3 ms·
Hahahaha, this bug was perfectly exploitable via TLS wrapped HTTP (so HTTPS, which is still HTTP as far as the PHP application is concerned). The SWIFT_client
by ryanlol 7y ago
Hahahaha, this bug was perfectly exploitable via TLS wrapped HTTP (so HTTPS, which is still HTTP as far as the PHP application is concerned).
The SWIFT_client cookie gets passed directly into unserialize(), TLS has literally nothing to do with this.
FWIW rasengan is one of the PIA founders, he should know much better.
This response is so utterly silly I must wonder if this is all just an incredible display of incompetence instead of malice.
- rasengan 7y agoSorry, I glanced at the link you pasted and wrote the response as I knew this was a non issue from the past. So, I spoke with our internal team and was able to find more details: - We haven't used that machine since that exploit was made public. - We were never exploited. - There was no sign of intrusion of any kind. - The specific machine was a backup helpdesk test server without any real user data. Thanks again for bringing this up!
- ryanlol 7y ago>- We haven't used that machine since that exploit was made public. So what? You were exploited before kayako patched this bug, it was glaringly obvious to anyone who ever looked at the cookies set by your site. >- We were never exploited. This simply isn't true, either you're misinformed or lying. >- The specific machine was a backup helpdesk test server without any real user data. The specific machine (Which you took down really fast after I pointed it out! :P) I linked probably did not even exist in 2015, I was talking about your prod env. I don't have a horse in this race, there's no incentive for me to lie about this. I know what you are saying isn't true.