3 ms·
From my read of the NSA sections of the snowden docs, they didn't decrypt anything- they observed the Google front end that talked to the user was the end of th
by dekhn 4y ago
From my read of the NSA sections of the snowden docs, they didn't decrypt anything- they observed the Google front end that talked to the user was the end of the TLS chain and from there, inside Google's networks, the front end talked to its backends without any encryption, so as long as you had physical access to Google's network (which included intercontinental fiber) you could trivially sniff packets. https://www.washingtonpost.com/world/national-security/nsa-infiltrates-links-to-yahoo-google-data-centers-worldwide-snowden-documents-say/2013/10/30/e51d661e-4166-11e3-8b74-d89d714ca4dd_story.html https://www.washingtonpost.com/world/national-security/nsa-i...
I can't find the article now but it had a picture of an NSA presentation with what was obviously a bigtable RPC (spend enough time debugging protocol buffers and stubby). From what I can glean, it seems likely that they tapped an undersea cable carrying Google's traffic between a frontend and a backend, although that's speculation.
- monocasa 4y agoThey did both. They tapped private inter-DC fiber of Google's that wasn't encrypted. Additionally they had a cute hack that could crack Diffie Hellman at a cost of about $100M a prime (and most of the internet at the time used the same primes). So most TLS/IPSEC/SSH etc was wide open to them too if they could catch the handshake.
- dekhn 4y agoCan you point me at the DH crack? This? https://www.theregister.com/2015/10/19/nsa_crypto_breaking_theory/ https://www.theregister.com/2015/10/19/nsa_crypto_breaking_t... I don't recall that ever being associated with breaking Google traffic although I recall there were some Google SSL changes around the time which may be related. I simply haven't seen any direct evidence, while a packet dump on a slide is direct evidence.