[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 427.

roy7

sr. member

Activity: 434

Merit: 250

Quote from: clown on February 11, 2014, 10:54:00 PM

Hey guys I've released the proxypool that sits between p2pool and miners, ensuring smaller miners get paid accordingly. It's currently in a beta stage and fully open source, I'd be great if you guy can help test.

It's at http://proxypool.doge.st

Very cool. Wasn't sure what you were actually doing. So the proxy will mine p2pool using a local wallet payment address, and then do payouts itself as people get enough coin balance saved up to justify a payout?

clown

member

Activity: 70

Merit: 10

Hey guys I've released the proxypool that sits between p2pool and miners, ensuring smaller miners get paid accordingly. It's currently in a beta stage and fully open source, I'd be great if you guy can help test.

It's at http://proxypool.doge.st

bkminer

full member

Activity: 216

Merit: 100

Don't let the nam-shub in your operating system.

Quote from: kano on February 11, 2014, 10:01:58 PM

Discussed almost 2 years ago:
https://bitcointalksearch.org/topic/new-attack-vector-8392

Check the history - this page was added a year ago:
https://en.bitcoin.it/wiki/Transaction_Malleability

Thanks for the clarification.

kano

legendary

Activity: 4634

Merit: 1851

Linux since 1997 RedHat 4

Quote from: bkminer on February 11, 2014, 09:32:08 PM

Quote from: smoothrunnings on February 11, 2014, 08:16:25 PM

Roy this issue you are pointing out isn't a bitcoin issue, its a site issue. The exchanges haven't been accountable for really anything that people have found an exploit which now requires them to be accountable for everything coming in and going out of their sites. So until they fix their own code nothing will change.

Actually it's a bitcoin network issue. The malleable transaction attack involves flooding the bitcoin network with transactions that have changed so the hash is no longer valid. It takes the bitcoin client time to sort out the bad transactions, hence my latency issue. I was running right on the edge with transaction fee's and the extra load bumped me up.

http://www.wired.com/wiredenterprise/2014/02/bitcoin-ddos/

Apparently some of the exchanges are having trouble processing the bad transactions...Mt. Gox had a custom implementation of the client that wreaked havoc with them.

(on my node the internet connection is an OC-3+ it's at a hosting facility)

What it is, is a bunch of phail sites (including MtGox) that ignored the issue that's been known for almost 2 years.
If you use txid to decide about transactions, you'll phail sooner or later and when that happens, if you do something stupid to repeat the transaction you may also lose out because of it.

Discussed almost 2 years ago:
https://bitcointalksearch.org/topic/new-attack-vector-8392

Check the history - this page was added a year ago:
https://en.bitcoin.it/wiki/Transaction_Malleability

bkminer

full member

Activity: 216

Merit: 100

Don't let the nam-shub in your operating system.

Quote from: smoothrunnings on February 11, 2014, 08:16:25 PM

Roy this issue you are pointing out isn't a bitcoin issue, its a site issue. The exchanges haven't been accountable for really anything that people have found an exploit which now requires them to be accountable for everything coming in and going out of their sites. So until they fix their own code nothing will change.

Actually it's a bitcoin network issue. The malleable transaction attack involves flooding the bitcoin network with transactions that have changed so the hash is no longer valid. It takes the bitcoin client time to sort out the bad transactions, hence my latency issue. I was running right on the edge with transaction fee's and the extra load bumped me up.

http://www.wired.com/wiredenterprise/2014/02/bitcoin-ddos/

Apparently some of the exchanges are having trouble processing the bad transactions...Mt. Gox had a custom implementation of the client that wreaked havoc with them.

(on my node the internet connection is an OC-3+ it's at a hosting facility)

smoothrunnings

hero member

Activity: 630

Merit: 501

Quote from: roy7 on February 11, 2014, 07:02:42 PM

Quote from: bkminer on February 11, 2014, 07:00:05 PM

mutable-tx attack on Bitcoin

Has anyone noticed any effects from this "attack?" A couple of days ago when it supposedly started my block template latency shot up from .1 to .45 and has been high every since. If there anyway to counter this kind of thing? Any know bot-net IPs to block?

