[4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool

[naypalm]

Thanks for keeping it running! My home is nice and toasty now

[OgNasty]

Pardon my ignorance of NMC. Where do I get a NMC wallet?

You can use mine.  My6ZwP3kVVdaVZ8uNMfes82cq2qEzZNkrF   

[slush]

On the connection issue; you are able to pick your exit nodes which should allow one to find nodes that have the best connection between them and the pool of choice.

Exit nodes are used only for accessing public Internet, they're not used for hidden services. It's good, because throughput of exit nodes is usually the main reason for low performance (you need to be real Tor supporter to run exit node; I hanged 10Mbit node after two years of fighting for it with my ISP).

Hidden services are building circuits (usually three hops) to few random long-running tor relays after startup. Then service publish it's descriptor with list of it's rendez-vous nodes. Client pick onion address and retrieve descriptor. Then choose one of rendez-vous node and create (usually three hops) circuit to them, which finally create pipe for routing TCP traffic.

Although rendez-vous nodes for given service are publicly known, Tor automatically re-route traffic when node is going down, so there's usually only few-second downtime, which make connection between user and hidden service pretty DDoS resilent.

When anonymity is not an issue, it's also possible to setup shorter routes than "(client) -> OR1 -> OR2 -> (rendez-vous) -> OR3 -> OR4 -> (hidden service)". Technically is necessary only "(client) -> (rendez-vous) -> (hidden service)", which makes connection much faster. Although I know how to setup one-hop connection manually using control port, it needs source changes to use only one hop circuits  together with automatic rendez-vous failover, so it will need some time to setup this on pool.

[slush]

So I guess for now we throw torify in front of the miner and if this takes off wait for (or help with) patching miners to try and proxy .onion domains?  Or maybe I should say, adding a new user, turning off LP and doing torify python phoenix.py ... Just Works.  And cgminer uses curl which supports SOCKS (and of course see the Tor FAQ about SOCKS and potential privacy issues) so it should Just Work too, but I haven't fiddled with that yet.

Yes, both torify or SOCKS5 proxy in miners should work. Actually it would be really amazing if somebody write short howto for running miners over Tor (maybe to bitcoin.it wiki?). Unfortunately torify works only in Linux, so Windows users need miner with SOCKS5 support.

[slush]

What do you think about paying protection money to a hypothetical multi-gbps attacker
at a rate of say -50 - -80% of what a datacenter will take for professional anti-DDoS/mitigation services?

Nonsense. Accepting attacker's game isn't a solution.

Another issue is that attacker is anonymous and with unknown intention, so there's nobody to bribe :-). But I'm strictly refusing an idea to pay for such "protection".

[slush]

A malicious attacker needs not to target the pool or hidden service;
since TOR is a relatively low bandwidth network it takes very few resources to grind all active onion routers to a halt.

It's low bandwidth, but very decentralized. It's like having hundreds of proxies before pool itself; you shut down one, but other still works. Don't forget that many of those relays are running by professionals with different skills and solutions. Shutting down them all at the same time is much harder than shutting down a single server.

DoS against the entire TOR network is also cheaper than a well hosted single site with high bandwidth.

Also using Tor network is cheaper than any DDoS mitigation service :-). I know that it won't work for 100%, mainly because only few people will configure their miners over Tor. But every piece counts.

[slush]

Downside: (prolexic) costs a lot of money (in the 4-5 figures) per month.

So it's something like 1-2 figures more than earnings from the pool at current price. Why are we considering this? :-)

If you persuade 'attacker' to voluntarily stop in exchange for a smaller sum than charged by Prolexic, both sides win.

There's nobody to bribe. May I trust somebody who send me a PM like "I'm an attacker, pay me xxx USD monthly and I will stop attacks"? Surely not.

[DutchBrat]

If you send me 500 BTC a month, I promise I will never ever START attacking you !

I don't have much hashing power so my price is low 

[Tartarus]

So I guess for now we throw torify in front of the miner and if this takes off wait for (or help with) patching miners to try and proxy .onion domains?  Or maybe I should say, adding a new user, turning off LP and doing torify python phoenix.py ... Just Works.  And cgminer uses curl which supports SOCKS (and of course see the Tor FAQ about SOCKS and potential privacy issues) so it should Just Work too, but I haven't fiddled with that yet.

Yes, both torify or SOCKS5 proxy in miners should work. Actually it would be really amazing if somebody write short howto for running miners over Tor (maybe to bitcoin.it wiki?). Unfortunately torify works only in Linux, so Windows users need miner with SOCKS5 support.

I'm game for some of the contents at least (no Windows here).  There's already a Tor page on the bitcoin.it wiki, but it's for bitcoind/IRC usage rather than mining, but the basic setup stuff is there.  Perhaps adding a Mining section, covering Linux and OS X (which I assume has torify too, but checking now) and hoping someone with Windows can confirm what I see googling about setting http_proxy in Windows and hoping cgminer+libcurl for Windows obeys?

