[ANN] [POW] [MSR] Masari - simple, scalable, and secure cryptocurrency

[babanana]

http://puu.sh/xPMOH/0cb7037f37.png
Someone keeps making the hash rate and difficulty skyrocket.  

This is just a guess, but I think they're exploiting the lag in Masari's re-target time.  If you make a huge spike of hash for a little bit, you will get blocks very fast because the difficulty won't rise right away.  Then, once it is risen, you pull your hash and let other people mine on high difficulty for very slow blocktimes, until diff comes back down.  Then you repeat.

was miming this and riding the difficulty downtrend, then someone sent ridiculous amount of hashrate, I pulled my miner for now until difficulty is normal.

Hi dev, you might want to address this exploit issue as it can drive miners away. there was 26+ miners before now its down to 13.

Looks like dev dont know what he is doing. This shit will get palyed. If he no smart then this is shit worth nothing. Dev do the right thing and change change what is needed. Or you dont know a thing?
Make your coin stronger. Protect it from bad actors/players/shitters/hash-jumpers/space travellers.

Maybe you dont care because you already have the biggest coin collection.

I have released a mandatory upgrade version v0.1.2.0, containing a protocol upgrade that changes the difficulty algorithm to handle these flash miners.

https://github.com/masari-project/masari/releases/tag/v0.1.2.0
Linux binaries: https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-linux-x64-v0.1.2.0.tar.gz

Windows and Mac binaries will be available tomorrow, and this is released early with Linux binaries to protect the exchanges and the pools involved

Everybody please update ASAP, as this hard fork could be due much sooner with the flash mining at play!

Good. Perhaps you know a thing or two.  

[1715396779]

1715396779

[1715396779]

1715396779

[1715396779]

1715396779

[miyagiman]

http://puu.sh/xPMOH/0cb7037f37.png
Someone keeps making the hash rate and difficulty skyrocket.  

This is just a guess, but I think they're exploiting the lag in Masari's re-target time.  If you make a huge spike of hash for a little bit, you will get blocks very fast because the difficulty won't rise right away.  Then, once it is risen, you pull your hash and let other people mine on high difficulty for very slow blocktimes, until diff comes back down.  Then you repeat.

was miming this and riding the difficulty downtrend, then someone sent ridiculous amount of hashrate, I pulled my miner for now until difficulty is normal.

Hi dev, you might want to address this exploit issue as it can drive miners away. there was 26+ miners before now its down to 13.

Looks like dev dont know what he is doing. This shit will get palyed. If he no smart then this is shit worth nothing. Dev do the right thing and change change what is needed. Or you dont know a thing?
Make your coin stronger. Protect it from bad actors/players/shitters/hash-jumpers/space travellers.

Maybe you dont care because you already have the biggest coin collection.

I have released a mandatory upgrade version v0.1.2.0, containing a protocol upgrade that changes the difficulty algorithm to handle these flash miners.

https://github.com/masari-project/masari/releases/tag/v0.1.2.0
Linux binaries: https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-linux-x64-v0.1.2.0.tar.gz

Windows and Mac binaries will be available tomorrow, and this is released early with Linux binaries to protect the exchanges and the pools involved

Everybody please update ASAP, as this hard fork could be due much sooner with the flash mining at play!

Good. Perhaps you know a thing or two.  

Glad to hear there's an update! I was very put off when that miner came in.

[nanona]

Good. Perhaps you know a thing or two.  

Dev don't know shit.
Some guy on slack had to spoon-feed this bloody noob for days until he got it.
Instead of acknowledging the problem and working on a solution dev got cocky like a triggered snowflake.

[thaer]

Good. Perhaps you know a thing or two.  

Dev don't know shit.
Some guy on slack had to spoon-feed this bloody noob for days until he got it.
Instead of acknowledging the problem and working on a solution dev got cocky like a triggered snowflake.

Grade-school comments such as "dev got cocky like a triggered snowflake" don't help anybody nor adds context, and make you look like a noob.


