---------- Treść przekazywanej wiadomości ----------
Temat: Strong Security Processes Require Strong Privacy Protections Data: piątek, 18 lipca 2014, 06:16:09 Od: coderman Do: Full Disclosure fulldisclosure@seclists.org, liberationtech liberationtech@mailman.stanford.edu, cpunks cypherpunks@cpunks.org, oss- security@lists.openwall.com
"Strong Security Processes Require Strong Privacy Protections"
A request for all security conscious organizations handling vulnerability reports to deploy privacy enhancing technologies.
---
With the Snowden disclosures and Google's Project Zero on the minds of security professionals everywhere, it is time to evaluate one more aspect of this renewed focus on 0day and targeted attacks: vulnerability submission to vendors. [0][1]
Software vulnerabilities of use to nation states and espionage organizations are recognized as a threat to privacy and basic human rights. Their impact no longer dismissable or discounted given evidence of misuse. I will not discuss hardware vulnerabilities in this treatment as they entail different considerations and constraints. [2]
Reporting vulnerabilities of this nature in turn requires strong privacy protections commensurate with the five and six digit monetary values they command, and the adversaries intent on discouraging their discovery or mitigation. [3][4]
---
Therefore, any organization handling vulnerability reports must support strong privacy for vulnerability submission. This is mandatory even if most or all issues received via this channel are not 0day, not high value, and entail very little risk to users.
The characteristics of a strong private reporting method are:
- Email must not be used. In the best circumstances email leaks too much information. In common situations it is passed around clear text, trivially interfered with, and winds through software with huge usability and vulnerability problems. Email for initial security vulnerability reporting must cease immediately. [5][6]
- Public web systems for vulnerability reporting must not be used. Like email, this leaks too much information and is vulnerable to a wide array of attacks destroying any privacy intended. [7][8]
- Submission of reports via hidden site required. This has become fashionable in media organizations as the "secure drop" for whistleblowers, and it is equally apropriate for vulnerability reporting. This significantly raises the cost of surveilling a vulnerability reporting service, and ensures that passive interception of reported vulnerabilities is impossible. [9]
- Encryption of submitted reports required. PGP and GPG are wonderful tools, despite encrypted email being a dismal failure. While the hidden drop may protect the privacy of the reporter, encryption of the report content to specific vulnerability researchers' keys ensures privacy to the receiver. A compromise of the hidden site must not lead to access of reported vulnerabilities. [10]
- Submitter anonymity the default. Submissions and communication must accomodate an anonymous identity. If a researcher wishes to claim credit they must opt-in and provide additional information. No psuedonymous account requirements, no key linking across submissions.
- Obfuscated disclosure should be available if desired. Capturing 0day in the wild used for espionage or cyber effects is a rare event. Publicly disclosing when, where, and how you obtained such captures ensures you're likely never to see any others. Researchers in position to observe and inspect such events should be able to report the vulnerabilities without credit and without indicating the origin. A vendor could provide a "cover story" for how the vulnerability was discovered internally, to best protect sources' ability to continue to discover these types of weaponized exploits in the wild.
Finally, it goes without saying that this privacy applies during reporting and mitigation phases of defect resolution. Once a patch is prepared and public the details of the vulnerability should be public as well, via email list, public blog, or any other useful medium.
---
As participants in the security industry it behooves us all to set an example for others and to demonstrate a committment to security and privacy via action.
Security conscious organizations handling vulnerability reports can support strong privacy and send a clear message deploying private reporting methods described above.
Security researchers must demand strong privacy from organizations they collaborate with, even in the most trivial or minor of circumstances, so that infrequent severe vulnerabilities may also be reported in confidence.
Privacy is a basic human right we must all support. Let's demonstrate our support by using privacy enhancing technologies to resolve risks to privacy!
best regards,
0. "The NSA Revelations All in One Chart" https://projects.propublica.org/nsa-grid/
1. "Announcing Project Zero" No link as the announcement is only supported over HTTP; attempt HTTPS and you're redirected to plain-text. This is an embarassment that should be fixed, Google Project Zero! (the other plain-text sites below have not unreasonable exuses ;)
2. "New technologies are radically advancing our freedoms but they are also enabling unparalleled invasions of privacy" https://www.eff.org/issues/privacy
3. "A Declaration of Cyber-War" http://www.vanityfair.com/culture/features/2011/04/stuxnet-201104 '''On July 15, the day Stuxnet’s existence became widely known, the Web sites of two of the world’s top mailing lists for newsletters on industrial-control-systems security fell victim to distributed-denial-of-service attacks...'''
4. "The Real Story of Stuxnet" http://spectrum.ieee.org/telecom/security/the-real-story-of-stuxnet '''Just as Kaspersky’s engineers were tricking Gauss into communicating with their own servers, those very servers suddenly went down,...'''
5. "Universal Email Encryption Specification" http://ritter.vg/blog-uee_email_encryption.html
6. "Pond" (not like email) https://pond.imperialviolet.org/tech.html
7. "Bullrun (decryption program)" https://en.wikipedia.org/wiki/Bullrun_%28decryption_program%29
8. "How secure is HTTPS today? How often is it attacked?" https://www.eff.org/deeplinks/2011/10/how-secure-https-today
9. "How the NSA Attacks Tor/Firefox Users With QUANTUM and FOXACID" https://www.schneier.com/blog/archives/2013/10/how_the_nsa_att.html (this is what a hard to attack system looks like, and keeping disclosure entirely within the network from clients to hidden sites amplifies the difficulty significantly.)
10. "The Rise of the Middle and the Future of End-to-End: Reflections on the Evolution of the Internet Architecture" https://www.ietf.org/rfc/rfc3724.txt
-----------------------------------------