EDIT: Page edited, OS X doesn't have torify so I just put Windows and OS X in the same section about "try setting http_proxy".  I don't know if more client specific information should go on the client pages (which are kinda sparse at times) or here.

[anthony.selby]

So I have two sites ... and one of them is working perfect ... the other :-/ ... connection problems

do I need to let you know the ips or is that something that will work itself out

[slush]

do I need to let you know the ips or is that something that will work itself out

Tell me IP, I'll watch logs.

[slush]

I'm playing with SOCKS5 support for poclbm, looks like very simple patch, unfortunately with external dependency to socksipy library. If it will work, I'll ask m0mchil to accept it to upstream. Worse thing is that poclbm is pretty sensitive to timeouts (default is 5sec, which sometimes isn't enough for Tor).

[Tartarus]

I'm playing with SOCKS5 support for poclbm, looks like very simple patch, unfortunately with external dependency to socksipy library. If it will work, I'll ask m0mchil to accept it to upstream. Worse thing is that poclbm is pretty sensitive to timeouts (default is 5sec, which sometimes isn't enough for Tor).

What's poclbm using for url parsing?  I was poking phoenix earlier (and hoping to have time today, but didn't end up with) to try and add a proxy option since there's examples for twisted (which phoenix uses) and urllib (which I hope others use, heh) for proxy.

[Iyeman]

just an fyi, looks like anonymous was able to ddos a site hosted through tor. Granted a child porn site deserved to be ddos'd but with the talks about using tor to avoid the pool going down from ddos may not be as good as it is in theory.

Anonymous’ hackers were able to exploit the PHP site with a SQL injection attack and extract the user database before launching a denial of service attack

http://arstechnica.com/business/news/2011/10/anonymous-takes-down-darknet-child-porn-site-on-tor-network.ars

[sadpandatech]

just an fyi, looks like anonymous was able to ddos a site hosted through tor. Granted a child porn site deserved to be ddos'd but with the talks about using tor to avoid the pool going down from ddos may not be as good as it is in theory.

Anonymous’ hackers were able to exploit the PHP site with a SQL injection attack and extract the user database before launching a denial of service attack

http://arstechnica.com/business/news/2011/10/anonymous-takes-down-darknet-child-porn-site-on-tor-network.ars

 OT; OT; I love this part,  “The server was using hardened PHP with escaping,” Anonymous said in its statement. “We were able to bypass it with with UTF-16 ASCII encoding.”

  What good is escaping if the parser will accept arbitrary commands in a differnt encoding? *facepalm* Why would php even parse a UTF-16 command if it was strictly a UTF-8 setup?


   There are a few things left out here; numbers! Anon has access to a huge magnitude greater ddos capacity than anyone attacking here thus far, one would assume.  Yes, it has been shown you can cripple nodes. But this report fails to reveal just how much zombie power was used to cripple the site. They have the capability to fingerprint sites and likely also the capability to make there node attacks much more focused than what your average botnet OP could. We also don't know if this Lolita site was setup with some sort of static route out of freedom hosting that would have made it much more vulnerable to ddos..

[slush]

TheGECK, sadpandatech: Thanks for text corrections (back on page 186 and 187), I finally found some time and updated homepage :-).

Edit: :-P

[sadpandatech]

TheGECK, sadpandatech: Thanks for text corrections (back on page 186 and 187), I finally found some time and updated homepage :-).

  That's all TheGeck's edits there. I did not quote him properly but thank you for pointing out that I did try to contribute. ;p

[melted349]

I have all kinds of questions for the NMC mining portion of this site.  I think I'll answer some of my own questions as I write this post so I'll try to keep this organized so others can follow.

Questions

• Why is it I seem to get payed the same amount of NMCs during different periods of time waiting for a BTC block to be found?
• Why isn't the NMC stats separate from the BTC stats?
• If a BTC block was found to be invalid, do I loose the NMC I earned? (Example below)
• Its said on the site that users that don't setup an NMC wallet are giving their NMC to other members.  How do I know I'm getting part of that and that you're not just keeping the extra NMC?
• Added: Why is it a NMC block seems to be found everytime a BTC block is found?

Example: Date:2011-10-23 17:16:46   Time:1:41:25   TotalShares:1772244   BitcoinEarned:0.03231212   NamecoinEarned:0.93021449    -    invalid

This bitcoin black was invalid.  But the combined NMC block I participated in mining were not.  So did I loose that .9 NMC? 

Also, seeing how quickly the NMC blocks are found, and just assuming there's generally an average time that NMC blocks are found.  I would expect that the longer a bitcoin block takes to find, the larger amount of NMC payout I will have with that bitcoin payout.  (The longer the bitcoin block takes to find, the more NMC blocks that are found, the higher my proportional payout will be).  Yet, I've seen bitcoin blocks take 1 hour to find, and I get the same amount of NMC as I do when it takes 2 hours for a bitcoin block to be found.