I'm asking because I haven't seen any discussion about it here...

Thanks

It's being worked on at a level above my pay grade.

https://bitcoinfoundation.org/blog/?p=422

Roy this issue you are pointing out isn't a bitcoin issue, its a site issue. The exchanges haven't been accountable for really anything that people have found an exploit which now requires them to be accountable for everything coming in and going out of their sites. So until they fix their own code nothing will change.

roy7

sr. member

Activity: 434

Merit: 250

Quote from: mmouse on February 11, 2014, 08:10:52 PM

Quote from: roy7 on February 11, 2014, 06:00:18 PM

I don't know if it's a rounding error or a concern, but the slower miners have target difficulties a bit lower than the network minimum it seems.

Saw that, too. Miner share diff is only updated when get_work() for the miner is computed. So the miner share diff can be some seconds older than the latest minimum pool difficulty shown.

Ahh ok. I was concerned I had some sort of bug or oversight.

Did you add a new api call for that info, or adjust an existing one? If so I'd love it if you'd private message it to me or post here. I might want to add "average time to share" per-worker sometime soon on my nodes.

mmouse

full member

Activity: 162

Merit: 100

Quote from: roy7 on February 11, 2014, 06:00:18 PM

I don't know if it's a rounding error or a concern, but the slower miners have target difficulties a bit lower than the network minimum it seems.

Saw that, too. Miner share diff is only updated when get_work() for the miner is computed. So the miner share diff can be some seconds older than the latest minimum pool difficulty shown.

smoothrunnings

hero member

Activity: 630

Merit: 501

Quote from: bkminer on February 11, 2014, 07:00:05 PM

mutable-tx attack on Bitcoin

Has anyone noticed any effects from this "attack?" A couple of days ago when it supposedly started my block template latency shot up from .1 to .45 and has been high every since. If there anyway to counter this kind of thing? Any know bot-net IPs to block?

I'm asking because I haven't seen any discussion about it here...

Thanks

I don't believe your latency is related to some fake attack as I have been looking around at other p2pool nodes to see what their front-ends look like and noticing their latency's are lower than mine. If you are having a latency issue I would check your internet connection first.

roy7

sr. member

Activity: 434

Merit: 250

Quote from: bkminer on February 11, 2014, 07:00:05 PM

mutable-tx attack on Bitcoin

Has anyone noticed any effects from this "attack?" A couple of days ago when it supposedly started my block template latency shot up from .1 to .45 and has been high every since. If there anyway to counter this kind of thing? Any know bot-net IPs to block?

I'm asking because I haven't seen any discussion about it here...

Thanks

It's being worked on at a level above my pay grade.

https://bitcoinfoundation.org/blog/?p=422

bkminer

full member

Activity: 216

Merit: 100

Don't let the nam-shub in your operating system.

mutable-tx attack on Bitcoin

Has anyone noticed any effects from this "attack?" A couple of days ago when it supposedly started my block template latency shot up from .1 to .45 and has been high every since. If there anyway to counter this kind of thing? Any know bot-net IPs to block?

I'm asking because I haven't seen any discussion about it here...

Thanks

roy7

sr. member

Activity: 434

Merit: 250

Quote from: mmouse on February 11, 2014, 03:35:08 PM

Quote from: roy7 on February 11, 2014, 03:22:20 PM

I like it! It's helpful to see the per-miner share difficulty targets.

Thanks. Do we have some repo, where we could work on this and some other improvements to p2pool for altcoins?

I'd really like to consolidate the many altcoin builds (one git repo per coin Wink

) into one build which works for most coins, for example...

I don't know if it's a rounding error or a concern, but the slower miners have target difficulties a bit lower than the network minimum it seems.

roy7

sr. member

Activity: 434

Merit: 250

Forrestv mentioned in irc he's willing to pull in all the coin settings from other repos into the main distribution. It's be nice if the coin devs would submit pull requests with their coin setups to the official distribution when launching their coins.

mmouse

full member

Activity: 162

Merit: 100

Quote from: roy7 on February 11, 2014, 03:22:20 PM

I like it! It's helpful to see the per-miner share difficulty targets.

