[BBR] Boolberry: Privacy and Security - Guaranteed Since 2014

[Neo.op]

I am bit confused. when I do show_hr it tells me 1090310 - is it 1.09 khs or something else?
@dga - now the solomine seems to be suffering from the same issue:

ubuntu@ip-10-0-0-214:~/boolberry/build/release/src$ ./simpleminer --pool-addr localhost:10102 --pass x --login 1AswkzJ3bYE6UZC7x8ssvyKMnRY8xX5vxghKWzKxgpFWhv7EZaozuXWDpXSpFR26qU4ig2b32FquuXt X9NsqdNK9SewvqTv
2014-Jun-23 13:16:18.511374 Connecting localhost:10102....
2014-Jun-23 13:16:18.512253 Connected localhost:10102 OK
2014-Jun-23 13:16:18.513835 Getting scratchpad...
2014-Jun-23 13:16:20.019629 Scratchpad received ok, size: 8432Kb, heigh=26283
2014-Jun-23 13:16:20.025188 Getting next job...

stuck but mining via boolbd works.

1090310 == 1.09mh/s.

I'll grab an AWS instance and see if I can get simpleminer to stick like that.  Does it get stuck also with extremepool?

Yep
ubuntu@ip-10-0-0-214:~/boolberry/build/release/src$ ./simpleminer --pool-addr bbr.extremehash.com:5555 --pass x --login 1AswkzJ3bYE6UZC7x8ssvyKMnRY8xX5vxghKWzKxgpFWhv7EZaozuXWDpXSpFR26qU4ig2b32FquuXt X9NsqdNK9SewvqTv
2014-Jun-23 14:30:23.322102 Connecting bbr.extremehash.com:5555....
2014-Jun-23 14:30:23.331967 Connected bbr.extremehash.com:5555 OK
2014-Jun-23 14:30:23.341861 Getting scratchpad...
2014-Jun-23 14:30:24.225721 Scratchpad received ok, size: 8446Kb, heigh=26326
2014-Jun-23 14:30:24.229200 Getting next job...

Do you use HVM instance ?

Ubuntu.
complete noob at this so ran the build in one window and started the daemon ./boolbd at one
In another window tried to solo mine.

[1714046844]

1714046844

[1714046844]

1714046844

[1714046844]

1714046844

[1714046844]

1714046844

[cubydu]

I am bit confused. when I do show_hr it tells me 1090310 - is it 1.09 khs or something else?
@dga - now the solomine seems to be suffering from the same issue:

ubuntu@ip-10-0-0-214:~/boolberry/build/release/src$ ./simpleminer --pool-addr localhost:10102 --pass x --login 1AswkzJ3bYE6UZC7x8ssvyKMnRY8xX5vxghKWzKxgpFWhv7EZaozuXWDpXSpFR26qU4ig2b32FquuXt X9NsqdNK9SewvqTv
2014-Jun-23 13:16:18.511374 Connecting localhost:10102....
2014-Jun-23 13:16:18.512253 Connected localhost:10102 OK
2014-Jun-23 13:16:18.513835 Getting scratchpad...
2014-Jun-23 13:16:20.019629 Scratchpad received ok, size: 8432Kb, heigh=26283
2014-Jun-23 13:16:20.025188 Getting next job...

stuck but mining via boolbd works.

1090310 == 1.09mh/s.

I'll grab an AWS instance and see if I can get simpleminer to stick like that.  Does it get stuck also with extremepool?

Yep
ubuntu@ip-10-0-0-214:~/boolberry/build/release/src$ ./simpleminer --pool-addr bbr.extremehash.com:5555 --pass x --login 1AswkzJ3bYE6UZC7x8ssvyKMnRY8xX5vxghKWzKxgpFWhv7EZaozuXWDpXSpFR26qU4ig2b32FquuXt X9NsqdNK9SewvqTv
2014-Jun-23 14:30:23.322102 Connecting bbr.extremehash.com:5555....
2014-Jun-23 14:30:23.331967 Connected bbr.extremehash.com:5555 OK
2014-Jun-23 14:30:23.341861 Getting scratchpad...
2014-Jun-23 14:30:24.225721 Scratchpad received ok, size: 8446Kb, heigh=26326
2014-Jun-23 14:30:24.229200 Getting next job...

Do you use HVM instance ?

Ubuntu.
complete noob at this so ran the build in one window and started the daemon ./boolbd at one
In another window tried to solo mine.

You don't understand me. EC2 has two type virtualization - paravirtualization and HVM . HVM it's hardware virtualization. For huge pages you must
use HVM.