Examples:

1 Date/Time: 2011-10-25 18:05:08   TimeToFind:2:00:13   TotShares:1934180   BTCEarned:0.03839732   NMCEarned:0.69285054
2 Date/Time: 2011-10-25 16:04:55   TimeToFind:1:08:01   TotShares:1113569   BTCEarned:0.03039822   NMCEarned:0.72321182
3 Date/Time: 2011-10-25 14:56:54   TimeToFind:2:20:50   TotShares:2290272   BTCEarned:0.03190931   NMCEarned:0.75832297

Here's all the NMC blocks that were found during the time period of the 3 BTC blocks above:

1 - 2767   2011-10-25 18:05:09   0:09:11   24386    92 confirmations left
1 - 2766   2011-10-25 17:55:58   0:09:02   24385    91 confirmations left
1 - 2765   2011-10-25 17:46:56   0:25:14   24384    90 confirmations left
1 - 2764   2011-10-25 17:21:42   0:04:48   24382    88 confirmations left
1 - 2763   2011-10-25 17:16:54   0:04:56   24381    87 confirmations left
1 - 2762   2011-10-25 17:11:58   0:03:49   24380    86 confirmations left
1 - 2761   2011-10-25 17:08:09   0:30:44   24379    85 confirmations left
1 - 2760   2011-10-25 16:37:25   0:04:15   24375    81 confirmations left
1 - 2759   2011-10-25 16:33:10   0:02:22   24373    79 confirmations left
1 - 2758   2011-10-25 16:30:48   0:17:31   24372    78 confirmations left
1 - 2757   2011-10-25 16:13:17   0:08:22   24370    76 confirmations left
2 - 2756   2011-10-25 16:04:55   0:04:55   24368    74 confirmations left
2 - 2755   2011-10-25 16:00:00   0:06:35   24367    73 confirmations left
2 - 2754   2011-10-25 15:53:25   0:10:56   24366    72 confirmations left
2 - 2753   2011-10-25 15:42:29   0:00:06   24363    69 confirmations left
2 - 2752   2011-10-25 15:42:23   0:04:13   24362    68 confirmations left
2 - 2751   2011-10-25 15:38:10   0:06:01   24361    67 confirmations left
2 - 2750   2011-10-25 15:32:09   0:09:58   24360    66 confirmations left
2 - 2749   2011-10-25 15:22:11   0:05:08   24359    65 confirmations left
2 - 2748   2011-10-25 15:17:03   0:09:29   24358    64 confirmations left
2 - 2747   2011-10-25 15:07:34   0:01:12   24357    63 confirmations left
2 - 2746   2011-10-25 15:06:22   0:08:14   24356    62 confirmations left
2 - 2745   2011-10-25 14:58:08   0:00:48   24355    61 confirmations left
2 - 2744   2011-10-25 14:57:20   0:00:26   24354    60 confirmations left
3 - 2743   2011-10-25 14:56:54   0:06:56   24353    59 confirmations left
3 - 2742   2011-10-25 14:49:58   0:17:04   24351    57 confirmations left
3 - 2741   2011-10-25 14:32:54   0:31:13   24347    53 confirmations left
3 - 2740   2011-10-25 14:01:41   0:07:51   24344    50 confirmations left
3 - 2739   2011-10-25 13:53:50   0:07:39   24343    49 confirmations left
3 - 2738   2011-10-25 13:46:11   0:21:37   24342    48 confirmations left
3 - 2737   2011-10-25 13:24:35   0:05:53   24340    46 confirmations left
3 - 2736   2011-10-25 13:18:42   0:06:08   24337    43 confirmations left
3 - 2735   2011-10-25 13:12:34   0:14:26   24336    42 confirmations left
3 - 2734   2011-10-25 12:58:08   0:11:30   24334    40 confirmations left
3 - 2733   2011-10-25 12:46:38   0:03:58   24332    38 confirmations left
3 - 2732   2011-10-25 12:42:40   0:06:26   24331    37 confirmations left
3 - 2731   2011-10-25 12:36:14   0:00:10   24330    36 confirmations left
3 - 2730   2011-10-25 12:36:04   0:06:30   24329    35 confirmations left
3 - 2729   2011-10-25 12:29:34   0:10:11   24328    34 confirmations left

39 NMC blocks found in this time period.  That's 1950 NMCs.   Example 1 period: 10 NMC blocks, 550 NMCs  -  Example 2 period: 13 NMC blocks, 650 NMCs  -  Example 3 period: 15 NMC blocks, 750 NMCs

Alright so in my example, I think I answered my own question about why I'm getting payed the same amount of NMC when the BTC block takes longer to solve.  Its just the way the blocks were found.  But, this also raised another question for me at least.  Why is it that a NMC block seems to be found every time a BTC block is found?  Is there something I'm missing there?    I'll add this to my questions up top. 

