|
Mastergerund
|
 |
June 16, 2013, 07:22:28 PM |
|
Any time a pool is down, other pools pick up a larger proportion of the total hash-rate, because the total hash-rate shrinks.
This is partially offset by the use of backup pools, but Bitminter is particularly susceptible to this type of attack because so many people use the java client. (And if your backup pool is also under attack.)
So you are technically correct that if the top pools are all DDOSed at the same time, ASICminer would benefit. Smaller non-targeted pools would also benefit, but not to the same extent as ASICminer because their total hash rate is so low.
|
|
|
|
|
|
|
|
|
|
|
|
|
The block chain is the main innovation of Bitcoin. It is the
first distributed timestamping system.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
|
|
|
DrHaribo (OP)
Legendary
 Offline
Activity: 2730
Merit: 1034
Needs more jiggawatts
|
|
June 16, 2013, 07:29:52 PM |
|
No, reducing the global hashrate doesn't have any immediate impact, as the difficulty is still the same.
Your chance to make blocks is only affected by your own hashrate and the current difficulty.
If they ddos a substantial amount of the global hashrate for a substantial amount of time then it may affect the next difficulty adjustment. You would have to ddos a lot to have any noticeable effect. Doesn't seem to be a likely reason for the attacks, unless the attackers are confused about how difficulty works.
|
|
|
|
|
Mastergerund
|
|
June 16, 2013, 07:35:44 PM |
|
I do not see how anyone gains by a DDoS attack on a pool, because the hashing just gets directed elsewhere temporarily.
This. Wherever the hashing gets directed benefits. A lot of bitminter people use the java client, so their hashing doesn't get redirected, it just goes idle until the pool is back up and they reconnect their miner. In that case, again they are just wasting their time because they only cause a disruption for the seconds it takes to connect to a backup pool. If the backup pool is taken out at the same time, their hashing would also go idle. The scenario given was an attack against several pools simultaneously. To avoid the scenario, you could have a long string of backup pools, but in most cases one or two backups are sufficient and most miners don't have more. Edit because DrHaribo posted while I was writing: No, reducing the global hashrate doesn't have any immediate impact, as the difficulty is still the same.
Your chance to make blocks is only affected by your own hashrate and the current difficulty.
If they ddos a substantial amount of the global hashrate for a substantial amount of time then it may affect the next difficulty adjustment. You would have to ddos a lot to have any noticeable effect. Doesn't seem to be a likely reason for the attacks, unless the attackers are confused about how difficulty works.
This is a good point. You don't find any more blocks, your opponents just don't find them. It could prolong the next block readjustment a little, allowing you to find more blocks before difficulty readjusts. I don't think this is what is happening, though. |
|
|
|
whitefeather
Member
Offline
Activity: 97
Merit: 10
|
|
June 16, 2013, 07:57:21 PM |
|
How feasible is it to register one's IP with Bitminter, to create a whitelist of IP addresses. Traffic from all other addresses not in the IP whitelist can then be dropped by the edge routers. Probably not realistic. But an idea nonetheless.
|
|
|
|
|
|
redtwitz
|
|
June 16, 2013, 08:07:17 PM |
|
How feasible is it to register one's IP with Bitminter, to create a whitelist of IP addresses.
Not feasible at all. Many users (myself included) have dynamic IP addresses. |
|
|
|
|
loshia
Legendary
Offline
Activity: 1610
Merit: 1000
|
|
June 16, 2013, 08:08:14 PM |
|
How feasible is it to register one's IP with Bitminter, to create a whitelist of IP addresses. Traffic from all other addresses not in the IP whitelist can then be dropped by the edge routers. Probably not realistic. But an idea nonetheless.
VPN+NAT (Proxy) to core bitminter server from IP located in US for instance. When Dos attack is launched, core sever shuts the VPN to attacked US machine and fires anothewr one to EU IP which is published here + DNS changes. So attackers will be not be so flexibble and they will be cutted of completely as long as real IP of bitminter stays hidden always. DOC just needs a couple of VPS (empty boxes) to be exposed all over the world  |
|
|
|
whitefeather
Member
Offline
Activity: 97
Merit: 10
|
|
June 16, 2013, 08:33:51 PM |
|
If they ddos a substantial amount of the global hashrate for a substantial amount of time then it may affect the next difficulty adjustment. You would have to ddos a lot to have any noticeable effect. Doesn't seem to be a likely reason for the attacks, unless the attackers are confused about how difficulty works.
