![Artwork](/static/images/128pixel.png)
Mettez-vous hors ligne avec l'application Player FM !
How do we patch the right things? - Josh Bressers - PSW #840
Manage episode 435590037 series 2591184
Every week here on the show we talk about vulnerabilities and exploits. Typically we recommend that organizations remediate these vulnerabilities in some way. But how? And more importantly, which ones? Some tools we have to help us are actually not all that helpful at time, such as:
- Mitre Att&ck - Don't get me wrong, this is a great project and Adam and team is doing a great job. However, its not a complete picture as we can't possibly know about every attack vector (or can we?). People seem to think if they cover everything in the framework they will be secure. You can't cover everything in the framework because each technique can be utilized by an attack in a hundred different ways.
- CVSS - Anyone can apply a score, but who is correct? Good that we have a way to score things, but then people will just use this as a basis for what they patch and what they do not. Also, chaining vulnerabilities is a thing, but we seem to lack any way to assign a score to multiple vulnerabilities at once (different from a technique). Also, some things don't get a CVE, how are you tracking, assessing risk, and patching these?
- CISA KEV - Again, love the project and Tod is doing amazing work. However, what about things that do not get a CVE? Also, how do you track every incident of an attacker doing something in the wild? Also, there is frequency, just because something got exploited once, does that mean you need to patch it right away? How are we tracking how often something is exploited as it is not just a binary "yes, its exploited" or "no, it is not".
- EPSS - I do like the concept and Wade and Jay are doing amazing work. However, there seems to be a "gut reaction" thing going on where we do see things being exploited, but the EPSS score is low. How can we get better at predicting? We certainly have enough data, but are we collecting the right data to support a model that can tell us what the attackers will do next?
This week: YAVD: Yet Another Vulnerable Driver, why bring your own when one already exists, backdoors in MIFARE Classic, wireless hacking tips, AMD sinkclose vulnerability will keep running, you down with SLDP yea you know me, Phrack!, IoTGoats, Pixel vulnerabilities, leaking variables, a DEF CON talk that was not cancelled, Telnet is still a thing, More CNAs, and the last thing Flint Michigan needed was a ransomware attack!
Visit https://www.securityweekly.com/psw for all the latest episodes!
Show Notes: https://securityweekly.com/psw-840
3010 episodes
Manage episode 435590037 series 2591184
Every week here on the show we talk about vulnerabilities and exploits. Typically we recommend that organizations remediate these vulnerabilities in some way. But how? And more importantly, which ones? Some tools we have to help us are actually not all that helpful at time, such as:
- Mitre Att&ck - Don't get me wrong, this is a great project and Adam and team is doing a great job. However, its not a complete picture as we can't possibly know about every attack vector (or can we?). People seem to think if they cover everything in the framework they will be secure. You can't cover everything in the framework because each technique can be utilized by an attack in a hundred different ways.
- CVSS - Anyone can apply a score, but who is correct? Good that we have a way to score things, but then people will just use this as a basis for what they patch and what they do not. Also, chaining vulnerabilities is a thing, but we seem to lack any way to assign a score to multiple vulnerabilities at once (different from a technique). Also, some things don't get a CVE, how are you tracking, assessing risk, and patching these?
- CISA KEV - Again, love the project and Tod is doing amazing work. However, what about things that do not get a CVE? Also, how do you track every incident of an attacker doing something in the wild? Also, there is frequency, just because something got exploited once, does that mean you need to patch it right away? How are we tracking how often something is exploited as it is not just a binary "yes, its exploited" or "no, it is not".
- EPSS - I do like the concept and Wade and Jay are doing amazing work. However, there seems to be a "gut reaction" thing going on where we do see things being exploited, but the EPSS score is low. How can we get better at predicting? We certainly have enough data, but are we collecting the right data to support a model that can tell us what the attackers will do next?
This week: YAVD: Yet Another Vulnerable Driver, why bring your own when one already exists, backdoors in MIFARE Classic, wireless hacking tips, AMD sinkclose vulnerability will keep running, you down with SLDP yea you know me, Phrack!, IoTGoats, Pixel vulnerabilities, leaking variables, a DEF CON talk that was not cancelled, Telnet is still a thing, More CNAs, and the last thing Flint Michigan needed was a ransomware attack!
Visit https://www.securityweekly.com/psw for all the latest episodes!
Show Notes: https://securityweekly.com/psw-840
3010 episodes
Tous les épisodes
×Bienvenue sur Lecteur FM!
Lecteur FM recherche sur Internet des podcasts de haute qualité que vous pourrez apprécier dès maintenant. C'est la meilleure application de podcast et fonctionne sur Android, iPhone et le Web. Inscrivez-vous pour synchroniser les abonnements sur tous les appareils.