[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 578.

Subo1977

sr. member

Activity: 344

Merit: 250

Flixxo - Watch, Share, Earn!

Quote from: PatMan on March 28, 2013, 02:33:42 AM

Quote from: Aseras on March 27, 2013, 09:14:45 PM

Quote from: gyverlb on March 27, 2013, 05:41:54 AM

Quote from: Aseras on March 26, 2013, 08:26:23 PM

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over.

Did you try the avalon branch of p2pool on github? A fix for stratum is there.

Yes it's not a fix for stratum. It disables work caching. I get 150mhash with it using stratum. Not exactly ideal.

Stratum for avalon on p2pool is terribly broken. I don't know if its p2pool, or cgminer version or a combination of one doing something the other doesn't like. I actually feel like it should be an easy fix since every other stratum pool works fine.

So far the only thing that works is using fix-protocol and getwork with a very high diff. I've been using /4000+16. That seems the best balance. +32 is all late. You have to remember the "miners" are only 300mhash each. It will bring a 3930k running Ubuntu with 32gb of ram to its knees. It pegs 4 processors trying to keep up and eventually crashes as it runs out of memory. I also is 25-35% DOA. You loose 1/3 of your hashrate.

Agreed, but I'm not sure it's a cgminer thing TBH - although kudos to ckolivas for putting the notion in. I've tried several different miners on p2pool since the stratum code was implemented in an effort to get it working correctly, but to no avail. Like you said, every other stratum pool works fine.

Although I'm sorry that you're not able to use your avalon on p2pool, (it must be very frustrating - & I was looking forward to seeing the results), it is actually quite a relief to finally get some confirmation that the stratum code on p2pool is borked after banging my head against a wall for so long, not to mention a few "experts" living in denial. If there's a plus to all this - then maybe now, finally & at long last, we can get this bug ironed out, get stratum working properly and p2pool back to 100% working order as it was before the stratum code was introduced. This can only be for the benefit of everyone.

It would be a disaster if this opportunity was not taken. Lets see......

+1

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Quote from: Aseras on March 27, 2013, 09:14:45 PM

Quote from: gyverlb on March 27, 2013, 05:41:54 AM

Quote from: Aseras on March 26, 2013, 08:26:23 PM

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over.

Did you try the avalon branch of p2pool on github? A fix for stratum is there.

Yes it's not a fix for stratum. It disables work caching. I get 150mhash with it using stratum. Not exactly ideal.

Stratum for avalon on p2pool is terribly broken. I don't know if its p2pool, or cgminer version or a combination of one doing something the other doesn't like. I actually feel like it should be an easy fix since every other stratum pool works fine.

So far the only thing that works is using fix-protocol and getwork with a very high diff. I've been using /4000+16. That seems the best balance. +32 is all late. You have to remember the "miners" are only 300mhash each. It will bring a 3930k running Ubuntu with 32gb of ram to its knees. It pegs 4 processors trying to keep up and eventually crashes as it runs out of memory. I also is 25-35% DOA. You loose 1/3 of your hashrate.

Agreed, but I'm not sure it's a cgminer thing TBH - although kudos to ckolivas for putting the notion in. I've tried several different miners on p2pool since the stratum code was implemented in an effort to get it working correctly, but to no avail. Like you said, every other stratum pool works fine.

Although I'm sorry that you're not able to use your avalon on p2pool, (it must be very frustrating - & I was looking forward to seeing the results), it is actually quite a relief to finally get some confirmation that the stratum code on p2pool is borked after banging my head against a wall for so long, not to mention a few "experts" living in denial. If there's a plus to all this - then maybe now, finally & at long last, we can get this bug ironed out, get stratum working properly and p2pool back to 100% working order as it was before the stratum code was introduced. This can only be for the benefit of everyone.

It would be a disaster if this opportunity was not taken. Lets see......

Aseras

hero member

Activity: 658

Merit: 500

Quote from: -ck on March 27, 2013, 06:08:26 PM

Quote from: Aseras on March 25, 2013, 11:19:58 PM

Well someone's gotta do it. Ill see what I can do. I can code and tinker. I just don't like to much. I'm gonna be gone a fair bit of April. So if you want to screw with my units you can but you may have to call one of my techs to hard reset them if you bork one and can't reset it remotely.