As context for everybody here, a discussion was had yesterday regarding the difficulty calculation and I joined it by saying: "I'll  look into the difficulty calculation as well to see what can be done there" (was going to look into it after work, PST, v0.1.2.0 being the result)

Somebody was upset at the situation we have (due to be fixed in the upcoming hard fork), went through an unnecessary temper tantrum because it wasn't done right that very second, and everything he had said was filled with logical fallacies, mainly ad hominem attacks, and had little in terms of facts to back up what he was saying.

The said person in passing had indirectly mentioned "a coin you mentioned on bitcointalk" which prompted me to look into Sumokoin, and because it was an easy thing to check, within minutes I found the solution that addresses it by considering the statistical distribution of blocks, as well as the explanation of the issue (https://www.overleaf.com/articles/difficulty-adjustment-algorithms-in-cryptocurrency-protocols/ytcxbjvzrpbp/viewer.pdf ).

I worked on this immediately after my employment requirements were done for the day, and submitted v0.1.2.0 due in the upcoming protocol upgrade / hard fork.

Humans are emotional, and the user's emotions in the Slack group were getting the better of him, even though they were well intentioned. We'll all go through this though and everything is okay.

Cheers,
Thaer

[thaer]

Windows binaries are now also available (MacOS binaries will be compiled later today):
https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-win-x64-v0.1.2.0.zip
sha256sum: 79b2efaa996ab58479c08d1b5e7839026100bdf39a4a85c7a7252883e388b0cb

Linux binaries for reference: https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-linux-x64-v0.1.2.0.tar.gz
sha256sum: 2d01c137bf1c0f8abc15bd898d7a4e755e2c77ea53d3a90bc206b5aa842c0044

[nanona]

Both pools adjust the difficulty up to 10000H no matter which port is being used, which is bad.
Putting a fixed difficulty behind the address [address.diff] won't work with the latest Claymore.
It gives a parsing error and likely discards everything after the address.

[daninimmer]

Both pools adjust the difficulty up to 10000H no matter which port is being used, which is bad.
Putting a fixed difficulty behind the address [address.diff] won't work with the latest Claymore.
It gives a parsing error and likely discards everything after the address.

You need last versión >= v9.7 and add "-allpools 1"

NsGpuCNMiner -a cryptonight -allpools 1 -o stratum+tcp://masari.superpools.net:5555 -u YOUR_WALLET_ADDRESS.5000 -p x

I use this program with AMD and it works without problems

[daninimmer]

if you want to build it on ubuntu 16.04 (thats what im using) here are the steps:

Code:

sudo apt-get install build-essential cmake pkg-config libboost-all-dev libssl-dev libunbound-dev libminiupnpc-dev libunwind8-dev liblzma-dev libldns-dev libexpat1-dev doxygen graphviz

sudo apt-get install libgtest-dev && cd /usr/src/gtest && sudo cmake . && sudo make && sudo mv libg* /usr/lib/

git clone https://github.com/masari-project/masari.git

cd masari 

make 

to run the node you need to:

Code:

cd build/release/bin/

./masarid 

i usually use "screen" to run various nodes so instead of doing the "./masarid" , you do

Code:

screen -S "msrnode" ./masarid

then you run the wallet (same way)

Code:

screen -S "msrwallet" ./masari-wallet-cli 

and you create your wallet


... hope that helps a bit

Cant get this to make. I am stuck at

Code:

[ 34%] Building CXX object src/rpc/CMakeFiles/obj_rpc.dir/core_rpc_server.cpp.o

Any one having same issue. Installing on ubuntu instance.

I compile it on Ubuntu. Make sure you have enough memory.
What error gives you?

[nanona]

You need last versión >= v9.7 and add "-allpools 1"

NsGpuCNMiner -a cryptonight -allpools 1 -o stratum+tcp://masari.superpools.net:5555 -u YOUR_WALLET_ADDRESS.5000 -p x

I use this program with AMD and it works without problems

v10.2 is the latest version.
Those parameters you are using are not valid anymore.
It is xpool, xval and so on.
No, it doesn't work adding anything behind the address.

Still the pool shouldn't ramp the share difficulty up to 10000 when port 3333 or 5555 are used, when Claymore doesn't submit a fixed diff.