[Neo.op]

You don't understand me. EC2 has two type virtualization - paravirtualization and HVM . HVM it's hardware virtualization. For huge pages you must
use HVM.

Oh sorry. How do I check my virtualisation?
Edit: Got it. raising another instance.

[Sy]

You can remove 79.135.200.108 from first post, it shutdown - only one pool left?

[dga]

Update to build instructions for wallet on MacOS specifically - the cmake prefix path should be slightly different:

cmake -D BUILD_GUI=TRUE -D CMAKE_PREFIX_PATH=/Users/dga/Qt/5.3/clang_64 -D CMAKE_BUILD_TYPE=Release ../..

With that and the MacOS build changes, it builds and runs perfectly on MacOS.  Still kicking around how to easily make an app bundle for it - any Mac dev gurus want to chime in?

[dga]

You don't understand me. EC2 has two type virtualization - paravirtualization and HVM . HVM it's hardware virtualization. For huge pages you must
use HVM.

Oh sorry. How do I check my virtualisation?
Edit: Got it. raising another instance.

Ahh, I see.  It sounds like hugepages are interacting poorly with Amazon's virtualization in PVM mode:

http://www.pythian.com/blog/linux-hugepages-for-oracle-on-amazon-ec2-possible-but-not-convenient-easy-or-fully-supported/

The miner will work properly without hugepages - just a little slower - so if you are running on a PVM, not HVM, amazon instance, please do *not* enable hugepages via sysctl.

[Neo.op]

You don't understand me. EC2 has two type virtualization - paravirtualization and HVM . HVM it's hardware virtualization. For huge pages you must
use HVM.

Oh sorry. How do I check my virtualisation?
Edit: Got it. raising another instance.

Ahh, I see.  It sounds like hugepages are interacting poorly with Amazon's virtualization in PVM mode:

http://www.pythian.com/blog/linux-hugepages-for-oracle-on-amazon-ec2-possible-but-not-convenient-easy-or-fully-supported/

The miner will work properly without hugepages - just a little slower - so if you are running on a PVM, not HVM, amazon instance, please do *not* enable hugepages via sysctl.

yep looks like it. Here's after I was able to do everything correctly. I guess finally mining at 1.3khs avg hr - 1389725

Let me make some BBR before I tip you   Btw, any ideas how can I get this running on reboot?

[cubydu]

Neo.op

Code:

ubuntu@ip-10-0-0-119:~$ sudo vim /etc/init/miner.conf

/etc/init/miner.conf

Code:

start on runlevel [2345]
stop on runlevel [06]

respawn

script

exec /usr/bin/sudo /home/ubuntu/boolberry/build/release/src/simpleminer  --pool-addr=10.0.0.145:7777 --login=1FUtZJ9Ao8qg1MiNa6RfRJKY4aFH885wD13Lf6U7xL4FS58f7i6CYL58s8dg2aV8maRzzaQG1c1kCTMDw8LKAEG5LQsW7M8 --pass=z --mining-threads 32
end script

Code:

sudo service miner start

check miner

Code:

sudo less /var/log/upstart/miner.log

And please, don't ask how use vim. You can choose another editor.

[Neo.op]

Neo.op

Code:

ubuntu@ip-10-0-0-119:~$ sudo vim /etc/init/miner.conf

/etc/init/miner.conf

I am noob but not a complete one   got it to run via cron

[btc-mike]

You can remove 79.135.200.108 from first post, it shutdown - only one pool left?

There is another pool - http://bbr.mine-hard.com/

I will get the first post updated.

[cubydu]

Neo.op

Code:

ubuntu@ip-10-0-0-119:~$ sudo vim /etc/init/miner.conf

/etc/init/miner.conf

I am noob but not a complete one   got it to run via cron

Upstart scripts run when server starts .It's not same as cron

[Sy]

Whats a good value for /proc/sys/vm/nr_hugepages?

Each hugepage is 2MB.  To be safe, I'd go for something like 64.  128MB isn't so much that you will run out of memory elsewhere, but is more than enough for the miner.

The miner will print out a warning if it's not able to allocate enough memory in hugepages, so you can also use that as a guide.

I'm getting good reports of both speed and stability from people, so I've submitted a pull request to c_z for the most recent changes:
  * Linux speed optimizations by hugepages
  * Fixing MacOS build (renamed a macro in the new windows optimized Keccak)

Cheers, i've set it to 128 earlier, testing 64.

