Scan Pipeline
Our data mining can be divided into three subsections.
- In the first step, an IP dataset is created for which the scans are to be performed.
- In the following step, zmap is used to check whether these IP addresses are reachable on specific ports, and zgrab2 is used to perform a protocol handshake. The resulting data is stored in a database and further enriched in the following steps.
- A Custom Platform Enumerator (CPE) is used to uniquely identify a software product. Therefore, in this step, if possible, we generate the corresponding CPE for each identified banner.
Using this data, we can now check whether the software is up-to-date. The product lifetime required for this are collected manually from the manufacturer websites.