Bitcoin Forum

One reason people think TOR is hard to DDOS is because it does not allow UDP.  Most DDOS is UDP based and TOR only allows TCP.  Still this does not make TOR impossible to DDOS it just means you need to flood a lot of nodes with TCP protocol.  I'd imagine using a lot of zombie PC with TOR clients one could map out TOR nodes and develop a fairly simple plan to flood enough nodes to make TOR fairly unusable.

Given how the TOR network operates and the bandwidth constraints, I imagine it's easier to take a site down in terms of traffic required. But harder to do, given that you need your zombies to connect to the Tor network first.

Given how the TOR network operates and the bandwidth constraints, I imagine it's easier to take a site down in terms of traffic required. But harder to do, given that you need your zombies to connect to the Tor network first.

I think it's just zombie pcs "refreshing". I read somewhere that the thing that supposedly would make DDOS agains the Tor network has not been implemented, because it would require a fair amount of resources. With a sizeable botnet, the zombies could just send legitimate requests over and over, making sites like silk road and such, with stone age servers, be almost unusable.

Would mirroring the site under many different addresses make it ddos proof in tor?

The problem is that all transactions are stored in a database.  In order to mirror the databases have to be sync'd.  When you have high speed transactions this gets difficult with distance due to latency.  Wall street gets around this having invested billions of dollars in fiber connects and centralizing everything in one spot.  Of course if Wall street was the new 104 story building about to be completed I'd compare MtGox or Silk road as a 2 man pup tent.  Not to be critical but they are small potatoes.

So after more research Silk Road has two larger competitors BMR and Atlantis. 

2. The speed of the server has pretty much nothing to do with DDOS resistance.  DDOS is accomplished by saturating the network connection, not attempting to overload a machine that is simply piping data from disk to the network port.

"UPDATE (04/30/2013 0035 UTC): We have come a long way in the battle, but still do not have the upper hand. It's looking more and more like a restructuring of the tor software or even the tor network will be required to mitigate the kind of attack we are under. If this can be solved by modifying the tor client software running the silk road .onion, then it will be a matter of patching and redeploying the hidden service. If it is a network issue, it will require the cooperation of the Tor developers, or running a new network of nodes. I haven't given up hope for a faster solution, but if one can not be found, then we will move to a semi private scheme where users will be given access through many private URLs. I'll keep everyone updated on how we will move forward, but please be prepared for a few more days at the least of no access."

where is that from?

Do these competitors use Bitcoin as well ?

Atlantis was Litecoin only last time I heard. Looks like selling my Litecoins was a massive mistake! :(

... Wait, silk road has been down for days and BTC price HASN'T tanked?

... Might this mean that BTC price isn't carried by Silk Road, i.e. we AREN'T still in 2011?

And according to some bull/bear schizos on here, the "speculators" have left, interest is down, and BTC is headed for the doldrums.

So if Silk Road is down, and "the speculators" "have left", and price is holding... major bullish, because it means price isn't based on drugs and wall st. wannabes, according to the bears' own arguments

 8) 8) 8) 8) 8)  

But couldn't all the mirrors be run from the same source?  Because the DDoSers would just be targeting the website distribution in general, as opposed to targeting the specific source computer (if they could locate/target the source server then what would be the point of Tor)

So couldn't the dread pirate roberts just run 100 instances of Tor on the same server all with different tor addresses, but still serving up the same data from the same database?

But then again I have a really limited knowledge of Tor so this could be completely wrong.

Price will go up if  SR is down hopefully. Less coins to be sold.

No dumb drug dealers cashing in by putting in market orders causing selloffs. That's one of my theories, anyways.

Atlantis was Litecoin only last time I heard. Looks like selling my Litecoins was a massive mistake! :(

Price will go up if  SR is down hopefully. Less coins to be sold.

That's incorrect.

While saturating a network connection is ONE way of causing a denial of service, overloading the target machine(s) so that they can't respond to requests even if the connection isn't saturated would also count as a DOS attack. As long as someone is intentionally trying to cause a network resource to become unavailable it's a DOS attack regardless of the means through which they cause that service denial. (If they're using a zombie botnet or some such then it's a DDOS).

But couldn't all the mirrors be run from the same source?  Because the DDoSers would just be targeting the website distribution in general, as opposed to targeting the specific source computer (if they could locate/target the source server then what would be the point of Tor)

So couldn't the dread pirate roberts just run 100 instances of Tor on the same server all with different tor addresses, but still serving up the same data from the same database?

But then again I have a really limited knowledge of Tor so this could be completely wrong.

... Wait, silk road has been down for days and BTC price HASN'T tanked?

... Might this mean that BTC price isn't carried by Silk Road, i.e. we AREN'T still in 2011?

And according to some bull/bear schizos on here, the "speculators" have left, interest is down, and BTC is headed for the doldrums.

So if Silk Road is down, and "the speculators" "have left", and price is holding... major bullish, because it means price isn't based on drugs and wall st. wannabes, according to the bears' own arguments

 8) 8) 8) 8) 8) 

Ashley, don't be silly, I've been telling you forever the price is solely based on speculation and not usage or adoption.  Right now my best guess is it is being used as a temporary cash store in hopes of good news later on.  There is nothing surprising here at all.

Also Silk Road has had on and off issues for a long while so it will take more than 3 days to lose confidence.  If MtFUX didn't bring BTC down to double digits that shows the irrational exuberance is still at an all time high.  I've predicted a slowly downward trend for a long time although a panic sell based on possible news scenarios wouldn't shock me.  Once Silk Road comes back who knows another irrational bubble as I first was thinking.  I figured it would be up by today though.

EDIT - Also those with BTC in silk road can't panic sell currently their Bitcoins are gone unless the site comes back.

I don't want to spread FUD (just trying to use logic) but the longer Silk Road is down the more I worry a big cash out when funds do become available.

The site is back up now, no telling how long.  I think over 50% have faith in DPR and will not try to mass sell.  At least I hope for the sake of bitcoin.  No telling how long this uptime will last once the jackasses return.  Let's hope they did some good engineering and things will remain up a while.

Any idea how the timing coincided with the recent downspike to $127?

Any idea how the timing coincided with the recent downspike to $127?

I wonder if this entire situation could have been engineered to produce a large downspike.

 the market was also severely overbought short- and long-term.

Can you explain this to me?

of course. :)

'overbought' and 'oversold' signals are usually given by oscillators, a special kind of technical indicator.

oscillators measure something that is comparable to the intuitive notion of the 'momentum' of price. since a general rule of markets is 'what goes up must go down', we can use oscillators to see the magnitude of the net 'upness' versus the magnitude of the net 'downness' in a given time frame, and anticipate corrections based on their proportions.

-===-

*1-month 6-hour scale*

*snip*

-===-

if the two oscillators are legible to you, the red splotches indicate periods of 'overbuying', where the net upness was high for a long period of time. notice how we have spent a lot of time in the overbought region in the mid- and long-term (mid-term circled in green, long-term not shown), and just left the overbought zone in the William's oscillator in the short-term (circled in blue).

further, we can interpret the current downward trending of both oscillators as a downward trend in 'price momentum', which both anticipated the sell-offs we just witnessed, as well as suggests that the market may be preparing for a major correction in the immediate future.

hope that was clear enough -- are you familiar with these indicators at all?

--arepo

Ah, thank you for that (detailed) explanation. Great contribution to the community ^.^
That is very interesting, my knowledge does not go past market depth at the moment :P.
I will have to look into that more, bit confused on the Y-axis units and how they relate to each other.

Does an equilibrium exist on these graphs? Again I’m not sure on how they are calculated so I’m not sure how to tell.