[ANN][YAC][POOL]p2pool for yacoin!!

[eule]

Looks like pos+pow combo is painfull when reorganize and orphan lots of blocks :/
Thats why I change payment for last 10 blocks, orphans will happen.

Also updated git to make share chain persist and not download it every time on pool restart.

So it should work now? Updated to try.  
@YacLives: To autostart services use crontab, i.e.

Code:

crontab -e user 

and put  

Code:

@reboot cd /home/user/ && ./yacoind
@reboot cd /home/user/p2pool-yac && ./p2pool_start_command

into it. Pay attention to the formatting of the file. To restart crashed services use an admin panel like webmin that has process monitoring, you can set rules what to do if a process crashes.
Not sure about errorchecking, it's probably possible to grep for some common errors but that's too high for me. ^^
edit: Just got a payout yay! [edit: it appears that payout was from the 10 minutes i ran the classic version...]
Slightly cluesless if i should go back to the classic version (with over 50% DOA and 5x the cpu usage) or not... [Yes i switched back for tonight and just got another payout]

[x3maniac]

Looks like pos+pow combo is painfull when reorganize and orphan lots of blocks :/
Thats why I change payment for last 10 blocks, orphans will happen.

Also updated git to make share chain persist and not download it every time on pool restart.

i had mine set to true since the started with your code few days ago.

[YacLives]

Looks like pos+pow combo is painfull when reorganize and orphan lots of blocks :/
Thats why I change payment for last 10 blocks, orphans will happen.

Also updated git to make share chain persist and not download it every time on pool restart.

So it should work now? Updated to try.  
@YacLives: To autostart services use crontab, i.e.

Code:

crontab -e user 

and put  

Code:

@reboot cd /home/user/ && ./yacoind
@reboot cd /home/user/p2pool-yac && ./p2pool_start_command

into it. Pay attention to the formatting of the file. To restart crashed services use an admin panel like webmin that has process monitoring, you can set rules what to do if a process crashes.
Not sure about errorchecking, it's probably possible to grep for some common errors but that's too high for me. ^^
edit: Just got a payout yay! [edit: it appears that payout was from the 10 minutes i ran the classic version...]
Slightly cluesless if i should go back to the classic version (with over 50% DOA and 5x the cpu usage) or not... [Yes i switch back for tonight]

Thanks mate ill play with cron again, and btw the orig network is stating its 25% atm not 50

and if your having issues with mem check ur mem usage here

Code:

free -m

you may need to make a swap file it could slow things down but wont be an issue if you are running out of mem and crashing anywas

Code:

 sudo dd if=/dev/zero of=/swapfile1 bs=1024 count=524288 
 sudo mkswap /swapfile1
 sudo chown root:root /swapfile1
 sudo chmod 0600 /swapfile1
 sudo swapon /swapfile1
# if you dont have nano 
 sudo apt-get install nano
#Then you edit your fstab
 sudo nano /etc/fstab

#apend the following

/swapfile1 swap swap defaults 0 0

#Then you can check it is working with 
free -m 

i got a node to work on a free ec2 micro instance doing this  

EDIT: i forgot to mention its mainly for a openvz setups, if you are getting the error "swapon failed: Operation not permitted" then its most likely your hosts does not permit it and need to consider changing hosts
 These guys have possibly the worst customer service on the planet and will take ages to get your server going in some cases not in mine but i was polite and kept sending emails
but once you are setup with one its great $15 euro a month https://host-unlimited.de/vserver

[eule]

No worries i transferred my node from my cheap 512 MB VPS to a quite powerful i7 root running some gameservers for friends. DOA improved significantly, but i wouldn't have done it if i hadn't messed up my python install yesterday on the VPS, no rescue system and too lazy now for manual debian install over ssh which will surely be a pita...
Running the classic version now aswell with low 25% DOA indeed which is nice, just ran into the "too many files open" error which i hopefully fixed now. YAC mining is pretty adventurous, unlike those boring sha256 and scrypt coins.  
edit: Since rav3n's p2pool found 3 valid blocks again i switched once more, hopefully the last time.  
edit2: LOL i have one vps at host-unlimited too, you're totally right about their customer service but the servers are pretty fast for the money (i had it hashing away at 350 kHash when Yacoin was released, now it's around 30) when you finally got it after a week or more. Post at their forum or ask per mail if you have RAM issues, they often give 2gb servers to 12 gb customers and hope nobody notices, check your free memory with top and mention kmemsize in the forum post/mail if you got less than the advertised 12 gb.

