Detecting Past and Present Intrusions through Vulnerability-Specific Predicates, SOSP 2005
Discussion lead: Niel Lebeck & Shrainik Jain
What is the relationship between a patch for a vulnerability and a predicate checking for exploitation of that vulnerability?
How does IntroVirt deal with the problem of target state changing between when a predicate checks that state and when the checked code executes?
Name one type of vulnerability that can be patched, and one type of vulnerability that cannot be patched but that can be checked in a predicate.
Discuss how IntroVirt supports predicates for applications. Do you think this approach will always work?
Provide a list of questions you would like to discuss in class. Feel free to provide any comments on the paper and related topics (e.g., which parts you like and which parts you find confusing).
/var/log/httpd/*log
, Replayer (CCS’06)int 3
), hw breakpoints (DRx)