[babanana]

Good. Perhaps you know a thing or two.  

Dev don't know shit.
Some guy on slack had to spoon-feed this bloody noob for days until he got it.
Instead of acknowledging the problem and working on a solution dev got cocky like a triggered snowflake.

I have to join slack.

You should give the developer time to react. It will not be a right away solution but definitely a something must be done and he did.
If someone spoon-fed him, well, thank you to that someone. May be dev can say thank you as well to finish this non-sense.

My view about it, The dev wants to do it right away but hesitating until he found a good solution done by others(or spoon-fed by others).
Hope this fix address the inability of cryptonote algorithm to adjust to sudden big hash.

Good luck to your coin

[nanona]

Both pools adjust the difficulty up to 10000H no matter which port is being used, which is bad.

Correction. After connecting on 7777 the diff gets ramped up like this: 10000H, 20000H, 40000H, 80000H, 100001H.
This is where it remains and it is much too high. Maybe nice because low server load. But very inefficient mining.

If you are using auto-diff it must be such that shares get submitted every few seconds.
At 100001H literally minutes pass without a share getting submitted, then a new round starts and all the work was wasted.
Next round eventually the same or with some luck one of those ultra high diff shares is found.
Mostly it isn't and the work expires.

When connecting on 3333 it starts at 100H, then 200H, 400H, 800H, 1600H, 3200H, 6400H, 12800H, 25600H, 51200H and then goes to the famous 100001H and remains there.
It seems this auto-diff is not able to adjust down. It only goes up to the max.

I remember that until yesterday this wasn't the case. It went only to 10000H and remained there, which was good for my setup.
Mining at 100001H makes no sense. Not with high-end GPUs and even less so with CPUs.

[nanona]

You need last versión >= v9.7 and add "-allpools 1"

NsGpuCNMiner -a cryptonight -allpools 1 -o stratum+tcp://masari.superpools.net:5555 -u YOUR_WALLET_ADDRESS.5000 -p x

I use this program with AMD and it works without problems

There is no such thing as -a cryptonight in Claymore v9.7
Here is what the readme.txt says.

-a   Select algorithm:
"-a 0" (default) means autoselection.
"-a 1" fastest mining mode for 290/290x and 270/270x cards.
"-a 2" low power consumption mode for 290/290x cards. A bit less speed (1%) but less power consumption (10-15%).
"-a 3" fastest mining for old cards (6xxx and 5xxx).
"-a 4" fastest mining for recent cards (280x, 270/270x etc).

As for putting the diff behind the address in the -u parameter, it results in the following error in the console: Error parsing line 2 of config.txt
That's the line with -u address.5000

I assume your -a cryptonight for Claymore is a copy-paste error from your CPU miner instructions on your getting started page because it makes no sense for Claymore.
Did you ever actually try the -u address.diff in Claymore v9.7 or later versions and observe if it really works? Keep an eye on the console right after start.
Is there the parsing error I mentioned, in red color? If it is there, it means Claymore is not sending the diff to the server and the server will auto-diff.

[elGuilleAR]

Hello miners.

We have been the Masari Pool upgraded.

You may join us at http://msr.pools.e-scavo.net.ar at only 0.5%

Thanks Dev for your quick reaction.

[debrus]

if you want to build it on ubuntu 16.04 (thats what im using) here are the steps:

Code:

sudo apt-get install build-essential cmake pkg-config libboost-all-dev libssl-dev libunbound-dev libminiupnpc-dev libunwind8-dev liblzma-dev libldns-dev libexpat1-dev doxygen graphviz

sudo apt-get install libgtest-dev && cd /usr/src/gtest && sudo cmake . && sudo make && sudo mv libg* /usr/lib/

git clone https://github.com/masari-project/masari.git

cd masari 

make 

to run the node you need to:

Code:

cd build/release/bin/

./masarid 

i usually use "screen" to run various nodes so instead of doing the "./masarid" , you do

Code:

screen -S "msrnode" ./masarid

then you run the wallet (same way)

Code:

screen -S "msrwallet" ./masari-wallet-cli 

and you create your wallet


... hope that helps a bit

Cant get this to make. I am stuck at

Code:

[ 34%] Building CXX object src/rpc/CMakeFiles/obj_rpc.dir/core_rpc_server.cpp.o

Any one having same issue. Installing on ubuntu instance.

