Internet Blocks, the Great Firewall and VPNs

[roddy]

What options are everyone using for getting around the blocks on sites like BBC News and Wikipedia, etc.

Like most people I think I originally used anonymouse, which was a bit slow and came with ads, but worked when you needed it to.

After a while I shifted to proxify.com which is similar to anonymouse but paid, ad-free and has some additional useful features.

Since then I've installed the Gladder plug in for Firefox, which has been fantastic - works virtually transparently, you just need to specify sometimes which sites you want it to work with. Free. I'll probably not renew my proxify subscription.

I've always been aware of other options such as Tor, but have never really felt a need to use them - other methods have always seemed easier than having to download and install anything.

Am adding a poll so you can select your most commonly used method - obviously sometimes you'll use more than one, just opt for the one you use most. Let me know if I forgot any.

[imron]

Something else which I will detail below . . .

Port forwarding over SSH to a remote proxy that runs on a private machine outside of China. Works a treat. Although technically this counts as tunneling, not vaulting

[roddy]

Added that to the poll now.

[ipsi()]

Would you care to detail that slightly more? Possibly in an email or something. I'll be heading over to China in about a week (talk about surprises! I've known about this for a day or two), and it'd be nice to put my Linux box to use over here while I'm gone.

[Senzhi]

By the way, the English version of Wikipedia is accessible through the Great Firewall again since a few days.

Not sure how long it'll last this time though ...

[gato]

Tor is quite easy to install and to use. Search on google for more info.

[muyongshi]

Hey Imron are you using a tunneling service or are you being clever and have built one yourself?

Anyone can pay for a SSH Tunnel now even at relatively low costs. Sometimes it can be too slow though.

[imron]

@Muyongshi, I'm using my own solution.

@ipsi() I didn't set the machine up myself so I can't help you out with specifics, however the basic solution is:

Install Squid and set up a web proxy on a given port (the default is 3128, but the machine I use has it set up on port 28 ).

Connect to that machine over ssh using port forwarding. This allows you to map a given port on your local machine to the port on the remote machine. For example the command I use to connect is:

ssh -p 32 -L 8080:remote.machine.com:28 imron@remote.machine.com

Where:

-p 32 : specifies the port number to ssh to on the remote machine. Note, the remote machine I use has the ssh daemon running on a non-standard port.

-L 8080:remote.machine.com:28 : Maps port 8080 on my local machine to port 28 on the remote machine.

imron@remote.machine.com : specifies the remote machine to connect to, and my username on the remote machine.

Once I run this command (set up as a shell script so I don't have to type it in manually every time), I get prompted for my password, and then logged in (if I need to log in from a windows machine, i'll use PuTTY)

In my browser, I set the internet proxy to be localhost on port 8080 (SwitchProxy works great under firefox). Then any browsing I do sends traffic to localhost:8080 which forwards it to port 28 on the remote machine, which browses the web for me, sends the traffic back through the ssh tunnel to my machine, which then gets shown by my browser.

Once it's set up, the whole process is pretty seamless.

This page talks about the process in a bit more detail.

[ipsi()]

Thanks. I've managed to set that up. Just going to have to reconfigure that when I get to China. Either way, it works fine over my local network (Why I'd want a proxy server to connect from my laptop to my linux box to my router, rather than just going directly to my router is a mystery though ).

[imron]

Also, if you want to save disk space, you can set Squid up so that it doesn't actually do any caching, and acts purely as a proxy. The Squid documentation has examples of how to set this up.

[赫杰]

I just give up and read the China Daily

^0^

ummmmmmm my really ghetto way is just using google as a proxy...would that count as a free proxy service? i mean its original intention is not for proxy use I'm sure, haha, so I don't know if this counts as proxy service per se. So I am not sure what to check...

HJ

Edit: BTW this is an awesome poll! 2, I am having trouble viewing pics on Flickr...I tried downloading a firefox pluggin but it just stops around 20% everytime...so...what should I do? Please help! thx

[Senzhi]

赫杰,

I have the same problems with my Flickr account: I can access the site, but no pics show up. All my visitors from the Mainland are complaining about the same issue.

Through a proxy or from Hong Kong I have no problems, which makes us assume the culprit for this poll is also the culprit for blocking Flickr pics.

And I have just a whole collection of hundreds of glamorous, descent, non-political pics about China !!!

[imron]

You guys could try using xerobank, which was formerly known as Torpark, but has since changed names to avoid confusion with the Tor project. It's basically a modified version of Firefox configured to do all browsing through the Tor network. Download, install and away you go.

[赫杰]

Thank you imron, that worked just fine!

[ipsi()]

Also, if you want to save disk space, you can set Squid up so that it doesn't actually do any caching, and acts purely as a proxy. The Squid documentation has examples of how to set this up.

No so concerned about that. I've got a reasonable amount of HD space left, so I should be fine there. Just got to pray I can find someone to look after my box while I'm away.

[zhwj]

Risks of using software proxy systems:

I use a software proxy (FG, not Tor - I'm not so concerned about absolute anonymity so I'm willing to trust my browsing to servers belonging to another, faster organization).

On Friday, my daily update to the AVG anti-virus software suddenly started identifying it as a Trojan horse. I've had to downgrade to an earlier version (was using 6.34p) which apparently lacks that particular signature.

[gougou]

While we're at it, does anybody know how to edit the filters of gladder? With wikipedia not being in my user-controlled list but in the supplied filters, the only way I can access it directly is by disabling gladder. However, by the time it finishes loading through proxy I usually will have lost interest already, so it would be nice to disable just this one filter. Is there any workaround?

[gougou]

I've been having a LOT of time-outs recently for sites that have never been a problem, and also now are not blocked. That can happen with pretty much any site (e.g. German news sites), so I'm not sure whether it is connected to some filter function of the firewall or a network problem.

However, I found a useful recourse: the Firefox plug-in Mr. Uptime will keep monitoring sites that timed out in the background until they become accessible again. I guess you could also use that to keep track of which sites get unblocked - just make sure to set the time span long enough.

[roddy]

I've now got Tor installed and while it's quite handy, especially with the firefox plug-in, it's painfully slow. I can't remember what dial-up was like (poor memory, not because I'm too young) but I suspect it might have been a bit like using Tor. Are there any faster alternatives? Zhjw, you mention FG? Is that worth trying?

[gato]

I've now got Tor installed and while it's quite handy, especially with the firefox plug-in, it's painfully slow.

Tor's performance is uneven, probably because it relies on people volunteering their machines to be proxies. I'm sure the Great Fire Wall's trying its best to defeat tor, as well, though with what success, I have no idea. My needs aren't that demanding as I use it mostly to read non-mainland Chinese newspapers. Sometimes it's quite sufficient for that purpose, that is, it's faster than 10KB/sec; other times, it crawls or that pages can't download at all.

I doubt that FG is consistently better, as it works pretty much on the same principles. Maybe it has a more dedicated set of volunteer servers?