Both the i7 870 and the i5 4440 run on pretty much the same hashrate, Ubuntu 14.04 Server.

[dga]

Btw - one suggestion for pool operators:  Could you please indicate on your announcement post and on your site whether you're setting the donation flag or not?  I'm a fan of this feature - I think it gives c_z and his team incentive to keep developing and increasing the value of BBR over time, compared to the bogosity of a pre-mine, and this benefits anyone taking a risk on the currency - and would like to make sure I mine in a way that supports it.  Other people may feel exactly the opposite and only want to use non-donate-supporting pools.  Thanks!

[uvt9]

I got this error by running my own compiled boolbd on Ubuntu 14.04:

Code:

2014-Jun-23 17:48:56.103704 ERROR /home/XXXXXXXX/boolberry/src/daemon/daemon.cpp:214[int main(int, char**)]Exception at [main], what=locale::facet::_S_create_c_locale name not valid

Other binaries such as simplewallet and simpleminer running fine, don't know what's the problem with boolbd.

[veritasBS]

Is the algo being fixed so one miner does not take the majority of the blocks? Please tell me somthing is being done to resolve this issue.

[ddink7]

Is the algo being fixed so one miner does not take the majority of the blocks? Please tell me somthing is being done to resolve this issue.

I *almost* started to get excited about this coin, until I did a quick Google search and remembered this is the one that Christian @ ccminer is "closed source" GPU mining. He claims to be making 12 BTC per day, and understandably will not open his mining source to anyone.

I'm afraid that using such an obscure algo is going to be the death of this project. There's simply no way that one person mining the plurality of your coins is healthy.

[otila]

I got this error by running my own compiled boolbd on Ubuntu 14.04:

Code:

2014-Jun-23 17:48:56.103704 ERROR /home/XXXXXXXX/boolberry/src/daemon/daemon.cpp:214[int main(int, char**)]Exception at [main], what=locale::facet::_S_create_c_locale name not valid

What boost version? Does it work if you start it with
LC_ALL=C ./boolbd

[dga]

Is the algo being fixed so one miner does not take the majority of the blocks? Please tell me somthing is being done to resolve this issue.

I *almost* started to get excited about this coin, until I did a quick Google search and remembered this is the one that Christian @ ccminer is "closed source" GPU mining. He claims to be making 12 BTC per day, and understandably will not open his mining source to anyone.

I'm afraid that using such an obscure algo is going to be the death of this project. There's simply no way that one person mining the plurality of your coins is healthy.

Actually he almost has a point. Yes, the project is currently 10% complete. The price, doubtless, reflects that--$90 DRK is coming =)

Feeling a little worried? :-)

I've used up my quota of pointing out the speed boosts that have gone in recently to the CPU miner, both for pooled and solo, to ensure that the playing field is even, the EC2 analyses that suggest that no covert GPU miner is managing nearly as much of the hash rate as you claim, or cbuchner1's statements that they were only mining that much at 1/5th the difficulty.

Instead, I'm just going to note that every time more FUD like this gets thrown around, I feel a strong urge to continue contributing as much code and time as I can to help this berry float in the face of such annoyingly childish attempts to promote ones own bag at the expense of promoting technological advancement and a discussion of the actual merits of the coins involved.  Crypto is a grand social, economic, and technological experiment -- let's try to make it a constructive one instead of trying to tear down things we view as potential competitors.

(end rant)

[veritasBS]

Is the algo being fixed so one miner does not take the majority of the blocks? Please tell me somthing is being done to resolve this issue.

I *almost* started to get excited about this coin, until I did a quick Google search and remembered this is the one that Christian @ ccminer is "closed source" GPU mining. He claims to be making 12 BTC per day, and understandably will not open his mining source to anyone.

I'm afraid that using such an obscure algo is going to be the death of this project. There's simply no way that one person mining the plurality of your coins is healthy.

Actually he almost has a point. Yes, the project is currently 10% complete. The price, doubtless, reflects that--$90 DRK is coming =)

Feeling a little worried? :-)

I've used up my quota of pointing out the speed boosts that have gone in recently to the CPU miner, both for pooled and solo, to ensure that the playing field is even, the EC2 analyses that suggest that no covert GPU miner is managing nearly as much of the hash rate as you claim, or cbuchner1's statements that they were only mining that much at 1/5th the difficulty.

