Bitcoin Forum

Slush's pool AND Deepbit are both getting hit big-time, repeatedly, with Distributed Denial of Service attacks!

What gives?

ZOMG!  its HAARP!!!! I knew IT!

HAARP can only create superstorms and influence the weather -- pretty powerful stuff, but it can't perform a DDoS.

Quit being silly.

Add my site to the list as well  >:(

It's like someone's trying to damage/bring down Bitcoin -- maybe they're selling a bunch of Bitcoin at the same time?

1. target ~65% of aggregate mining pool with hired zomB net DDoS
2. Continue to mine your own 20-50 Ghash minifarm solo
3. Increase personal likelihood of mining whole blocks by factor of 1000
4. Profit

We are being challenged, we can come through.

My friend had the idea of "what if someone's going DDoS EVERYONE then MINE ALL ALONE so that the difficulty would be theroretically next to nothing.
I told him "dude everyone'll go solo"
He said that "yeah but if you take down the pools then most miners are pointed AFK for awhile, And if your the fastest guy in solo you'd maybe get a block"
I said that it wouldnt be worht the effort

Why you should always have a fallback to solo mining.

The issue here is you'd have to wait until the next difficulty change for it to be worth it - as until then/stopping DDoS, you're just having to mine slowly? :s

I just realized though.. removing all that hashing from the pool wouldn't reduce the diff.rate (immediately) so your personal odds of mining a block don't actually change.. I think you'd have to effectively remove a shit-ton of hashing power for an extended period of time and get the diff.rate to go back down for this to be an effective attack vector..

Math isn't necessarily my strong suit. Someone smarter than me please comment.

So basically someone is either just griefing because they can?

Or they want to take out the network because they dislike bitcoins.

(less pools = solo = too much variance for most)

Yeah, It would take probably months for the number of blocks to get solved for the difficulty to change.

A denial of service attack costs money and resources (infected PCs valued at $0.25 to $3 each depending on country and bandwidth available).
It's a service that has a tangible real world value.

So it's either someone very bored who has access to his own botnet, or someone is actually paying thousands of dollars per hour just to annoy miners.

There isn't really any financial gain here, since difficulty level only changes every 2016 blocks.
Attackers chances of finding blocks does not go up simply because he forces other miners out for a few hours.

Your probability is based on the difficulty not the current global hash rate.  You would need to run #1 long enough (2 weeks) to affect a difficulty decrease so #2 is profitable

Yes.  This is correct.

The only suspects who would be motivated by profit to DDoS a pool would be a competing pool.

Why are you implying that I said they would gain financially by doing it? :P

Nope ...


Correct.  Mining isn't a race.  Other people finding (or not finding) blocks doesn't affect your odds in any way.  In the long run, if you manage to knock out significant mining capacity for a long period of time, then you can affect the difficulty level, which so long as your own mining isn't affected, does mean you will end up mining more in the long run.  However, targeting pools is not a good way to do this, because the individual miners will simply hop between pools as one goes down and you won't affect the hashing rate for very long at all (not long enough to have a big long term effect).

Any challenge to the current fiat controlled banking system will be attacked wether you are a country, a company, or a cryptocurrency not controlled by them.  This is shown time and time again around the world.

http://chart.googleapis.com/chart?chs=350x200&chd=t:29.59,2.63,1.50,18.26,40.82,0.00,6.12,1.08&cht=p&chf=bg,s,00000000&chl=deepbit|BitcoinPool|bitcoins.lc|btcguild|other|Eligius|btcmine|swepool

"Other" has 40% ?



Compared to .49% 2 days ago

June 8th 2011
http://chart.googleapis.com/chart?chs=350x200&chd=t:49.40,3.05,21.96,1.62,10.46,0.49,4.06,8.16,0.80&cht=p&chf=bg,s,00000000&chl=deepbit|BitcoinPool|slush|bitcoins.lc|btcguild|other|Eligius|btcmine|swepool

That is certainly interesting. Someone must have +2thash/s of power just idling for attacks. Then again, collective hash rate did drop a lot. It was 400ghash/s at slush at one point.

Where is slush in the top pie chart?

Slush is not showing any status on bitcoinwatch.com


all graphs are from there bitcoinwatch.com

Does this mean miners are going solo because the pools are going down or misbehaving?

Those graphs have a one day lag for "other", because total network hash rate takes a long time to measure. "other" is simply the average total hash rate over the last day - the current reported hash rate of all the pools. Don't get worried until it's like this for longer than a day.

And don't worry then either. Other is decentralized miners doing their thing!  :D

The 100% of this pie chart is calculated by the rate of new block generation in the network, but all the pools there are accounted by their hashare.
Hashrate and block generation rate are somehow similar, but NOT the same.
0.49% for other was wrong number.

and now people know why Tycho (correctly) did not close registrations for deepbit despite all the people wigging out a couple days ago.

Also when right after a Difficulty Bump the equations go out of wack. Because difficultly isn't continuously updated, but instead in bursts. When calculating network hash speed it is based on Share/Block and Difficulty. The ratio of Share/Block to Difficulty, right after a Difficulty bump creates extra unaccounted for hash rates that are just purely a calculation error. It takes awhile to balance out.

For instance you guys remember how after each Difficulty jump it says something absurd like the network speed just jumps up like 1-4 Thash instantly.

Example: I don't know the exact equation but.

Say:
(Say the equation is Diffculty / Shares per second)

5 / 1 = 5 Hash/s

Right after an difficulty bump of +5:

10 /1 = 10 Hash/s

The Calculated hash rate jump but technically the Real hash speed hasn't changed. Right after the Difficulty bump the Share per second increases as Real Speed increases (Due to more users). That's why right after the huge jump after a difficulty bump you steady see overall hash speeds going down.

That is whats accounting for that giant portion of (Other) you probably got it right after the bump to 500k which was around 2 days or something ago.