Bчepa ycтaнoвил нa Mint 18.2 бeз пpoблeм

[daninimmer]

You need last versión >= v9.7 and add "-allpools 1"

NsGpuCNMiner -a cryptonight -allpools 1 -o stratum+tcp://masari.superpools.net:5555 -u YOUR_WALLET_ADDRESS.5000 -p x

I use this program with AMD and it works without problems

v10.2 is the latest version.
Those parameters you are using are not valid anymore.
It is xpool, xval and so on.
No, it doesn't work adding anything behind the address.

Still the pool shouldn't ramp the share difficulty up to 10000 when port 3333 or 5555 are used, when Claymore doesn't submit a fixed diff.

True, it is September 30 and I had not seen it.
I try it

[daninimmer]

I am preparing a pool update (https://masari.superpools.net/) with:
* See the pending mature for account
* Different variable difficulty setting per port
* Maximum hasrate limit in the pool? This is not whether or not you like the community and that's why I ask. And what value would you put? 10kH / s, 50kH / s?

If you have any more ideas, please comment.

I will create a pool of tests, I will ask one of you to change the pool to test these changes (there will be multiple micro-cuts) in exchange for 0% fee and some compensation.

Soon more news, thank you all

[nanona]

I am preparing a pool update (https://masari.superpools.net/) with:
* See the pending mature for account
* Different variable difficulty setting per port
* Maximum hasrate limit in the pool? This is not whether or not you like the community and that's why I ask. And what value would you put? 10kH / s, 50kH / s?

If you have any more ideas, please comment.

I will create a pool of tests, I will ask one of you to change the pool to test these changes (there will be multiple micro-cuts) in exchange for 0% fee and some compensation.

Soon more news, thank you all

Your pool increased the share difficulty even faster than masaripool. 2000H, 12000H, 72000H, then 200007H and remained there.
This is like a DOS attack on the connected miners because they can hardly submit any share before the next block starts and their work stales.
Six hours ago during the recent hashrate surge I guess the attacker got all shares for all blocks while all other miners running at 200007H produced only heat for their owners 

Your auto-diff must adjust the share difficulty in such a way that shares get submitted more or less every X seconds. For a 120s blocktime I'd recommend X to be around 12s.
When blocks get faster, it should get lowered. Other pools with other coins typically have one port for auto-diff and other ports for fixed diff.
Flypool allows you to set a fixed diff in the password field.

I think blocking users based on their hashrate is censorship and not acceptable at all. Some pools however block botnets.
When there are many workers mining with the same payment address but from many different IPs, it can only be a botnet or the dev fee for Claymore.
Claymore or any other dev will however contact the pools so their address gets whitelisted.

[AzCrTe]

I am preparing a pool update (https://masari.superpools.net/) with:
* See the pending mature for account

I see it up - nice

Now if there was just a gui wallet.....  

[Littledragons]

Im in, this looks promising.
Monero isnt the best at the whole anonymity aspects, hopefully there can be some innovation here!

[thaer]

Mac binaries for v0.1.2.0 are now available! https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-mac-x64-v0.1.2.0.tar.gz

For reference, list of supported binaries:

Linux - https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-linux-x64-v0.1.2.0.tar.gz
sha256sum 2d01c137bf1c0f8abc15bd898d7a4e755e2c77ea53d3a90bc206b5aa842c0044

Windows: https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-win-x64-v0.1.2.0.zip
sha256sum: 79b2efaa996ab58479c08d1b5e7839026100bdf39a4a85c7a7252883e388b0cb

MacOS: https://github.com/masari-project/masari/releases/download/v0.1.2.0/masari-mac-x64-v0.1.2.0.tar.gz
sha256sum: 0941b70a98f2ae614a5b7401405e8d6baacfb2e4a4811514aa0da3db11e6380a