Pages:
Author

Topic: [ANN][BLC] Blakecoin Blake-256 for GPU/FPGA With Merged Mined Pools Stable Net - page 10. (Read 409639 times)

sr. member
Activity: 409
Merit: 250
Next things now:

I moved through the eloipool installation guide from Ignatius, but it seems the pool is not working yet. I doublechecked the needed changes in config.py mentioned by Ignatius. These should be fine.

How i start:
First start blakecoind by run-blakecoind.sh

Code:
./blakecoind \
        -datadir=/home/osboxes/Desktop/Blake/BlakeData/ -daemon\
        -conf=/home/osboxes/Desktop/Blake/BlakeData/blakecoin.conf \
        -blocknotify=./newblock.sh

There the problems begin (i think).. Anytime a new block is found on the network (blakecoind has to be running properly, otherwise it wouldnt notice a new block on the network) it starts the notify process for the pool

Code:
#!/bin/sh

killall -USR1 /home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py

but as answer i just get:

Code:
/home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py: no process found

While eloipool.py is running issue the command "ps", what path is shown for eloipool.py? If it differs that is the issue, otherwise experiment to find the right killall usage. What OS was this again?


The path on newblock.sh is correct, when i try to just start eloipool.py with this path i get:

Code:
osboxes@osboxes:~/Desktop/Blake$ /home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py
Traceback (most recent call last):
  File "/home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py", line 82, in
    import jsonrpc
ImportError: No module named 'jsonrpc'

Install python-jsonrpc package? It looks like eloipool crashes on startup due to a missing dependency. Fix this issue first. The killall is probably correct, it fails to kill eloipool because eloipool is not running.  You are very close now, good luck and welcome to the Blake ecosystem!
jr. member
Activity: 75
Merit: 3
Next things now:

I moved through the eloipool installation guide from Ignatius, but it seems the pool is not working yet. I doublechecked the needed changes in config.py mentioned by Ignatius. These should be fine.

How i start:
First start blakecoind by run-blakecoind.sh

Code:
./blakecoind \
        -datadir=/home/osboxes/Desktop/Blake/BlakeData/ -daemon\
        -conf=/home/osboxes/Desktop/Blake/BlakeData/blakecoin.conf \
        -blocknotify=./newblock.sh

There the problems begin (i think).. Anytime a new block is found on the network (blakecoind has to be running properly, otherwise it wouldnt notice a new block on the network) it starts the notify process for the pool

Code:
#!/bin/sh

killall -USR1 /home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py

but as answer i just get:

Code:
/home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py: no process found

The path on newblock.sh is correct, when i try to just start eloipool.py with this path i get:

Code:
osboxes@osboxes:~/Desktop/Blake$ /home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py
Traceback (most recent call last):
  File "/home/osboxes/Desktop/eloipool_Blakecoin/eloipool.py", line 82, in
    import jsonrpc
ImportError: No module named 'jsonrpc'
So its there, but i think if i try to start it manually there are arguments missing. Or is jsonrpc really missing? Thought the path to the rpc is given in the runit.sh (python-bitcoinrpc)

Code:
#!/bin/sh

PYTHONPATH=/path/to/python-bitcoinrpc:/path/to/python-base58:/path/to/midstate \
     nohup ./eloipool.py 2>&1 >/dev/null &

-----

Second thing is - if i run the runit.sh i get one line of console output

Code:
osboxes@osboxes:~/Desktop$ ./runit.sh
osboxes@osboxes:~/Desktop$ nohup: redirecting stderr to stdout

Thats it. The logfile is empty - even after minutes and hours. So i think there is exactly nothing running.

Any thoughts? Any gladly done mistakes, that i may have overseen?
Maybe there are again problems with the dependancies. I had several problems to get them under the links, that were mentioned by Ignatius, but otherwise found them on your Git folder.
legendary
Activity: 1509
Merit: 1030
Solutions Architect
Yup, block matured and after that the coins could be transfered properly into my main wallet.