It's an offer if you are interested in doing it right Cheesy

Ps I've probably given you 50 coins by now, over the past couple years. Keep up the good work.

Thanks. With xiangfu currently working on bringing the code into line with mainline cgminer, it would be helpful for me to test on an actual unit, and I have some time next week. However I would not be touching the p2pool software code at all since it is in python and a complete mystery to me.

Yes, agreed. I tried today to go looking since I used to program python and perl. God some of that code is nuts. Forrest is a genius or a mad hatter. Lol.

It's just too complex for me to dive in and know even where to look. I'm sure if Forrest had an avalon and a few hours he could nail it down.

Aseras

hero member

Activity: 658

Merit: 500

Quote from: gyverlb on March 27, 2013, 05:41:54 AM

Quote from: Aseras on March 26, 2013, 08:26:23 PM

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over.

Did you try the avalon branch of p2pool on github? A fix for stratum is there.

Yes it's not a fix for stratum. It disables work caching. I get 150mhash with it using stratum. Not exactly ideal.

Stratum for avalon on p2pool is terribly broken. I don't know if its p2pool, or cgminer version or a combination of one doing something the other doesn't like. I actually feel like it should be an easy fix since every other stratum pool works fine.

So far the only thing that works is using fix-protocol and getwork with a very high diff. I've been using /4000+16. That seems the best balance. +32 is all late. You have to remember the "miners" are only 300mhash each. It will bring a 3930k running Ubuntu with 32gb of ram to its knees. It pegs 4 processors trying to keep up and eventually crashes as it runs out of memory. I also is 25-35% DOA. You loose 1/3 of your hashrate.

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Quote from: Aseras on March 25, 2013, 11:19:58 PM

Well someone's gotta do it. Ill see what I can do. I can code and tinker. I just don't like to much. I'm gonna be gone a fair bit of April. So if you want to screw with my units you can but you may have to call one of my techs to hard reset them if you bork one and can't reset it remotely.

It's an offer if you are interested in doing it right Cheesy

Ps I've probably given you 50 coins by now, over the past couple years. Keep up the good work.

Thanks. With xiangfu currently working on bringing the code into line with mainline cgminer, it would be helpful for me to test on an actual unit, and I have some time next week. However I would not be touching the p2pool software code at all since it is in python and a complete mystery to me.

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Quote from: kano on March 27, 2013, 12:08:46 AM

Quote from: Aseras on March 26, 2013, 08:26:23 PM

...
The highest p2pool would let me go is 6535. Any higher number just comes back as 6535.
...

Better get that fixed fast ...

That and the rest...... Tongue

If ever there was a perfect opportunity for the devs here to go through the p2pool stratum code with a fine comb & spyglass - this is it. Some of us here have been suggesting this for months.....ie: since it was introduced.

But that's another story.....or is it? Grin

stevegee58

legendary

Activity: 916

Merit: 1003

Quote from: BR0KK on March 27, 2013, 08:17:06 AM

Could someone point me a way to installing p2pool on a raspberry pi (256 Model)?

Is that even possible or something i shouldn't do?

I was able to get it working on a BeagleBone with Ubuntu. It was a PITA since so many of the required Python packages were missing and I had to download them one at a time until it worked.
I couldn't come up with a reason to actually use that since I already have a spare Linux PC running in the house and I use the BB for other things.

BR0KK

hero member

Activity: 784

Merit: 500

Could someone point me a way to installing p2pool on a raspberry pi (256 Model)?

Is that even possible or something i shouldn't do?

gyverlb

hero member

Activity: 896

Merit: 1000

Quote from: Aseras on March 26, 2013, 08:26:23 PM

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over.

Did you try the avalon branch of p2pool on github? A fix for stratum is there.

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Quote from: Aseras on March 26, 2013, 08:26:23 PM

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over. There's something wrong with the way it reads the stratum response. Someone, Jeff I think, said it has a problem with double byte responses that p2pool makes. On getwork, it just hammers the server and the DOA rate is >25%. Using a very high diff 3000-4000 helps a tiny bit. The highest p2pool would let me go is 6535. Any higher number just comes back as 6535.

