[ANN] Stratum mining protocol - ASIC ready

[newguy05]

Hi guys, apologize in advance for the newbie question. I just got my stratum proxy setup on my mac osx following the standard instructions, all default setting.  Everything seem to be running fine and i am seeing a spam of the below text repeated.

1) Is the below text means everything is running okie? the lines that starts with WARNING is that anything to worry about?
2) Also for the same WARNING lines, it ends with "diff 1"  what does that mean exactly?  difficulty 1? still dont understand what that means.

Thank you!

Code:

2013-06-03 20:38:53,237 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,318 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,365 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,515 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,524 INFO proxy jobs.submit # Submitting 238b0469
2013-06-03 20:38:53,553 INFO proxy jobs.submit # Submitting 025c9983
2013-06-03 20:38:53,592 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,601 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,616 WARNING proxy getwork_listener._on_submit # [91ms] Share from 'gagaliya.worker1' accepted, diff 1
2013-06-03 20:38:53,645 WARNING proxy getwork_listener._on_submit # [92ms] Share from 'gagaliya.worker1' accepted, diff 1
2013-06-03 20:38:53,697 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,750 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,842 INFO proxy jobs.submit # Submitting 64c43d35
2013-06-03 20:38:53,852 INFO proxy jobs.submit # Submitting 21705f0e
2013-06-03 20:38:53,873 INFO proxy jobs.submit # Submitting fcb480cb
2013-06-03 20:38:53,892 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,932 WARNING proxy getwork_listener._on_submit # [89ms] Share from 'gagaliya.worker1' accepted, diff 1
2013-06-03 20:38:53,944 WARNING proxy getwork_listener._on_submit # [92ms] Share from 'gagaliya.worker1' accepted, diff 1
2013-06-03 20:38:53,948 INFO proxy getwork_listener._on_authorized # Worker 'gagaliya.worker1' asks for new work
2013-06-03 20:38:53,963 WARNING proxy getwork_listener._on_submit # [90ms] Share from 'gagaliya.worker1' accepted, diff 1

[MineForeman.com]

Hi All,

I feel a bit of a dead head, usually when I have questions I just go to the code but I am having a bit of a problem with stratum-mining-proxy and my python knowledge is very sub par.  I have been aware of the custom-user capabilities of it and want to take advantage of it with;-

./mining_proxy.py -v --host=stratum.bitcoin.cz --port=3333 --custom-user=MinePeon.Donate --custom-password=Donate

The problem is it still uses the authorise credentuals supplied by the miner dispite showing;-

Code:

2013-06-19 01:16:25,727 WARNING proxy mining_proxy.on_connect # Authorizing custom user MinePeon.Donate, password Donate
2013-06-19 01:16:25,728 DEBUG protocol protocol.writeJsonRequest # < {"params": ["MinePeon.Donate", "Donate"], "id": 1, "method": "mining.authorize"}

In the startup.  Am I missing something?  Is this not the intention of those parameters?

Neil

[MineForeman.com]

Found a ticket for the problem on github and added some more info;-

https://github.com/slush0/stratum-mining-proxy/issues/22

If anyone is willing to tackle this I am willing to put a bounty on it, would 1BTC be acceptable?

Neil

[MWNinja]

Hi All,

I feel a bit of a dead head, usually when I have questions I just go to the code but I am having a bit of a problem with stratum-mining-proxy and my python knowledge is very sub par.  I have been aware of the custom-user capabilities of it and want to take advantage of it with;-

./mining_proxy.py -v --host=stratum.bitcoin.cz --port=3333 --custom-user=MinePeon.Donate --custom-password=Donate

The problem is it still uses the authorise credentuals supplied by the miner dispite showing;-

Code:

2013-06-19 01:16:25,727 WARNING proxy mining_proxy.on_connect # Authorizing custom user MinePeon.Donate, password Donate
2013-06-19 01:16:25,728 DEBUG protocol protocol.writeJsonRequest # < {"params": ["MinePeon.Donate", "Donate"], "id": 1, "method": "mining.authorize"}

In the startup.  Am I missing something?  Is this not the intention of those parameters?

Neil

I got it working on the beaglebone, your parameter usage isn't correct.

should be:
./mining_proxy.py -o stratum.bitcoin.cz -p 3333

then you connect your miner to the proxy and provide the pool user/pass there
./cgminer -o 127.0.0.1:3333 -u MinePeon.Donate -p Donate


If you are going to use the custom_user options, then don't specify a user/pass on the miner, or make them null ""; the point is so that miners can remote connect without revealing user/pass to eliminate a potential man-in-the-middle attack.
EDIT doesn't seem to work, it connects to the pool but either uses the credentials supplied by the miner or pukes. I'm no python expert either but will take a crack at it.

MOAR edit. I think I've figured out what the feature does finally.  It overrides the submission of shares to go to the account you specify, but it looks like you still need to supply valid credentials at the miner.  Going to make a few accounts and test.



If you are feeling generous, hit the donation address in my sig.  We have similar projects, so I look forward to collaboration (rather than name calling ala kano/luke-jr).

[MineForeman.com]

MOAR edit. I think I've figured out what the feature does finally.  It overrides the submission of shares to go to the account you specify, but it looks like you still need to supply valid credentials at the miner.  Going to make a few accounts and test.

That is indeed what it is meant to do (or at least I think that is what is meant to do).  It does not seem to do it at the moment though.  I will get around to having a look, python doesn't look that hard but time is short for me at the moment.

