I'll be restarting the pool shortly for further upgrades and bugfixes.
There were more improvements to remote block submission and it should now reset the share counter properly on a block find. Restarts complete, mine on! I'm bored and wish to make one more configuration change which means one more restart which will happen shortly. All complete. Block on. Time for yet another restart for a minor tweak shortly. All complete. Keep on blocking. |
|
|
|
This has happened loads of times why do people have to keep posting about it every time this backup happens
Because it gives sigspammers something to post about. Oh wait... |
|
|
|
Yes but look back before 2013 when everyone was their own pool, when people used to mine with CPU or GPU.
You're mistaken here. I've been writing mining software and mining since June 2011 and virtually everyone was mining on a pool back then. If you want to use that argument you need to go back to 2010 or earlier... basically before bitcoin became popular at all. |
|
|
|
Is there a need for a high performance well connected public p2pool node? I could set one up behind a ckpool sanitiser to optimise performance and propagation speed. Last time I did this however, hardly any miners actually used the service.
Where geographically would this be located? I would consider powering down my node if I had a good alternative. Mid USA. Hmmm, I am in Hawaii, so that might not be good, but I would still try it out. Alas it hardly seems worth if you're far away and going to be the only miner on it. I'll leave the offer open in case many more people raise their hand but at this stage won't be going ahead with setting it up. |
|
|
|
Just enlighten me please, isn't confirming the transactions called mining? also if you find a block you'll get the fees and if you run a node with desktop computer you are technically mining but will never find a block and never receive any thing whatsoever.
It's the pools that should run as many full nodes as they can but are they?
One thing we should let people know, you can mine with 5m rigs and link them just to 1 node or you can have 5m rigs and link each one to a node to have 5m nodes, but if you want to find a block faster than others it's better if you do everything with that one node.
But the network problem is way over that, and it's about the limitations in the source code itself, so if you run 10m full nodes you'll be limited to find a block every 10 minutes and in each block only a limited number of transactions can fit.
Confirming the transactions is indeed done by mining (and mining only) but running a node on your computer - or any computer - is not mining. That stopped being possible 5 years ago... Mining bitcoin is done with dedicated ASIC mining hardware only and most hardware is directing its power towards pools that construct work for them that then submit blocks. The pools only need one node to submit the block, running more nodes does nothing for them (except perhaps create more places to distribute the block from.) You misunderstand a lot of what's going on. |
|
|
|
Also, does anyone have any suggestions on how I can reduce my total bandwidth / traffic use if I have multiple (20+) S9's running on the same network? I will be using a cellular data plan, so it is important to minimize my data usage. As wmabern suggested, a proxy will dramatically decrease your bandwidth. You can use ckpool (linux only) in ckproxy mode to put as many miners as you like to connect to one worker on your pool. The ckpool link is in my signature. |
|
|
|
Is it possible it's the miners themselves doing this to put pressure on the core team to make the necessary improvements?
After all, the miners will reclaim a portion of their own transaction fees – with enough collusion between large miners they could do this almost for free
Unless it's the miners that are injecting the transactions in the first place, no, the miners aren't doing anything to perpetuate this issue. All blocks are about as full as they possibly can be and all signs are that they are filled according to transaction fee/size ratio which is the default bitcoind selection process. |
|
|
|
Bitcoin is heading towards Centralization that way. Kinda looks like miners now control the network. We now have to "bribe" Miners with high attractive Transactions fees to get our transactions confirmed? full bullshit. The idea of increasing fees to get confirmed is a crab.
If it continues this way, Miners will crumble the Bitcoin network by not confirming Transactions over people's unwillingness to use higher fees.
I hate to break it to you, but miners have always controlled the network... That's bitcoin by design. |
|
|
|
Is there a need for a high performance well connected public p2pool node? I could set one up behind a ckpool sanitiser to optimise performance and propagation speed. Last time I did this however, hardly any miners actually used the service.