Honestly I feel if there were a few more stats available to me, I'd be more satisfied.  If I could see my current hash rate, If I could see how many blocks I've contributed to each BTC block and to each NMC block, so I can verify I'm getting paid properly.  It just feels like the way it's setup now, it's all about putting faith into the "scoring" system, which I don't fully understand.  And with the few stats that are shown to us, it's near impossible to verify our earnings.  For all I know, you could be taking a 10% fee, I can't verify it.  So I guess I'm looking for some clarification here.  Any help would be greatly appreciated.   

[slush]

Why is it I seem to get payed the same amount of NMCs during different periods of time waiting for a BTC block to be found?

You answered this by yourself. It's not "the same", but it is proportional to number of found NMC blocks in the meantime of BTC block.

Why isn't the NMC stats separate from the BTC stats?

I explained this when I was introducing Merged mining; My pool was first one programmed back in December 2010. Back at this time, I didn't have an idea about alternative chains, so pool natively support only one blockchain. It is all about performance and scalability, I simply calculate shares and other stats once instead of twice (separate for BTC and NMC). Because of this, I decided to pick solution described above; Pool is collecting Namecoins and split collected NMCs between users using score from bitcoin rounds. It is perfectly fair mechanism, however I'm unable to provide fancy stats and detailed graphs about mining of namecoins. I'm planning to rewrite pool core for full support of more blockchains, but (as you will see in few days) there're more important stuff coming, so please be patient.

If a BTC block was found to be invalid, do I loose the NMC I earned? (Example below)

No. This is just side effect of merged stats - it *looks like* your NMC reward is also "invalid", but this state is related only for bitcoin block. As far as Namecoin blocks are valid, you'll get it.

Its said on the site that users that don't setup an NMC wallet are giving their NMC to other members.  How do I know I'm getting part of that and that you're not just keeping the extra NMC?

It's same as with bitcoin mining - you need to trust me. Pools aren't cheat proof from side of operators and every operator can cheat pool members easily.

You cannot recalculate if your NMC reward is correct, because you don't know who subscribed for mining namecoins and who does not. However if I'll show on site "60.3% of hashpower are NMC miners" (or something like this), you still need to trust this number. Also in pool in completely open stats (where people can list who submitted how many shares when), cheating is still possible, because you need to trust the operator that those members are real.

Added: Why is it a NMC block seems to be found everytime a BTC block is found?

This is a little technical, but: every time single share is submitted, it is compared not only against Bitcoin target, but also against Namecoin target. When submitted share has difficulty above Bitcoin difficulty, it leads to valid Bitcoin block. When share has difficulty above Namecoin difficulty, it leads to valid Namecoin block. And because NMC difficulty is currently lower than Bitcoin difficulty, everytime some miner find a valid Bitcoin block, it is automatically also Namecoin block.

This bitcoin black was invalid.  But the combined NMC block I participated in mining were not.  So did I loose that .9 NMC?  

No.

Also, seeing how quickly the NMC blocks are found, and just assuming there's generally an average time that NMC blocks are found.

Yes, current Namecoin difficulty is around 156000, Bitcoin difficulty is 1460000. So there should be (1460k / 156k) ~ 9 Namecoins block inside one Bitcoin block in average.

Honestly I feel if there were a few more stats available to me, I'd be more satisfied.

Rome wasn't built in a day. I introduced merged mining as a first large pool, so I don't think it's so bad . As I explained above, collecting detailed stats need major rewrite of pool accounting internals which wasn't in a scope of recent merged mining project. And now I'm working on another important stuff for the pool.

If I could see my current hash rate

Your hashrate is still the same, merged mining didn't affect it at all.

it's all about putting faith into the "scoring" system, which I don't fully understand.

Actually score system is still pretty easy, I'm thinking about geometric system, which is hell more complicated. Score system is basically normal proportional method, only weight of every share is rising in time. For basic math you still can use share count instead of score and it will +/- fit expectations.

And with the few stats that are shown to us, it's near impossible to verify our earnings.  For all I know, you could be taking a 10% fee, I can't verify it.

Actually you can. Those calculations are done on all pool history and it fits expectations. All calculations based on earnings of a single person (if you're comparing 'expected earnings' with 'real earnings') are subject of local pool luck, but that math done in linked thread shows that pool is as lucky as it should be.

P.S. Sorry for my crappy english, I'm really tired and this stuff is hard to explain for me . I hope that everything is much clearer for you now.

[slush]

User bars - finally! Thanks twmz for hell fast support. Please visit btcstats.net to pick yours. I'll integrate userbars to website soon.