We have similar projects, so I look forward to collaboration (rather than name calling ala kano/luke-jr).

I was not even aware of your project!  It is good there is someone to take care of the BeagleBone crowd!

I must warn you that I have one (and a lot of other devices) sitting here and do plan a MinePeon port one of these days (as you probably have a Pi port in the plans).

But yeah, I have no need of an internet arch enemy and would love to collaborate with you.  In bitcoin it is VITALLY important that there as much diversification as possible.

Neil

[MWNinja]

I must warn you that I have one (and a lot of other devices) sitting here and do plan a MinePeon port one of these days (as you probably have a Pi port in the plans).

Yeah, I have a Pi sitting here too
I checked out your distribution, and we did some things similarly.  There's a good Arch linux distribution for the bone so I don't expect it will take too much effort for you to port.  If I do a Pi release it will likely also be Arch based, as Angstrom on Pi looks like a nightmare.

[kano]

...
If you are feeling generous, hit the donation address in my sig.  We have similar projects, so I look forward to collaboration (rather than name calling ala kano/luke-jr).

Name calling?
Moron

[cypherdoc]

when i monitor this port on my server:  http://yourserveripaddress:8889/ any hashrate that i see is getting to the Bitcoin Network, correct?

[-ck]

Have other pools lost interest in implementing reconnect support?

[doublec]

Have other pools lost interest in implementing reconnect support?

What pools do and what pools don't support it?

[kano]

Have other pools lost interest in implementing reconnect support?

What pools do and what pools don't support it?

Good pools support it (so they don't force you to throw away your shares when a disconnect happens)
Bad pools don't support it.

[-ck]

Have other pools lost interest in implementing reconnect support?

What pools do and what pools don't support it?

As far as I'm aware, only eligius supports it. This is why I'm bringing it back up since it's not in the pool ops' interests to implement it but it is in the miners' interests.

[doublec]

As far as I'm aware, only eligius supports it. This is why I'm bringing it back up since it's not in the pool ops' interests to implement it but it is in the miners' interests.

I'm under the impression that my pool, bitparking, and fireduck's pool, HHTT, support it unless I'm misunderstanding what you mean by reconnect support.

[-ck]

Have other pools lost interest in implementing reconnect support?

What pools do and what pools don't support it?

As far as I'm aware, only eligius supports it. This is why I'm bringing it back up since it's not in the pool ops' interests to implement it but it is in the miners' interests.

I see all the pool ops are doing an excellent job of ignoring this since most users are unaware of the lack of this support, so I'm reminding them to hopefully give this issue a higher profile.

[ATC]

Hi all,

I got many many errors as I use mining_proxy in win7 system. It reported as following:

Code:

stratum.custom_exceptions.TransportException: Not connected
Unhandled error in Deferred:
Unhandled Error
Traceback (most recent call last):
  File "twisted\web\server.pyo", line 156, in process

  File "twisted\web\server.pyo", line 191, in render

  File "twisted\web\resource.pyo", line 216, in render

  File "mining_libs\getwork_listener.pyo", line 163, in render_POST

--- <exception caught here> ---
  File "twisted\internet\defer.pyo", line 134, in maybeDeferred

  File "mining_libs\worker_registry.pyo", line 37, in authorize

  File "stratum\socket_transport.pyo", line 93, in rpc

It repeated those errors and only a very small fraction of work/shares ask and submit. So I'm wondering what's the matter?

BTW, how can I get the version information of mining_proxy?

Thanks.

[-ck]

Slush's document stipulated that it should refresh at least every 60 seconds. If it's much longer than that, since you're dealing with raw sockets that can quietly die, there is no way for the miner software to know if the socket has disappeared, so cgminer uses 90 seconds between messages to say the pool has died so you cannot increase it to much more. Updating it more frequently includes newer transactions but requires more refreshes on your mining software, and vice versa.

[DrHaribo]

Would faster refreshes give miners a slight edge of finding easy blocks?  If everybody and their mother is getting merke_branches every 60 seconds, and you get them every 35 seconds, if there is easy block to find, you would have a better chance of starting to work on it first.  But then again, it might be a toss since working longer on a given merkle_branch might actually help miner solve the block.

No, it won't give you any edge. As long as you are changing the nonce or merkle root and not hashing the same data as anyone else, any block variation is as likely as any other to meet the current target when you hash it. And don't believe people who say you have to try the entire nonce range before changing the merkle root, otherwise you'll "miss something". It's all superstition and misunderstandings.

[kano]

A 30s refresh (vs a 60s refresh) will increase your chance of getting higher transaction fees ...

[af_newbie]

A 30s refresh (vs a 60s refresh) will increase your chance of getting higher transaction fees ...

Thanks Kano.

Will try reducing the refresh interval.  Right now, I'm running with 150.  
I just found my first block: 255822 !!! Only 0.0667 in transactions.

[smoothrunnings]

So I gave the windows binaries a try, I installed the proxy on my Windows 7 64bit VM and connected my Avalon miners and ASIC USB key machine to the proxy. I noticed I get around 109ms and a lot of rejects show up on the avalons, yet if I connect direct to Slush's site I don't get nearly as many rejects on my Avalon's.

Is there a reason for this?

My network consists a Dell 6224 Gbit switch, CAT6 cable and a Cisco Gbit firewall.

Thanks,