or delaying the rest.
- retry storm — A flood of retries from many clients at once that hits an already struggling service harder than the original traffic.
- load shedding — Deliberately rejecting some requests when overloaded so the requests that are accepted can still be served well.
- Slowloris — A denial-of-service attack that opens many connections and sends each request extremely slowly, so the server runs out of connections for real users.
- hacktivist — An attacker driven by a political or social cause, who leaks data or defaces websites to make a point.
- DNS — The system that turns names such as learnsome.tech into IP addresses.
- STRIDE — A threat-modelling checklist of six threat types: Spoofing, Tampering, Repudiation, Information disclosure, Denial of service and Elevation of privilege.
- unrestricted resource consumption — OWASP's name for APIs that let callers spend unlimited processor time, memory, connections, rows or money.
Taught in
- What A Hash Function Promises — Algorithms & Data Structures for Working Engineers
- Twelve Alerts, Ten Services — AWS Certified Cloud Practitioner (CLF-C02) Exam Prep
- Security Sprint: Fifteen Questions' Worth — AWS Certified Cloud Practitioner (CLF-C02) Exam Prep
- The Public Subnet That Was Not — AWS Certified Cloud Practitioner (CLF-C02) Exam Prep
- Confusable Services Sprint: One Clue Each — AWS Certified Cloud Practitioner (CLF-C02) Exam Prep
- The Benefits, In Microsoft's Words — Microsoft Azure Fundamentals (AZ-900) Exam Prep
- Zero Trust, Defence In Depth And Defender For Cloud — Microsoft Azure Fundamentals (AZ-900) Exam Prep
- The Two A.M. Page — CompTIA Cloud+ (CV0-004) Exam Prep
- Allowed In, Lost On The Way Back: Security Controls — CompTIA Cloud+ (CV0-004) Exam Prep
- The Bill That Jumped Overnight: Spotting Attacks — CompTIA Cloud+ (CV0-004) Exam Prep
- Resilience Engineering: Circuit Breakers & Retries — Production Full-Stack Engineering
- Traffic Management: Sliding Window Rate Limiting — Production Full-Stack Engineering
- Native HTTP Server: Request & Response Lifecycles — Node.js Internals & Backend Services
- Adversary Tactics, Techniques & Procedures (TTP) Mapping — Security Operations & Threat Hunting (SOC)
- The Threat Model Habit — Application Security & Threat Modeling for Engineers
- Rate Limiting And Resource Consumption — Application Security & Threat Modeling for Engineers
References
- NIST glossary: denial of service
- OWASP Denial of Service Cheat Sheet
- CISA: Understanding denial-of-service attacks
- MITRE ATT&CK: Network Denial of Service