[YacLives]

No worries i transferred my node from my cheap 512 MB VPS to a quite powerful i7 root running some gameservers for friends. DOA improved significantly, but i wouldn't have done it if i hadn't messed up my python install yesterday on the VPS, no rescue system and too lazy now for manual debian install over ssh which will surely be a pita...
Running the classic version now aswell with low 25% DOA indeed which is nice, just ran into the "too many files open" error which i hopefully fixed now. YAC mining is pretty adventurous, unlike those boring sha256 and scrypt coins.  
edit: Since rav3n's p2pool found 3 valid blocks again i switched once more, hopefully the last time.  
edit2: LOL i have one vps at host-unlimited too, you're totally right about their customer service but the servers are pretty fast for the money (i had it hashing away at 350 kHash when Yacoin was released, now it's around 30) when you finally got it after a week or more. Post at their forum or ask per mail if you have RAM issues, they often give 2gb servers to 12 gb customers and hope nobody notices, check your free memory with top and mention kmemsize in the forum post/mail if you got less than the advertised 12 gb.

yes i found if i email their customer service at 12pm german time when they start work....looooooool i get pretty fast response i also got my server running in under a day doing the same thing on 3 different occasions i just give them my id and the payment method used so they can track it and set it up

 i think their biggest problem is that they dont have any IT customer service workers so if your reading this live near Gifhorn in Germany apply to them because they are looking for workers as they have over 20k customers and are not copeing lol http://bieber-edv.de/jobs https://host-unlimited.de/kontakt

[Aggrophobia]

so you know how to setup p2pools for yac, did you forked them alone?

i ask due to the fact that Onecoin which uses another scrypt didn't have a p2pool

anyone willingly to help at this point? :/

[x3maniac]

i think it's happeninig again. last 5 block orphns

Code:

94188	Fri Jun 14 2013 22:00:12 GMT-0400 (EDT)	0000000014041567626a53b12a712f3c2ad5db2af5e316397048efc24067a097	→
94187	Fri Jun 14 2013 21:59:07 GMT-0400 (EDT)	00000000d6e1d199dbb257297ee821bf97ef2bba6865bc9512a34e997f084fe9	→
94167	Fri Jun 14 2013 21:41:25 GMT-0400 (EDT)	00000000b4ecce6c8c9fb8f70774af458b8b1cb7a59f9d3c815f42ed72f09bd6	→
94154	Fri Jun 14 2013 21:33:25 GMT-0400 (EDT)	0000000062ae0b29fdd2e384d0cdb04160bbab273cc7089b47364820bcd8f59c	→
94123	Fri Jun 14 2013 21:09:31 GMT-0400 (EDT)	00000000acdc3b237f89848bbd8a5057785265bfba9faa7ca32d264230297f56	→
94104	Fri Jun 14 2013 20:48:40 GMT-0400 (EDT)	00000000b71495cb571d8c1cca72899442d269d48cc9b884cec5e830bd78cbe2	→

[tamehuang]

i can't run windows binary on win2008(64bit) server... it shows me some error..

I guess the "some" part is important here ....  

Lol. Isn't "some" error fixed with the "any" key?

not yet. i got the error message like this when i run the binary which you provide. it seems like that the variable 'bundle_files' cause that.
mybe a win32 binary can't runs fine on a win2008 64bit system.
hope that you can compile a win64 one for us