Where geographically would this be located? I would consider powering down my node if I had a good alternative. Mid USA. |
|
|
|
|
Ancient mining software for hardware that is irrelevant to modern mining cannot mine directly to the bitcoind. Bitcoind removed most irrelevant mining interfaces years ago and ancient mining software that mines on useless mining hardware (by today's standards) does not speak modern mining communication languages. There are ways to make the latin speaking ancient mining software speak to the esperanto speaking bitcoind but it's a futile and pointless exercise to put all the necessary interfaces between them since it only consumes a lot of electricity and does not remotely resemble what modern bitcoin mining does today and cannot realistically ever mine anything of any value in one's lifetime.
|
|
|
|
The earlier of these spam attacks has been discussed to death here: https://bitcointalk.org/index.php?topic=1762690.0The graph of rise in transaction sets at the time did not follow the normal stochastic nature of random transaction events. There was a perfect curve with a rise in the number of transactions starting from 2017-01-24 at 06:00 until it peaked at 17:00. After that there have been random injections into the network to keep the value high. The perfect curve is consistent with a single entity constructing a careful crafted set of injections into the network. See: http://statoshi.info/dashboard/db/transactionsChange the graph to see the last seven days and you will see why this is not just some random rise in transactions. It's once again likely a political stunt to bring attention to blocksize and/or segwit. |
|
|
|
some pools will let you set a value for your miner in there interface.
on cgminer, you can tag a suggested diff to try to use during startup negotiating.
--suggest-diff 32
This will work on any ckpool based pool such as solo.ckpool.org . I'm not sure how much vh's cgminer fork varies from the master cgminer but it should work, worst case scenario it might take up to a minute to negotiate the diff, and solo does a redirect after a successful share so it may happen twice. There is no lower limit on hashrate to work on solo and there are many very low hashrate stick miners there. |
|
|
|
|
Is there a need for a high performance well connected public p2pool node? I could set one up behind a ckpool sanitiser to optimise performance and propagation speed. Last time I did this however, hardly any miners actually used the service.
|
|
|
|
I'm getting the following errors: checking for gcc... gcc
checking whether the C compiler works... *** Error in `gcc': double free or corruption (top): 0x01a43588 ***
no
configure: error: in `/home/pi/git/vthoang/cgminer':
configure: error: C compiler cannot create executables
See `config.log' for more details
My Raspbian is fully up to date and I'm running kernel v4.4 An error this bad during just the configuration process usually implies a corrupt SD card/image in your RPi. |
|
|
|
so the current difficulty is the numbers your *bestshare* must hit to solve a block ?
Yes, equal or better. |
|
|
|
i will get an s7 soon....
bestshare im havin now is 19572430 so where or how to know if i was close to solve a block?
Current difficulty is 392,963,262,344 so you're billions of light years away still. Check the current difficulty here: https://bitcoinwisdom.com/bitcoin/difficulty |
|
|
|
Total noob question. How does one update Cgminer in an Antminer S5?
IMs appreciated, Im still learning so dont bash me too bad.
The cgminer in an antminer S5 is usually precompiled in firmware provided by bitmain and is based on an old forked version of cgminer. It is not supported by the official cgminer version and would be far from trivial to be supported and build a binary for the embedded controller of the S5. In short you cannot build the latest cgminer for S5. I built a temporary binary for the S5 a while ago which you can copy to the S5 here: http://ck.kolivas.org/apps/cgminer/antminer/s5/But it does not stay on the device across reboots and is based on 4.9.0 code. If you are not having trouble with your S5 there is little to gain by updating the software anyway, accepting the shortcomings of the forked older version from bitmain. Except in the case of mining on p2pool (which I recommend against with any antminer because of their code problems), most of the issues with their version of cgminer are not showstoppers. |
|
|
|
|
That was not a meaningful argument, lacking any substance in your response to my factually based critique. Time for me to leave this thread since answering nonsense responses is a waste of my time.
|
|
|
|
This is directly related to the FLAW in BTC design, the fact it only does a difficulty adjustment very 2 weeks.
Most Alts switched to a difficulty adjustment after every block to avoid this problem.
Block finding is a stochastic process and sampling based on one block is a lousy way to estimate the current network hashrate. What happens when two blocks are found through pure luck only milliseconds apart? What about when a block is 90 minutes later even when the network hashrate hasn't changed, simply because block finding isn't a deterministic process? In short, changing diff after each block is a terrible idea. |
|
|
|
|
Show Posts