Instead, I'm just going to note that every time more FUD like this gets thrown around, I feel a strong urge to continue contributing as much code and time as I can to help this berry float in the face of such annoyingly childish attempts to promote ones own bag at the expense of promoting technological advancement and a discussion of the actual merits of the coins involved.  Crypto is a grand social, economic, and technological experiment -- let's try to make it a constructive one instead of trying to tear down things we view as potential competitors.

(end rant)

There is no FUD attempt by me...this is the claim on this board and elsewhere...i own over 50btc worth of BBR. I think BBR is a great coin in every respect other than the GPU miner that claims to be taking many of the blocks....i would like an explanation as to what is being done to even the playing field.

[dga]

Is the algo being fixed so one miner does not take the majority of the blocks? Please tell me somthing is being done to resolve this issue.

I *almost* started to get excited about this coin, until I did a quick Google search and remembered this is the one that Christian @ ccminer is "closed source" GPU mining. He claims to be making 12 BTC per day, and understandably will not open his mining source to anyone.

I'm afraid that using such an obscure algo is going to be the death of this project. There's simply no way that one person mining the plurality of your coins is healthy.

Actually he almost has a point. Yes, the project is currently 10% complete. The price, doubtless, reflects that--$90 DRK is coming =)

Feeling a little worried? :-)

I've used up my quota of pointing out the speed boosts that have gone in recently to the CPU miner, both for pooled and solo, to ensure that the playing field is even, the EC2 analyses that suggest that no covert GPU miner is managing nearly as much of the hash rate as you claim, or cbuchner1's statements that they were only mining that much at 1/5th the difficulty.

Instead, I'm just going to note that every time more FUD like this gets thrown around, I feel a strong urge to continue contributing as much code and time as I can to help this berry float in the face of such annoyingly childish attempts to promote ones own bag at the expense of promoting technological advancement and a discussion of the actual merits of the coins involved.  Crypto is a grand social, economic, and technological experiment -- let's try to make it a constructive one instead of trying to tear down things we view as potential competitors.

(end rant)

There is no FUD attempt by me...this is the claim on this board and elsewhere...i own over 50btc worth of BBR. I think BBR is a great coin in every respect other than the GPU miner that claims to be taking many of the blocks....i would like an explanation as to what is being done to even the playing field.

Sorry if I responded too grumpily.  I've been putting a lot of work into trying to improve the miners for exactly this reason, and I'm probably being trolled too easily (not by you).

I can't speak for the project itself - I'm just doing this because I too own a lot of BBR at this point and am trying to increase its value by contributing code.  Kind of a fun experiment, really.  But there are a few prongs to it that I've seen:

(a)  Reality vs hype:  How much GPU mining is really taking place, and is it a significant factor in the coin?

It could be, but there's also a lot of "headroom" in the coin at this point.  Compared to, e.g., XMR, where the price of the coin is almost exactly pegged to the spot price on Amazon (see aminorex's analyses of this), BBR is more-often EC2 profitable.  Given that EC2 is one of the most expensive possible ways to mine a coin, this suggests to me that the private GPU miner is not the dominant factor in BBR's difficulty (and thus, volume of coins).  I've posted some analyses of these along the same lines as aminorex's ongoing analysis of the marginal cost of XMR mining to try to make sure that the mining situation is transparent and clear to people participating.

(b)  Improved pool mining:  I've been contributing a sequence of optimizations to simpleminer to improve the efficiency of pool mining.  There's a reason to do it beyond efficiency:  Smaller miners are more likely to pool mine, and I'm trying to put them on a more even footing with the cloud miners who solo... (myself included.  I do a lot of EC2 mining of BBR when it's profitable.)

This effort is not yet complete, but I should have the final chunk of it by next Monday (longpoll support).

(c) Improved CPU mining efficiency:  C_Z imported an optimized Keccak that boosted Windows mining speed by roughly 15% about two days ago, and I have an outstanding pull request for he just merged my optimizations that improve the CPU miner's efficiency by up to 57% on Linux (realistically, 15-20% on machines with lots of L3 cache, such as EC2 machines, and 50% on smaller desktop CPUs).

(d)  A growing bounty for stratum support, which was the prerequisite to cbuchner1 open sourcing his miner.

This one's the most obvious one, and I encourage you and anyone else who already has a significant amount of BBR to contribute to the bounty.  It will leave BBR in a very healthy position from a technical standpoint - an open source GPU miner, which is more than is available for the other cryptonotes at this point.

As you probably gathered from my grumpy and hasty response to the previous post, I'm annoyed about this situation mostly because it seems like it provides the FUD machine with a lot of ammunition.  If you have other suggestions for things, please do chime in - I will try to find time to implement some more.