Search

In this body of work we describe preliminary work implementing a Berkely Packet Filter, in its original conception, in an FPGA. The purpose is packet filtering and ingress traffic shaping in security-relevant applications in distributed embedded nodes. We specifically target PROFINET nodes in hard real-time applications where network security is an open issue. We describe the motivation, implementation and verification including performance characteristics. We conclude that such a filter can be used to not only for protection against simple denial-of-service attacks but also for ingress protocol management and potentially for the implementation of system-wide security policies.

© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., We propose Downright, a novel framework based on Seccomp, Berkeley Packet Filter, and PTrace, that makes it possible to equip new and existing C applications with a request broker architecture. An extensive configuration language allows AppArmor-like configuration that supports programmers in building rules for system call parameter validation and result sanitization. Access to these privileged function calls can be restricted both within Linux kernel and user spaces. Downright's main strength compared to related approaches is that it implements a complete mediation request broker architecture, in which all system calls are vetted before execution, either by the kernel or by a request broker, which runs as another process. This isolates the main program from many failures due to programming bugs and attacks, which would have to pass not only the attacked program, but the request broker also. We argue that this makes acquiring and releasing elevated privileges easier and safer. Downright eliminates the need to write Seccomp programs, instead allowing policies to be expressed declaratively through a rich policy language. We demonstrate the viability of this approach by instrumenting nginx, an industrial-strength web server and reverse proxy. While this instrumentation takes only a single line of code, we argue that even this effort can be avoided by suitable C runtime code. We show that Downright's overhead is substantial, halving nginx's perfomance, but propose measures for optimisation.

In our digital world, we have become well acquainted with the login form – username shown as plaintext, password shown as asterisks or dots. This design dates back to the early days of terminal computing, and despite huge changes in nearly every other area, the humble login form remains largely untouched. When coupled with the ubiquity of smartphones, this means we often find ourselves entering complex passwords on a tiny touchscreen keyboard with little or no visual feedback on what has been typed. This paper explores how password masking on mobile devices affects the error rate for password entry. We created an app where users entered selected passwords into masked and unmasked password fields, measuring things like typing speed, error rate, and number of backspaces. We then did an exploratory data analysis for the data, and our findings show that, perhaps unexpectedly, there is no significant difference between masked and unmasked passwords for any of these metrics.

The Internet is full of middleboxes that change packets and flows. In fact, there is probably no IP or TCP header that is not affected by at least one middlebox. Obviously, middleboxes impede path transparency, i.e., the idea that an exchange of messages results in more or less the same packets, no matter what path the packets takes. But no one seems to have a truly global view of what middleboxes do to packets on what Internet paths, which would however be an essential knowledge for new transport protocols to be successfully deployed. We address these concerns in the MAMI project by building an observatory of path transparency measurements. The project hosts an extensive set of path transparency measurements – we believe it to be the first dataset to deal specifically with middlebox involvement. In this paper, we describe that Observatory and a number of questions that we want to address with the data in that Observatory. Eventually, the project will provide public access to that Observatory so that researchers and the interested public can ask their own questions about path transparency issues and middlebox involvement.

The deployment of encrypted transport protocols imposes new challenges for network operations. Key in-network functions such as those implemented by firewalls and passive measurement devices currently rely on information exposed by the transport layer. Encryption, in addition to improving privacy, helps to address ossification of network protocols caused by middleboxes that assume certain information to be present in the clear. However, “encrypting it all” risks diminishing the utility of these middleboxes for the traffic management tasks for which they were designed. A middlebox cannot use what it cannot see. We propose an architectural solution to this issue, by introducing a new “path layer” for transport-independent, in-band signaling between Internet endpoints and network elements on the paths between them, and using this layer to reinforce the boundary between the hop-by-hop network layer and the end-to- end transport layer. We define a path layer header on top of UDP to provide a common wire image for new, encrypted transports. This path layer header provides information to a transport- independent on-path state machine that replaces stateful handling currently based on exposed header flags and fields in TCP; it enables explicit measurability of transport layer performance; and offers extensibility by sender-to-path and path-to-receiver communications for diagnostics and management. This provides not only a replacement for signals that are not available with encrypted traffic, but also allows integrity-protected, enhanced signaling under endpoint control. We present an implementation of this wire image integrated with the QUIC protocol, as well as a basic stateful middlebox built on Vector Packet Processing (VPP) provided by FD.io.

We show in an empirical study of 3241 Red Hat packages that software vulnerabilities correlate with dependencies between packages. With formal concept analysis and statistical hypothesis testing, we identify dependencies that decrease the risk of vulnerabilities (beauties) or increase the risk (beasts). Using support vector machines on dependency data, our prediction models successfully and consistently catch about two thirds of vulnerable packages (median recall of 0.65). When our models predict a package as vulnerable, it is correct more than eight out of ten times (median precision of 0.83). Our findings help developers to choose new dependencies wisely and make them aware of risky dependencies.

We study the vulnerability reports in the Common Vulnerability and Exposures (CVE) database by using topic models on their description texts to find prevalent vulnerability types and new trends semi-automatically. In our study of the 39,393 unique CVEs until the end of 2009, we identify the following trends, given here in the form of a weather forecast: PHP: declining, with occasional SQL injection. Buffer Overflows: flattening out after decline. Format Strings: in steep decline. SQL Injection and XSS: remaining strong, and rising. Cross-Site Request Forgery: a sleeping giant perhaps, stirring. Application Servers: rising steeply.

We study the vulnerability reports in the Common Vulnerability and Exposures (CVE) database by using topic models on their description texts to find prevalent vulnerability types and new trends semi-automatically. In our study of the 39,393 unique CVEs until the end of 2009, we identify the following trends, given here in the form of a weather forecast: PHP: declining, with occasional SQL injection. Buffer Overflows: flattening out after decline. Format Strings: in steep decline. SQL Injection and XSS: remaining strong, and rising. Cross-Site Request Forgery: a sleeping giant perhaps, stirring. Application Servers: rising steeply.

We show in an empirical study of 3241 Red Hat packages that software vulnerabilities correlate with dependencies between packages. With formal concept analysis and statistical hypothesis testing, we identify dependencies that decrease the risk of vulnerabilities (beauties) or increase the risk (beasts). Using support vector machines on dependency data, our prediction models successfully and consistently catch about two thirds of vulnerable packages (median recall of 0.65). When our models predict a package as vulnerable, it is correct more than eight out of ten times (median precision of 0.83). Our findings help developers to choose new dependencies wisely and make them aware of risky dependencies.