Just to say it: I wouldnt wonder if i made a mistake at some point, as i'm not used to compiling manually - i'm a Windows user (you know what i want to say).
It just seems i have problems with boost. Two days ago i had another error, so i couldnt start the blakecoind. In the error message stood something of boost too - i had to reboot the system, the it worked again. So maybe its the Ubuntu version or just stupidity of myself^^

I do think ubuntu version can make a difference as it dictates which dep versions you have access too at least without compiling them all from source

will look into making some compatibility changes to source to help with these as you are not the only one to report issues so its not stupidity  Wink

it is clear if you found a block and it has matured ok that its working just has an issue with cleanly exiting most likely related boost  

think its good that you have had a go as nothing can beat a bit of real experience and hopefully feedback from you and other can go some way to making current code better(16.04+) even the new code base I have been working on has issues that give me a hard time, it all sounds easy but reality is that even for btc the code base is a real monolithic spaghetti monster  Cheesy  
jr. member
Activity: 75
Merit: 3
Yup, block matured and after that the coins could be transfered properly into my main wallet.

Just to say it: I wouldnt wonder if i made a mistake at some point, as i'm not used to compiling manually - i'm a Windows user (you know what i want to say).
It just seems i have problems with boost. Two days ago i had another error, so i couldnt start the blakecoind. In the error message stood something of boost too - i had to reboot the system, the it worked again. So maybe its the Ubuntu version or just stupidity of myself^^
legendary
Activity: 1509
Merit: 1030
Solutions Architect
hmm strange if I get chance will try to replicate the issue but it should not be doing that either way for mining if the wallet is running it should be fine even if its not exiting cleanly, is the block you found maturing ok?
jr. member
Activity: 75
Merit: 3
I tried to stop the server with

Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData stop

but it didnt work first. When i tried to restart the server, it said it seem to be running (cannot obtain lock on data directory). I again had to kill the process.


Then i tried

Code:
sudo ./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData stop

and got the following response on the first try:

Code:
terminate called after throwing an instance of 'boost::exception_detail::clone_impl >'
  what():  boost: mutex lock failed in pthread_mutex_lock: Invalid argument

The server was stopped (obv. by the exception).

After that i tried it again. The server stopped without any exception and could be restarted without any process kill. But sometimes i have to call the stop command two times until the server really stops (sometime the try to lock the data dir fails, when i only stopped the server once).

If i just try
Code:
sudo ./blakecoind stop
it fails, because the datadir isnt the one blakecoind wants to set without params.


BTW. i used the linked guide from Ignatius to install the eloipool. But as for now, i just want to make sure blakecoind is running well before trying to use mining it over eloipool (and then trying to get into merged mining, if i come that far).
legendary
Activity: 1509
Merit: 1030
Solutions Architect
As eloipool returned as a dead pool, i tried to run blakecoind in solo config first - just to test if it works alone.
After a few struggles with connecting my miners to the blakecoind on my virtual ubuntu machine, i got it to work and it even found a block this night (wanted to test the whole functionality). The block is still immature, but i think this will work too.

I now have the following problem:
i start blakecoind by
Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData -daemon
and it says "Blakecoin server starting"

Now i have a problem to stop that server.
What i did:
Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData -rpcusername=myusername -rpcpassword=mypassword stop
and it says "Blakecoin server stopping"

So far so good. But if i want to restart the server i get "Error: Cannot obtain a lock on data directory /home/osboxes/Desktop/Blake/BlakeData. Blakecoin is probably already running."

If first have to kill the process by
Code:
sudo pkill -9 -f blakecoind
to be able to restart the server.

So now, the question is: Whats wrong. Did i miss something?

