Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The NSA’s Hidden Spy Hubs in Eight U.S. Cities

theintercept.com

Post reply on HN
Print view View original

121–130 of 192 posts

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#121

Post by schoen »

AndrewKemendo wrote: post #116
schoen wrote: post #110

Earlier quoted context omitted.

> you can't completely conceal metadata You can do it with a very high cost (in overhead, latency, and availability) by having a large number of people all send and receive messages, on a fixed or randomized schedule, exceeding their maximum possible amount of communication with one another. Then someone monitoring the network knows that each of the participants in this system could have communicated with any other p…

Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly. The bottom line is that you are going to leave a signature of some sort through communications - the question is, can you properly bui…

> Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly.

To eliminate the statistical observability of metadata, the padding needs to reach or exceed the maximum capacity of the channel. So you can't have people sending more messages than the padded channel permits per time period. In your example, packets "with variable length/size content" would need to be absolutely prohibited, or else all packets' length would need to be randomized, and message data would need to be sent following strictly the same distribution as padding messages.

For example, you and I could have a rule of exchanging exactly 1 MB of data per day, at a specified time, every day. Then an observer wouldn't be able to tell whether, on a particular day, we had actually communicated something to each other or just allowed the padding data to go out. Clearly in this system we're not ever allowed to use it to transmit more than 1 MB per day, without destroying the metadata unobservability property. An attacker still knows that you and I are part of a system that offers us an otherwise unobservable channel, but not when we do or don't make use of that channel.

There are lots of variants that also allow many-to-many messaging, again at a high cost in overhead, latency, and availability.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#122

Post by dekhn »

closeparen wrote: post #103
stephengillie wrote: post #84

The Seattle location - 1122 3rd Ave - is an interesting spot. According to Google Maps[0], next door is the FBI Seattle Division office. On street view, the building itself has a street-level office with signs for both AT&T and CenturyLink. 3rd Ave itself is slightly notorious - hosting the surface entrances for the Seattle bus tunnel, and a large number of inter-city routes, the area has a reputation for crime and t…

It seems odd to have an ISP office in the middle of downtown?

Once I learned to recognize what an telephone exchange looks like, I have a hard time not seeing them. All over the place. Downtown, neighborhoods, etc. Big, unmarked buildings with no windows, mostly concrete, lots of infrastructure on top, and various telco trucks parked around at all hours.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#123

Post by jlgaddis »

judge2020 wrote: post #108
coldtea wrote: post #23

Earlier quoted context omitted.

> One wonders if a CDN might advertise, Yes, but they could also be lying about it.

Wouldn't it be easy to traceroute between the CDN and your server, or the CDN and your house to see if it hits servers that WHOIS to at&t?

From a network engineering perspective, it's pretty easy not to decrement the TTL (keeping devices from "appearing" in a traceroute).

Hell, this is pretty much the norm where MPLS is concerned. Your packet may hop through a dozen routers along its way without showing up in a traceroute -- you just see it go in one side (then it goes through a dozen routers) and you see it come out the other side.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#125

Post by walrus01 »

bhouston wrote: post #120

This building in downtown Ottawa looks similarly imposing: https://goo.gl/maps/6xEgv4KwJ4v

99% odds that is either a ILEC telco central office or a large electrical transformer substation. If electrical grid, similar to the big part with no windows at 970 Burrard in Vancouver. BC Hydro headquarters site with giant substation that later became office and residential condos.

edit: I just moved down the street a bit and it says Bell Canada on the building. It's a CO.

Another way you can tell for sure with a building like that, is that there will be a locked panel or set of highly protected hose ports for generator diesel fuel refill, from alley or street side tanker truck delivery.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#126

Post by ucaetano »

dekhn wrote: post #122
closeparen wrote: post #103

Earlier quoted context omitted.

It seems odd to have an ISP office in the middle of downtown?

Once I learned to recognize what an telephone exchange looks like, I have a hard time not seeing them. All over the place. Downtown, neighborhoods, etc. Big, unmarked buildings with no windows, mostly concrete, lots of infrastructure on top, and various telco trucks parked around at all hours.

They look surprisingly similar to electricity substations, with the key differences being style (substations are usually older) and the vehicles parked in front.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#127

Post by syshum »

jstarfish wrote: post #78
syshum wrote: post #55

Earlier quoted context omitted.

>Right now, you can do TLS termination ("Flexible SSL") Which sane people call Man in the Middle and should not be allowed at all. I have seen people doing this Flexiable SSL with Credit Card data and other PII believing it is "secure" Cloudflare may have started out with security in mind but their new services centered around centralization of key services (dns) and this kind of security breaking product means IMO t…

It has its uses though, even if it's misused. It's an easy box to check to pretend to offer HTTPS so you don't get penalized by Google. Before Let's Encrypt there was no free way to get a legit cert for your cat blog. Faking it via Flexible SSL was the next best thing.

This is false, there have always been ways to get a free DV cert, lets encrypt made it easier

Furhter before Lets Encrypt you would not have gotten dinged by google, Google only went that path when wide spread DV Certs where freely avaliable

Further still, the minor costs per year to get a paid DV cert should be factor when choosing to host your own content versus paying a 3rd party to do it, many of those 3rd parties provided SSL as part of their services.

There is zero need for a Man in the Middle for SSL,

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#128

Post by walrus01 »

darksaints wrote: post #43

Wow, just made a crazy realization. A while back I was downtown with my 5yo kid, and we were by the bus stop right next to this building (You can actually see the bus stop in the picture). The bus was late, and my kid was hyper, and he tried going through the revolving doors. They were locked, no big deal. After a while, he tried going through the doors again. At this point, three security guards with ballistic vests…

There is also a lot of security theater in telecom/datacenter stuff to impress the rubes. The Switch datacenter in Las Vegas had armed guards with AR-15s and body armor for a while. Network engineers who visit and enter a lot of these facilities joke about it all the time.

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#129

Post by 394549 »

schoen wrote: post #121
AndrewKemendo wrote: post #116

Earlier quoted context omitted.

Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly. The bottom line is that you are going to leave a signature of some sort through communications - the question is, can you properly bui…

> Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly. To eliminate the statistical observability of metadata, the padding needs to reach or exceed the maximum capacity of the channel. So…

>>>> but you can't completely conceal metadata.

> For example, you and I could have a rule of exchanging exactly 1 MB of data per day, at a specified time, every day.

Depending on the size and popularity of the relay network, the fact the two parties are connected to it could be valuable metadata.

If you really wanted to minimize the amount of metadata to something that's almost useless, you'd probably need to use something like a continuously-operating broadcast numbers station.

https://en.wikipedia.org/wiki/Numbers_station

Re: The NSA’s Hidden Spy Hubs in Eight U.S. Cities

#130

Post by AndrewKemendo »

schoen wrote: post #121
AndrewKemendo wrote: post #116

Earlier quoted context omitted.

Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly. The bottom line is that you are going to leave a signature of some sort through communications - the question is, can you properly bui…

> Even ignoring the practicality part, it becomes a timing game, because "empty" messages - even if they were filled with unintelligible "random" hex - would traverse the network differently than ones with variable length/size content and would be able to be filtered out pretty quickly. To eliminate the statistical observability of metadata, the padding needs to reach or exceed the maximum capacity of the channel. So…

On it's face such a scheme seems theoretically robust, but for frequency correlation only. I'd be curious if in practice it would be possible to eliminate all other variability though, of which there are many. For example I'm unaware of any true solution to latency triangulation.

My hunch is that it wouldn't be possible, and there would be a side-channel vulnerability somewhere.

Post reply on HN