Thanks. Do we have some repo, where we could work on this and some other improvements to p2pool for altcoins?

I'd really like to consolidate the many altcoin builds (one git repo per coin Wink

) into one build which works for most coins, for example...

roy7

sr. member

Activity: 434

Merit: 250

Quote from: mmouse on February 11, 2014, 02:48:01 PM

I did some more minor changes to work.py and web.py to get information for the web frontend about network diff, pool diff and share diffs - I was interested to actually see what's going on Wink

You can have a look at vtc.coinpools.de:9171.

I like it! It's helpful to see the per-miner share difficulty targets. In fact, knowing the share diff target and their hash rate one could also provide an "average time per share", so miners know what to expect.

mmouse

full member

Activity: 162

Merit: 100

Quote from: mmouse on February 11, 2014, 06:56:01 AM

Quote from: roy7 on February 10, 2014, 05:40:14 PM

PS- It'd be nice if anyone else has tested my earlier diff to share feedback on it.

I implemented your patch to work.py for my VTC node at vtc.coinpools.de:9171. Looks good so far.

I did some more minor changes to work.py and web.py to get information for the web frontend about network diff, pool diff and share diffs - I was interested to actually see what's going on Wink

You can have a look at vtc.coinpools.de:9171.

bitpop

legendary

Activity: 2912

Merit: 1060

Or proxy port 80 to duplicate it

probably what they did

twmz

hero member

Activity: 737

Merit: 500

Quote from: smoothrunnings on February 11, 2014, 11:42:17 AM

Quote from: twmz on February 11, 2014, 11:30:15 AM

Quote from: smoothrunnings on February 11, 2014, 11:16:18 AM

Quote from: bitpop on February 11, 2014, 10:38:07 AM

No 9332 is both mining and web page
9333 is p2p

So I am back to my original question. How do I change the website front-end port?

The answer is still -w. You can't change the port for only the website front-end without changing the port for miners. The port for the front-end and the miners is always the same port.

That's not true. I have seen P2Pool sites that have bitcoin set to port 9332 but when you try to access the front-end get a message that no page exists, and some have bitcoin on port 9332 and the front-end on a different port like port 80.

I'll clarify that it's not possible with the standard p2pool node software and built-in command line options. Anything is possible if you are willing to change the code yourself. The sites you have seen must have customized nodes. You can too! Just break out your favorite text editor and hack away.

bitpop

legendary

Activity: 2912

Merit: 1060

Quote from: smoothrunnings on February 11, 2014, 11:42:17 AM

Quote from: twmz on February 11, 2014, 11:30:15 AM

Quote from: smoothrunnings on February 11, 2014, 11:16:18 AM

Quote from: bitpop on February 11, 2014, 10:38:07 AM

No 9332 is both mining and web page
9333 is p2p

So I am back to my original question. How do I change the website front-end port?

The answer is still -w. You can't change the port for only the website front-end without changing the port for miners. The port for the front-end and the miners is always the same port.

That's not true. I have seen P2Pool sites that have bitcoin set to port 9332 but when you try to access the front-end get a message that no page exists, and some have bitcoin on port 9332 and the front-end on a different port like port 80.

Show us

smoothrunnings

hero member

Activity: 630

Merit: 501

Quote from: twmz on February 11, 2014, 11:30:15 AM

Quote from: smoothrunnings on February 11, 2014, 11:16:18 AM

Quote from: bitpop on February 11, 2014, 10:38:07 AM

No 9332 is both mining and web page
9333 is p2p

So I am back to my original question. How do I change the website front-end port?

The answer is still -w. You can't change the port for only the website front-end without changing the port for miners. The port for the front-end and the miners is always the same port.

That's not true. I have seen P2Pool sites that have bitcoin set to port 9332 but when you try to access the front-end get a message that no page exists, and some have bitcoin on port 9332 and the front-end on a different port like port 80.

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 427. (Read 2591964 times)