Even soloing to a bitcoind alone will crash. It's too much. Hence the need to run a buffer like eloipool between it.

https://github.com/forrestv/p2pool/commit/5f061e6c6753adf93acf04b8463badef88c4106e

Well cgminer has changed a lot since the version included with Avalon with lots of bugfixes and improvements, and I can't attest to any code added to the Avalon release so... usual disclaimers apply. On the upside, Xiangfu has said to me on IRC that he will be updating the code to bring it in line with mainline cgminer. But then I'm still out of the loop and only obliged to audit his code if he wants to push it upstream to me.

rav3n_pl

legendary

Activity: 1361

Merit: 1003

Don`t panic! Organize!

Quote from: kano on March 27, 2013, 12:08:46 AM

Quote from: Aseras on March 26, 2013, 08:26:23 PM

...
The highest p2pool would let me go is 6535. Any higher number just comes back as 6535.
...

Better get that fixed fast ...

I think it is in getwork.py:

Code:

'target': pack.IntType(256).pack(self.share_target).encode('hex'),

kano

legendary

Activity: 4592

Merit: 1851

Linux since 1997 RedHat 4

Quote from: Aseras on March 26, 2013, 08:26:23 PM

...
The highest p2pool would let me go is 6535. Any higher number just comes back as 6535.
...

Better get that fixed fast ...

Aseras

hero member

Activity: 658

Merit: 500

Quote from: -ck on March 26, 2013, 08:03:14 PM

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

Avalon Cgminer without fix protocol connects on stratum but resends all work over and over. There's something wrong with the way it reads the stratum response. Someone, Jeff I think, said it has a problem with double byte responses that p2pool makes. On getwork, it just hammers the server and the DOA rate is >25%. Using a very high diff 3000-4000 helps a tiny bit. The highest p2pool would let me go is 6535. Any higher number just comes back as 6535.

Even soloing to a bitcoind alone will crash. It's too much. Hence the need to run a buffer like eloipool between it.

https://github.com/forrestv/p2pool/commit/5f061e6c6753adf93acf04b8463badef88c4106e

-ck

legendary

Activity: 4088

Merit: 1631

Ruu \o/

Sounds like the problem is all in the p2pool end anyway? Can you tell if it's actually successfully connecting directly stratum because that would be essential to avoid p2pool sending a lot and higher diffs would be essential to avoid p2pool receiving a lot.

mdude77

legendary

Activity: 1540

Merit: 1001

Quote from: Aseras on March 26, 2013, 07:11:08 PM

Well I can't get it stable. I tried it under Ubuntu, under python, under windows binaries and python. Still crashes once python/p2pool eats ~1.2gb of ram it blows up and takes down bitcoind. I tried the stratum proxy, old getwork mining proxies,you name it. Nothing will run more than a few hours.

I may try baking my own firmware with some alternative builds, but for now its not working. I can mine on btcguild or slush or eloipool solo and not have any problems.

I'm going to offer a starting bounty of 10BTC for anyone who gets the avalon working properly and stable* on p2pool.

*Ill define properly as a reasonable hashrate and DOA rate ( ~25% on fix-protocol ) and stable as 1 week uptime without p2pool or bitcoind crashing. The avalon itself is buggy and restarts sometimes.

Why is python/p2pool using so much memory? I thought the memory leaks were fixed?

What's different about an Avalon from a normal GPU? The obvious thing to comes to mind is the massive increase in hash rate. Does p2pool properly scale to keep the difficulty to the miner at a reasonable pace? Did you try manually forcing the difficulty higher?

According to https://en.bitcoin.it/wiki/P2Pool, I think you just add /xxxxx to your username to increase the share difficulty. At 60gh, to get 5 shares an hour, I think it's saying use /10000. But as many times as I've looked at this page, and the link it refers to https://bitcointalksearch.org/topic/m.816322, I still don't understand what it's trying to say.