2013-06-10 23:45:59.106000 > Error handling message: (see RECV line)
2013-06-10 23:45:59.108000 > Traceback (most recent call last):
2013-06-10 23:45:59.108000 >   File "twisted\internet\tcp.pyc", line 209, in _dataReceived
2013-06-10 23:45:59.109000 >
2013-06-10 23:45:59.109000 >   File "p2pool\p2p.pyc", line 154, in new_dataReceived
2013-06-10 23:45:59.113000 >
2013-06-10 23:45:59.116000 >   File "p2pool\util\p2protocol.pyc", line 26, in dataReceived
2013-06-10 23:45:59.118000 >
2013-06-10 23:45:59.118000 >   File "p2pool\util\datachunker.pyc", line 40, in _DataChunker
2013-06-10 23:45:59.122000 >
2013-06-10 23:45:59.123000 > --- <exception caught here> ---
2013-06-10 23:45:59.124000 >   File "p2pool\util\p2protocol.pyc", line 54, in dataReceiver
2013-06-10 23:45:59.128000 >
2013-06-10 23:45:59.129000 >   File "p2pool\p2p.pyc", line 94, in packetReceived
2013-06-10 23:45:59.130000 >
2013-06-10 23:45:59.131000 >   File "p2pool\util\p2protocol.pyc", line 68, in packetReceived
2013-06-10 23:45:59.135000 >
2013-06-10 23:45:59.136000 >   File "p2pool\p2p.pyc", line 316, in handle_sharereply
2013-06-10 23:45:59.137000 >
2013-06-10 23:45:59.138000 >   File "p2pool\data.pyc", line 53, in load_share
2013-06-10 23:45:59.138000 >
2013-06-10 23:45:59.139000 >   File "p2pool\data.pyc", line 345, in __init__
2013-06-10 23:45:59.139000 >
2013-06-10 23:45:59.140000 >   File "p2pool\bitcoin\networks.pyc", line 45, in <lambda>
2013-06-10 23:45:59.141000 >
2013-06-10 23:45:59.142000 >   File "zipextimporter.pyc", line 98, in load_module
2013-06-10 23:45:59.144000 >
2013-06-10 23:45:59.153000 > exceptions.ImportError: MemoryLoadLibrary failed loading yac_scrypt.pyd
2013-06-10 23:45:59.158000 Lost peer 95.138.185.176:12579 - Connection was close
d cleanly.

[eule]

Did you install the yac scrypt module (python setup.py install in the yac-script dir)?

Yeah orphan rate is pretty high today... luckily it's not permanent. Still wonder why this p2pool version is so prone to orphans.

[tamehuang]

Did you install the yac scrypt module (python setup.py install in the yac-script dir)?

Yeah orphan rate is pretty high today... luckily it's not permanent. Still wonder why this p2pool version is so prone to orphans.

no. i have not compiled the code. i just run the binary. shouldn't a binary exe file includes all lib and pyd what it needs?

[eule]

Well i never ran p2pool on Windows, so I'm not sure. But i can tell you that on Linux it won't work without installing the yac scrypt module, i guess it's the same for Windows.

[x3maniac]

Did you install the yac scrypt module (python setup.py install in the yac-script dir)?

Yeah orphan rate is pretty high today... luckily it's not permanent. Still wonder why this p2pool version is so prone to orphans.

i'm pretty sure this is the problem. you gotta check your router / firewall settings. using this new code for 10 sec shares causes high amounts of tcp connections to be open, hence you guys were getting "too many files open" with this new code over the original p2pool source. digging deeper my firewall is dropping connection once it reaches it's peak open connection count causing orphans. well really causing our pools to be out of sync per say.

to resolve this issue: (I believe, i can be wrong about this)

Main node must be restarted to drop all the connection and start again. but with in 8 - 10 hours it will be the same again.

or

change firewall to drop tcp connection every 5 min or less.

[eule]

By main node i take it you mean the bootstrap address from the networks.py? Interesting observations.

[x3maniac]

By main node i take it you mean the bootstrap address from the networks.py? Interesting observations.

yup

[eule]

3 hours of orphans this morning...  
edit: Would adding more nodes to the bootstrap resolve the issue maybe? Don't think so, but just a thought.

[rav3n_pl]

My node is offline, some major electric work in my home. Sorry.
Will add all nodes I have to GIT in few hrs.
You can post addresses to run new nodes using -n to catch in.

[eule]

OK i replaced your address with the bitvolcano address for now in the networks.py. It's a bit weird, the bitvolcano and procrypto nodes seem to run the same version (and found the same orphan blocks), but when i changed the bootstrap address my node started "fresh", i.e. empty graph.

[rav3n_pl]

No need to mess in code, just run it:
run_p2pool -n address
It will bootstrap from that address.

[caish5]

OK i replaced your address with the bitvolcano address for now in the networks.py. It's a bit weird, the bitvolcano and procrypto nodes seem to run the same version (and found the same orphan blocks), but when i changed the bootstrap address my node started "fresh", i.e. empty graph.

To fix that you need to change persist to True in the networks.py file.

[eule]

Thanks for both your answers, very helpful.