I just want to clear things for myself before trying further with eloipool and maybe merged mine. But if i fail to use the simple solo version, how should i succeed in the more complicated.

no need to pass arg when stopping

Code:
./blakecoind stop

and if you want to make life simpler just set all your details in conf and avoid passing via command

example blakecoin.conf would be like (note the rpc allow can also be like 192.168.1.* and the rpc port)

Code:
listen=1
rpcallowip=127.0.0.1
rpcuser=myusername
rpcpassword=mypassword
rpcport=8772
logtimestamps=1
server=1
daemon=1
addnode=146.185.135.24
addnode=104.238.177.36
addnode=104.238.133.35
addnode=107.191.58.189
addnode=108.61.242.204
addnode=107.170.178.92

Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData

Ignatius did a private merged pool setup guide here Here
jr. member
Activity: 75
Merit: 3
As eloipool returned as a dead pool, i tried to run blakecoind in solo config first - just to test if it works alone.
After a few struggles with connecting my miners to the blakecoind on my virtual ubuntu machine, i got it to work and it even found a block this night (wanted to test the whole functionality). The block is still immature, but i think this will work too.

I now have the following problem:
i start blakecoind by
Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData -daemon
and it says "Blakecoin server starting"

Now i have a problem to stop that server.
What i did:
Code:
./blakecoind -datadir=/home/osboxes/Desktop/Blake/BlakeData -rpcusername=myusername -rpcpassword=mypassword stop
and it says "Blakecoin server stopping"

So far so good. But if i want to restart the server i get "Error: Cannot obtain a lock on data directory /home/osboxes/Desktop/Blake/BlakeData. Blakecoin is probably already running."

If first have to kill the process by
Code:
sudo pkill -9 -f blakecoind
to be able to restart the server.

So now, the question is: Whats wrong. Did i miss something?

I just want to clear things for myself before trying further with eloipool and maybe merged mine. But if i fail to use the simple solo version, how should i succeed in the more complicated.
legendary
Activity: 1509
Merit: 1030
Solutions Architect
Thank you Bluedragon.
It seemed i did it the wrong way (sorry, i have no experience with building, or its to long ago).

Now i tried it the way described in src/build-unix.md - so did
Code:
make -f makefile.unix

out of the src folder and it ran through (besides a ton of warnings) - it even seems blakecoind is syncronizing (folder is growing).

The issues mentioned with boost... i made the change that was described there before running make.

But at all... be aware of more questions and despair to come Wink

no problem  Grin

the most common commands I run to build for ubuntu 12.04-16.04
Code:
sudo git clone https://github.com/BlueDragon747/Blakecoin
cd blakecoin/src

cd src/leveldb
chmod +x Makefile
chmod +x build_detect_platform
sudo make libleveldb.a libmemenv.a
cd ..

sudo make -f makefile.unix USE_UPNP=0

sudo strip blakecoind

if your on a home network behind nat/firewall best to set UPNP=1 but for public nodes with public ips I turn it off to avoid exploit crashing nodes
jr. member
Activity: 75
Merit: 3
Thank you Bluedragon.
It seemed i did it the wrong way (sorry, i have no experience with building, or its to long ago).

Now i tried it the way described in src/build-unix.md - so did
Code:
make -f makefile.unix

out of the src folder and it ran through (besides a ton of warnings) - it even seems blakecoind is syncronizing (folder is growing).

The issues mentioned with boost... i made the change that was described there before running make.

But at all... be aware of more questions and despair to come Wink
legendary
Activity: 1509
Merit: 1030
Solutions Architect
So now the beginning of a long long torturous way Wink

Trying to compile blakecoind on Ubuntu 17.04 32 Bit i get the following error:

Code:
/usr/bin/ld: cannot find -lboost_system-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_filesystem-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_program_options-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_thread-mgw49-mt-s-1_55
collect2: error: ld returned 1 exit status
Makefile:315: recipe for target 'blakecoin-qt' failed
make: *** [blakecoin-qt] Error 1

a few packages that were mentioned as dependancies could not be installed in the original way as mentioned on github. I installed them as followed:

Quote
sudo apt-get install git-core build-essential libssl-dev libboost-all-dev libdb5.3-dev libdb5.3++-dev libgtk2.0-dev libminiupnpc-dev qt4-qmake mingw-w64 synaptic qt-sdk qt4-dev-tools libqt4-dev libqtcore4 libqtgui4 libdb++-dev

The bold marked packages are the ones i had to change, to get them installed. Maybe these needed changes are the problem?
Any other hints on the fail while doing make?

