Methods of detecting DNS flooding attack according to characteristics of type of attack traffic
AssigneeElectronics and Telecommunications Research Institute
InventorByoung Koo Kim
Life of this patent
Estimating the term…
Abstract
Disclosed are methods of detecting a domain name server (DNS) flooding attack according to characteristics of a type of attack traffic. A method of detecting an attack by checking a DNS packet transmitted over a network in a computer device connected to the network, includes determining whether the number of DNS packets previously generated within a threshold time with the same type of message, the same specific address and the same field value as in the transmitted packet is greater than or equal to a given number, and determining the transmitted DNS packet as a packet related to the attack if the number of DNS packets previously generated within the threshold time is greater than or equal to the given number.
Claims4
A method of detecting an attack by checking a DNS packet transmitted over a network in a computer device connected to the network, comprising:
- when the DNS packet includes a first query message:
- determining whether a first previous entry corresponding to the DNS packet exists;
- when the first previous entry does not exist, creating a new entry for the DNS packet, the new entry including an entry generation time;
- and
- when the first previous entry exists, dropping the DNS packet when a receive time of the DNS packet is within a threshold time of the entry generation time of the first previous entry and updating the entry generation time of the previous entry using a current time;
- and
- when the DNS packet includes a response message:
- determining whether a second previous entry corresponding to a previous DNS packet including a second query message corresponding to the response message exists;
- determining the DNS packet as a packet related to the attack when the second previous entry does not exist;
- and
- deleting the second previous entry when the second previous entry exists.
Dependencies are inferred from the claim text.