M

Aseras

hero member

Activity: 658

Merit: 500

Well I can't get it stable. I tried it under Ubuntu, under python, under windows binaries and python. Still crashes once python/p2pool eats ~1.2gb of ram it blows up and takes down bitcoind. I tried the stratum proxy, old getwork mining proxies,you name it. Nothing will run more than a few hours.

I may try baking my own firmware with some alternative builds, but for now its not working. I can mine on btcguild or slush or eloipool solo and not have any problems.

I'm going to offer a starting bounty of 10BTC for anyone who gets the avalon working properly and stable* on p2pool.

*Ill define properly as a reasonable hashrate and DOA rate ( ~25% on fix-protocol ) and stable as 1 week uptime without p2pool or bitcoind crashing. The avalon itself is buggy and restarts sometimes.

aspirez

newbie

Activity: 41

Merit: 0

Hi i have question,
can i run public pool, using p2pool?
Is there some limit of the max connections?

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Oh lordy.

I wonder how many other people from this pool have ordered Avalons thinking they're gonna work wonders on p2pool?

Maybe p2pool will be the last bastion for gpu mining?...... Grin

Aseras

hero member

Activity: 658

Merit: 500

Quote from: -ck on March 25, 2013, 11:13:30 PM

Quote from: Aseras on March 25, 2013, 10:57:48 PM

I'm using the latest git pull.

What do you need? the avalon modified cgminer source code? Need to light a fire under Yifu, zhang and xiangfu to get a real working copy. https://github.com/BitSyncom/cgminer.git

You can modify the software/firmware. I've been doing it all day. I pulled down the latest firmware and hacked the crap out of it and got rid of things I didn't want. Freed up some room. Enough to tinker and add my own scripts and such.

https://en.bitcoin.it/wiki/Avalon#How_to_Compile_Your_Own_WR703N_Firmware_for_Avalon

Well I don't know why you're taking this on yourself... I appreciate the gesture but it's not remotely trivial. Let me explain why it's now not that simple. As the avalon code was developed in secret outside of cgminer development, they worked on a codebase that is now redundant, hacking into it in a way where the only driver that works is for avalon. Cgminer's usb code and queueing has completely changed since then. To bring the avalon code into line would require large chunks of code to be rewritten to suit these changes. This is the danger of writing code out of the main tree. Now porting it to the new driver and queueing model and requires quite a bit of time effort and testing, and needs someone to support it (or not if it's abandoned). This is not in the scope of "lend me access to your mining hardware for a few hours over ssh". Undecided

I suspect Xiangfu will eventually be forced to keep his code in sync with the main cgminer git tree.

Yes I'm bitter about the whole Avalon experience, seeing them mine 10s of thousands of dollars worth of bitcoin each day with software mostly written by us, without them engaging us at all (except to get into a bunfight with Kano) or us earning a cent for our part, but it's not your fault. Lips sealed

Well someone's gotta do it. Ill see what I can do. I can code and tinker. I just don't like to much. I'm gonna be gone a fair bit of April. So if you want to screw with my units you can but you may have to call one of my techs to hard reset them if you bork one and can't reset it remotely.

It's an offer if you are interested in doing it right Cheesy

Ps I've probably given you 50 coins by now, over the past couple years. Keep up the good work.

Aseras

hero member

Activity: 658

Merit: 500

Quote from: kano on March 25, 2013, 11:06:51 PM

Quote from: Aseras on March 25, 2013, 11:03:42 PM

And p2pool crashed again... Keeps imploding and dropping connection to bitcoind. Gonna have to send Forrest some logs.

The avalons are mining fine, better since I went to U/2000+1 without diff they hammer the piss out of the server. Still 10-15% DOA, but better than 25% I started with.

The BFLs will work fine on p2pool - get one of them ... whenever they are available Tongue

Meanwhile ... now what did I say to GitSyncom about "proper" Avalon support ... they sure aren't providing anything like that ...

Well it's better than anything anyone else has right now, save asicminer. And he's probably using the same thing just in bigger cards I bet.

I've had enough fun for the day. I gotta get to bed. I have to fly to Italy and go play with a lidar unit on a NATO ship. Fun fun.. Glad I got my units in before I left.

Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool - page 578. (Read 2591920 times)