Anyone have thoughts on the motive behind these attacks? Is there a shakedown coming in the near future? Who is benefiting at this moment, when we are down? |
|
|
|
|
|
Thracian
|
|
June 16, 2013, 08:50:56 PM |
|
An option for backup pool to the java client would be useful  |
|
|
|
|
|
m3ta
|
|
June 16, 2013, 09:17:47 PM |
|
An option for backup pool to the java client would be useful +1 |
|
|
|
|
btceic
|
|
June 16, 2013, 09:26:35 PM |
|
anything we can do to help?
Hosting?
Can we move it to amazon?
|
|
|
|
ewibit
Legendary
Offline
Activity: 2955
Merit: 1049
|
|
June 16, 2013, 09:54:59 PM |
|
An option for backup pool to the java client would be useful +1 +10 |
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4102
Merit: 7765
'The right to privacy matters'
|
|
June 16, 2013, 09:57:55 PM |
|
well now that difficulty has moved to 19 mill the attack will stop soon. Dr H. I did not fully explain my question earlier . I need to break it down further. How much 7Th of your 7Th is like my piece 0 backup pool no shift just lost hash. I am all java client and when you get DDoS I get 0 mining. So the last 25 blocks at 15 mill diff I got 0 BTC. Now I am fighting against 19 mill.
Next part because no backup pool is my choice.
If BTC Guild and 50 BTC and Bitminter go down for the last 25 blocks at 15 mill diff. it would be a lot of hash power around 90Th maybe 100 Th leaving 50 to 60 th left of which asic miner has 30-32th so for those last few blocks 23-25 they have easy pickings. Since even though the diff is 15 mill they have half the available hash rate. This gives them 12 or 13 of those 25 blocks. No DDoS the last 25 blocks at 15 they have 30 of 150 or 20 percent they get 5 blocks. 5 blocks vs 12 blocks. I realize that the DDoS would need to be against the top 3 pools but I am sure my idea has real merit. This means any large share holder of AM has a big incentive to do simultaneous 3 pool DDoS attacks. I sure this logic it correct.
For that matter this top 3 pool attack at any time leaves AM in the catbird seat. Not good.
|
|
|
|
|
btceic
|
|
June 16, 2013, 10:32:56 PM |
|
An option for backup pool to the java client would be useful +1 +10 We desperately need a backup pool, I had to temporarily switch to slush's pool for now until you can get yours back up and running. |
|
|
|
Paulizei
Member
Offline
Activity: 98
Merit: 10
|
|
June 16, 2013, 11:01:41 PM |
|
The pool is completly offline?!
|
|
|
|
philipma1957
Legendary
Offline
Activity: 4102
Merit: 7765
'The right to privacy matters'
|
|
June 16, 2013, 11:08:36 PM |
|
BTC Guild has found 11 out of the last 25 blocks, which equates to a 44% of the last 25 blocks.
same idea as mine except the gainer was BTC Guild not AM. the concentration of hash in 3 spots BTC Guild , 50 BTC , AM , all at 32 Th plus then bitminter at 7 Th makes these attacks a bad thing. If you know you are able to do this to a few of the top miners at the last 20-30 blocks you get an edge. it is very easy to fix have an auto switching agreement among the larger miners/pools. end of DDoS incentive. this DDoS issue along with the crackdown on moving coins to cash will kill off BTC if not addressed. |
|
|
|
kano
Legendary
Offline
Activity: 4466
Merit: 1800
Linux since 1997 RedHat 4
|
|
June 16, 2013, 11:12:23 PM |
|
Simply use backup pools and the problem doesn't exist ... it's been well known for a VERY LONG TIME ... so if software STILL doesn't have it ... use something written in the last ... 18-24 months that does ...
|
|
|
|
|
btceic
|
|
June 16, 2013, 11:15:07 PM |
|
The pool is completly offline?!
Yes and has been for at least a couple of hours as far as I can tell. |
|
|
|
|
btceic
|
|
June 16, 2013, 11:17:04 PM |
|
Simply use backup pools and the problem doesn't exist ... it's been well known for a VERY LONG TIME ... so if software STILL doesn't have it ... use something written in the last ... 18-24 months that does ...
Understood, but for newbs and the like it is really easy to use his java applet and it has a nice interface to boot. I choose it because cgminer 3.02(I think it was) couldn't see my second GPU. I downloaded 3.2.1 and am finally up and running with a CLI, which I may just stick with. |
|
|
|
not.you
Legendary
Offline
Activity: 1726
Merit: 1018
|
|
June 16, 2013, 11:29:25 PM |
|
Back up for the moment at least...
|
|
|
|
|
DrHaribo (OP)
Legendary
Offline
Activity: 2730
Merit: 1034
Needs more jiggawatts
|
|
June 16, 2013, 11:31:48 PM |
|
Mining service is back online again.
I am working on getting backup pools in the client, multiple mining servers, and some other things that will help.
|
|
|
|
|