I found 16.04 great but 17.04 not so much need to make some changes and its just a PITA

if you change the deps need to make changes to the makefile.unix to reflect which ones you do have *from the error they are all boost issues so check which one and edit makefile

you can get issues with newer boost check this out

https://github.com/bitcoin/bitcoin/pull/6114/files

or panopolis did some updates for newer distros here

https://github.com/panopolis/Blakecoin
jr. member
Activity: 75
Merit: 3
So now the beginning of a long long torturous way Wink

Trying to compile blakecoind on Ubuntu 17.04 32 Bit i get the following error:

Code:
/usr/bin/ld: cannot find -lboost_system-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_filesystem-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_program_options-mgw49-mt-s-1_55
/usr/bin/ld: cannot find -lboost_thread-mgw49-mt-s-1_55
collect2: error: ld returned 1 exit status
Makefile:315: recipe for target 'blakecoin-qt' failed
make: *** [blakecoin-qt] Error 1

a few packages that were mentioned as dependancies could not be installed in the original way as mentioned on github. I installed them as followed:

Quote
sudo apt-get install git-core build-essential libssl-dev libboost-all-dev libdb5.3-dev libdb5.3++-dev libgtk2.0-dev libminiupnpc-dev qt4-qmake mingw-w64 synaptic qt-sdk qt4-dev-tools libqt4-dev libqtcore4 libqtgui4 libdb++-dev

The bold marked packages are the ones i had to change, to get them installed. Maybe these needed changes are the problem?
Any other hints on the fail while doing make?
newbie
Activity: 21
Merit: 0
Okay so it worked on my usb windows version. Must be some conflicting software I guess then. Probably the sdkbi installed to compile some other miner

yeah must be a conflict but glad to hear you got it working in the end  Cool

It's a band-aid solution to say the least it still doesn't work on my pc.
I Even uninstalled many programs to see if that work, it didn't.
Only thing left is that it's either avast or some windows update got corrupt on me.

Since my portable windows has no update on it nor any antivirus ATM. I'll try installing avast on it and even upgrading the drivers and see how it goes.

legendary
Activity: 1509
Merit: 1030
Solutions Architect
Okay so it worked on my usb windows version. Must be some conflicting software I guess then. Probably the sdkbi installed to compile some other miner

yeah must be a conflict but glad to hear you got it working in the end  Cool
newbie
Activity: 21
Merit: 0
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry

Yes I had used DDU. But it failed even then. Trying on a fresh install windows on usb with 16.9.2 if it works then it means there's something conflicting in windows if it doesn't idk what.

The thing is someone I know had it running on latest driver with 580, and on another PC with some old driver with 390

But it doesn't run on mine

yeah that is weird not sure what else you could try it should be working  Undecided

can you try these and see if any give different result on 16.9.2
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password
and
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T

oh and if your not using cgminer.conf rename it something like cgminer.conf.backup

Okay so it worked on my usb windows version. Must be some conflicting software I guess then. Probably the sdkbi installed to compile some other miner
newbie
Activity: 21
Merit: 0
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry

Yes I had used DDU. But it failed even then. Trying on a fresh install windows on usb with 16.9.2 if it works then it means there's something conflicting in windows if it doesn't idk what.

The thing is someone I know had it running on latest driver with 580, and on another PC with some old driver with 390

But it doesn't run on mine

yeah that is weird not sure what else you could try it should be working  Undecided

can you try these and see if any give different result on 16.9.2
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password
and
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T

oh and if your not using cgminer.conf rename it something like cgminer.conf.backup

It crashes the first one just shows
Code:
 [2018-01-09 17:02:47] Started cgminer 3.1.1 

and closes

The other command I ran as is and with
Code:
--verbose
pasting the results. Both closed down as well.
Code:
 G:\cgminer-alt-pool-la1>cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T
 [2018-01-09 17:02:47] Started cgminer 3.1.1
 [2018-01-09 17:02:47] Probing for an alive pool

