Live data from Hacker News

How the great firewall of China detects and blocks fully encrypted traffic [pdf]

gfw.report

11–20 of 289 posts

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#11
post #2

Seems like UDP is completely exempt, which would allow UDP-based VPNs, like Wireguard through. SSH is also exempt...

I'd go for ssh if I was trying to bypass it. At least legally I can claim that I'm just sshing to my aws server and not be jailed for using vpn.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#12

The algorithm found seems so unintuitive that I wonder if it was not found by the AI. " Allow a connection to continue if the first TCP payload (pkt) sent by the client satisfies any of the following exemptions: Ex1: popcount(pkt) len(pkt) ≤ 3.4 or popcount(pkt) len(pkt) ≥ 4.6. Ex2: The first six (or more) bytes of pkt are [0x20,0x7e]. Ex3: More than 50% of pkt’s bytes are [0x20,0x7e]. Ex4: More than 20 contiguous by…

This is just some experimentation results, it's not algorithm.

An algorithm is just a bunch of rules to follow to perform an operation, so this looks like an algorithm to me.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#13
post #2

Seems like UDP is completely exempt, which would allow UDP-based VPNs, like Wireguard through. SSH is also exempt...

Why would they let that happen? Doesn't seem to make any sense to me if it's how you describe it.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#14

The algorithm found seems so unintuitive that I wonder if it was not found by the AI. " Allow a connection to continue if the first TCP payload (pkt) sent by the client satisfies any of the following exemptions: Ex1: popcount(pkt) len(pkt) ≤ 3.4 or popcount(pkt) len(pkt) ≥ 4.6. Ex2: The first six (or more) bytes of pkt are [0x20,0x7e]. Ex3: More than 50% of pkt’s bytes are [0x20,0x7e]. Ex4: More than 20 contiguous by…

> I wonder if it was not found by the AI.

Do you mean "found" by the CCP, or "found" by the researchers? In the case of the CCP it was likely generated through basic statistical analysis, and tuned to minimize side effects and collateral damage below some threshold of acceptability (~0.6% of global traffic unintentionally blocked). In the case of the researchers, the paper details the basic statistical analysis used to discover these rules.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#15
post #10

I remember having to deal with the early GFW about 20 years ago when I was working for a company that had some employees on a site in Shanghai. Every morning, our colleagues in China would open their mail client and it would connect to our server abroad. The first person would usually be OK, but for everyone else, the connection would fail. At the time, almost nothing was known of the GFW and it wasn't as clever as i…

We have a satellite office in Dubai. I know their static IP. When they connect to our imap/smtp server they are coming in from another IP. I never looked into it deeply but assumed their connection is being diverted for inspection. (If true, they would probably not be below performing industrial espionage with the data they are accessing)

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#17
post #2

Seems like UDP is completely exempt, which would allow UDP-based VPNs, like Wireguard through. SSH is also exempt...

Went to china some years ago and my pptp vpn blocked after a day. Switched to ssh and after a day it was rate limited to basically nothing, but I could avoid that by switching port every morning.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#18
On the other hand, this shows GFW authors are more, and more considerate of the collateral damage, which is a surprise. It seems GFW has indeed became good enough to frustrate casual users to trigger uproar when windows update, or AWS ip ranges go belly up, or something.

VPN authors should chose the maximum collateral damage strategy to frustrate GFW authors, make China as close as possible to completely cutting off outside internet. No need to completely evade fingerprinting, instead, do the complete opposite, and try to mimic common protocols, and critical applications as much as possible.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#19
But does the paper imply that something like chunked encoding smuggled HTTP requests with an encrypted payload after the second chunk would work?

That is, assuming entry nodes are available as e.g. nginx proxies inside the Chinese ASNs and are allowed to operate serving websites to ASNs from foreign countries.

I'm mentioning nginx because there were some related bypass vulnerabilities in the past, and one could argue that they just missed updating them.

Re: How the great firewall of China detects and blocks fully encrypted traffic [pdf]

#20
post #2

Seems like UDP is completely exempt, which would allow UDP-based VPNs, like Wireguard through. SSH is also exempt...

I'd go for ssh if I was trying to bypass it. At least legally I can claim that I'm just sshing to my aws server and not be jailed for using vpn.

Trying to get off the hook on a technicality isn't going to work. Lots of people use VPNs completely in the open without getting jailed, because they're not otherwise of interest, but if you are being targeted, nobody is going to care about your "sshing to aws" excuse. And ssh tunneling web traffic looks quite different from normal ssh usage anyways.
Post reply on HN