Oh SSH-it, what's my fingerprint? A Large-Scale Analysis of SSH Host Key Fingerprint Verification Records in the DNS

TL;DR

This study analyzes the prevalence and security implications of SSHFP DNS records across large internet datasets, revealing low adoption and widespread insecure deployment.

Contribution

It provides the first large-scale measurement of SSHFP record deployment and highlights security risks due to lack of DNSSEC usage.

Findings

Approximately 1 in 10,000 domains have SSHFP records.

Over half of SSHFP records are deployed without DNSSEC.

Low adoption of SSHFP records limits their security benefits.

Abstract

The SSH protocol is commonly used to access remote systems on the Internet, as it provides an encrypted and authenticated channel for communication. If upon establishing a new connection, the presented server key is unknown to the client, the user is asked to verify the key fingerprint manually, which is prone to errors and often blindly trusted. The SSH standard describes an alternative to such manual key verification: using the Domain Name System (DNS) to publish the server key information in SSHFP records. In this paper, we conduct a large-scale Internet study to measure the prevalence of SSHFP records among DNS domain names. We scan the Tranco 1M list and over 500 million names from the certificate transparency log over the course of 26 days. The results show that in two studied populations, about 1 in 10,000 domains has SSHFP records, with more than half of them deployed without using DNSSEC, drastically reducing security benefits.