G:\cgminer-alt-pool-la1>cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T --verbose
 [2018-01-09 17:02:55] Started cgminer 3.1.1
 [2018-01-09 17:02:55] CL Platform 0 vendor: Advanced Micro Devices, Inc.
 [2018-01-09 17:02:55] CL Platform 0 name: AMD Accelerated Parallel Processing
 [2018-01-09 17:02:55] CL Platform 0 version: OpenCL 2.0 AMD-APP (2117.13)
 [2018-01-09 17:02:55] Platform 0 devices: 1
 [2018-01-09 17:02:55]  0       Hawaii
 [2018-01-09 17:02:55] GPU 0 AMD Radeon (TM) R9 390 Series hardware monitoring enabled
 [2018-01-09 17:02:55] Probing for an alive pool
 [2018-01-09 17:02:55] Testing pool stratum+tcp://eu3.blakecoin.com:3334
 [2018-01-09 17:02:56] Stratum authorisation success for pool 0
 [2018-01-09 17:02:56] Pool 0 stratum+tcp://eu3.blakecoin.com:3334 alive
 [2018-01-09 17:02:56] Init GPU thread 0 GPU 0 virtual GPU 0
 [2018-01-09 17:02:56] CL Platform vendor: Advanced Micro Devices, Inc.
 [2018-01-09 17:02:56] CL Platform name: AMD Accelerated Parallel Processing
 [2018-01-09 17:02:56] CL Platform version: OpenCL 2.0 AMD-APP (2117.13)
 [2018-01-09 17:02:56] List of devices:
 [2018-01-09 17:02:56]  0       Hawaii
 [2018-01-09 17:02:56] Selected 0: Hawaii

G:\cgminer-alt-pool-la1>
newbie
Activity: 21
Merit: 0
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry

Yes I had used DDU. But it failed even then. Trying on a fresh install windows on usb with 16.9.2 if it works then it means there's something conflicting in windows if it doesn't idk what.

The thing is someone I know had it running on latest driver with 580, and on another PC with some old driver with 390

But it doesn't run on mine

yeah that is weird not sure what else you could try it should be working  Undecided

can you try these and see if any give different result on 16.9.2
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password
and
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T

oh and if your not using cgminer.conf rename it something like cgminer.conf.backup

Will do. Taking a bit of time to set up usb windows. Once I am done with it I'll boot into normal windows and do this Smiley
legendary
Activity: 1509
Merit: 1030
Solutions Architect
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry

Yes I had used DDU. But it failed even then. Trying on a fresh install windows on usb with 16.9.2 if it works then it means there's something conflicting in windows if it doesn't idk what.

The thing is someone I know had it running on latest driver with 580, and on another PC with some old driver with 390

But it doesn't run on mine

yeah that is weird not sure what else you could try it should be working  Undecided

can you try these and see if any give different result on 16.9.2
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password
and
Quote
cgminer.exe --blake256 -o stratum+tcp://eu3.blakecoin.com:3334 -u loneSilentWolf.r9390 -p password -T

oh and if your not using cgminer.conf rename it something like cgminer.conf.backup
newbie
Activity: 21
Merit: 0
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry

Yes I had used DDU. But it failed even then. Trying on a fresh install windows on usb with 16.9.2 if it works then it means there's something conflicting in windows if it doesn't idk what.

The thing is someone I know had it running on latest driver with 580, and on another PC with some old driver with 390

But it doesn't run on mine
legendary
Activity: 1509
Merit: 1030
Solutions Architect
I tried with 16.9.2 aswell ut it's the same result it crashes after that.

could it be a problem because of amd sdk installed on my pc to compile some other miner?

could you help me compile it with latest sdk?

Did you use DDU to remove drivers?  I have a 290 here running win 10 and driver 16.9.2 and its working not sure there is any difference between the 290 - 390 apart from faster ram

yeah maybe

I dont have it setup for compiling the miner atm sorry
Pages:
Jump to: