How does P2Pool handle log file management as far as size/time goes?
In a short time my log has grown to 83.9MB, is their a point where it is truncated or will it just keep growing?
Hoping someone can give me some insight on this If I'm reading the code correctly, it should top out at 100MB.
|
|
|
mdude77 congrats on finding that block
|
|
|
I just checked out HUC. Interesting concept - build mining into a game. I'll look into setting up my node to merge mine it. Will have to compile the daemon from source (no Qt version for my Mac, and I really don't feel like trying to screw with the code to make one). I've had nothing but problems trying compile Qt clients - consistently get weird errors. It's probably something in my environment, but I haven't had the time to figure it out.
|
|
|
Well for the patches I've written, I have to fork the repo into my own name on github, make a branch, push the changes to the branch, and then submit the pull request to the maintainer. I'm pretty sure what I provided will work as expected though. OK, thanks. Just tried it - how does this look to you?: Seems OK to me - or have I replaced the whole repo? You did it correctly. You've replaced the listed files with the ones from the auto-worker-diff tag.
|
|
|
Yes, I agree!!
Should I add to my username +1024 for each of my miners? I am using several AntMinerS1 all with the same wallet address. What do you think?
I'd like to see some more informed members reply to this, however it is my understanding that allowing P2Pool to dynamically set your difficulty is a more efficient choice. The p2pool code looks at the node's total hash rate and assigns difficulty based upon that. There are forks of the code that address the issue of trying to assign difficulties by address/miner. See here: https://bitcointalk.org/index.php?topic=269042.280.
|
|
|
Trying to keep it as simple as possible...
VPS is what pretty much every service out there offers you. You get a virtualized machine with superuser access allowing you to install whatever you want on it. Sign up at Digital Ocean, AWS, etc, you get a VPS.
A dedicated server is bare metal. You get the entire processing power of that box to yourself. Unless you own your own data center, or rent space in one, chances are exceptionally slim you'll get an entire server to yourself, or if you do you're going to pay a fortune for the privilege.
Cloud mining means that somebody is hosting miners in a facility and is selling you hashing power on those machines. If I were to relate this to VPS and dedicated servers, I would do so as follows:
VPS - services like cex.io, pbmining, etc offer you the ability to buy hashing contracts Dedicated - You buy your own mining hardware and someone hosts it in their data center for you. Effectively, you'd pay rent (hosting fees) for your machine using up their space/electricity.
|
|
|
mdude77, I read about this in the p2pool thread. If I understand it correctly, for every block of NMC, IXC, DVC your p2pool node finds, you will be converting them into BTC and donating to the entire p2pool network.
What a great idea, and a bit of incentive for folks to mine on p2pool. I really like it, and should my own local node ever be lucky enough to find another block I will follow through and do the same as I currently merge mine as well.
|
|
|
Yes, I have a bad feeling, maybe it is still downloading the chain???
I did this:
bitcoind getmininginfo { "blocks" : 285045, "currentblocksize" : 7276, "currentblocktx" : 14, "difficulty" : 2621404453.06461525, "errors" : "", "genproclimit" : -1, "networkhashps" : 17699612012185874, "pooledtx" : 14, "testnet" : false, "generate" : false, "hashespersec" : 0 } is it up to date or not yet?? thank you
It's close, but not there yet. Check blockchain.info, the last block is 299153. M Now it is up and running perfectly, efficiency 110%, DOA 1-2% p2poolthailand.com or p2pool.asia Glad to see you got everything resolved! The problem was that you had not completed downloading the block chain, so your miners were finding and submitting shares that could not be rectified. Learning experience for everyone who wants to setup their own node - absolutely ensure that you've completed synchronizing the block chain before mining on your p2pool node
|
|
|
I am restarting, and installed a fresh copy of the p2pool, let's see what happen
But why yesterday I was expecting 0.06444 BTC and now 0.01, I mean why I did not change anything, at least in the miners.
bitcoin.conf has only
server=1 daemon=1 rpcuser=xxxxxxxx rpcpassword=xxxxxxxxxx
(not xxx but I am just hiding the content)
I figured you weren't really using 'xxxxxxxxxx' Monitor your node for a while. Like I wrote, when I looked at it, you had a 10% DOA rate. Now, you're up to 23%. You have a problem with your network connectivity between your miners and your node, or your miners are configured incorrectly.
|
|
|
9 total (8 orphaned, 1 dead) Efficiency: 0.000%
But I don't understand why yesterday it was working ok.
I am trying to reinstall p2pool from 0
Or I suspect there is something wrong in bitcoind....
anyway the pool is p2ptoolthailand.com:9332
That's precisely your problem. Your miners found 9 shares, all of which were lost. The 8 orphaned mean they got to the p2pool network too late. The dead one means it didn't even make it out to the network at all. You've got a connectivity issue from your p2pool node, and possibly a miner configuration issue. How many inbound/outbound connections do you have? Are you running a fully synched Bitcoin node? Have you setup some limits in your configuration files to limit the number of connections? EDIT: I just looked at your stats page. You made a typo in your link, but I figured you didn't really mean p2ptool Anyway, looking at the stats from the past few minutes since you've fired up your node, you've got a 10% DOA. That's about 9% too high. What is the latency between your miners and the machine running the p2pool node? Between the node and the bitcoind?
|
|
|
no guys, there is something wrong, I put back the standard web-static and I see my efficiency al 0%!!!!
how come??
I have more than 1.5 TH/s
How many shares does your node report it has found (how many orphan, how many dead)? What does your node report for current hash rate?
|
|
|
Oh, sorry, I never explained myself, I'm not mining SHA256, I'm mining Scrypt altcoins! Gotcha. That makes far more sense. Your Uncle is a very nice guy if he's letting you peg all those CPUs to 100% all the time. You should tell him to pick up a couple of those Scrypt blade miners and hook them into a single machine in his data center. You'd get greater hash rate, and he'd not have to worry about you burning out his machines. Well, those computers are used mainly for storage, so the CPU resources are never really used in any profitable way, that is why he granted me the logins. What are those scrypt blade miners, how do they work, and more importantly, how much do they cost? The Scrypt blade miners are ASIC hardware for mining coins like LTC, DOGE, etc. At present Gridseed manufactures them, along with their 5-chip dual miners. The miners are dual blades, with each blade having 40 chips, for a total of approximately 5.2 MH/s from the unit, all for about 140 watts. You can pick them up for about $1500 USD give or take. If you were to pick one up for yourself, you'd have to deal with VAT and customs fees (depending on where in Europe you are). I'm not sure what kind of tax/import fees there are in Brazil. That's pretty expensive, considering that I can get about 7 Mh/s for free. LOL... you might be getting it for free, your Uncle certainly isn't. He's paying for the electricity costs of all of those 1500 instances, which by the way will certainly increase because they are now drawing far more power than they were before when they were sitting idle doing nothing but storage as you said. He's also dealing with the risk associated with you pegging all of those CPUs to 100%. To him, it is likely very worth the investment of a couple of those blade miners. He'd make the investment back on power savings alone.
|
|
|
Oh, sorry, I never explained myself, I'm not mining SHA256, I'm mining Scrypt altcoins! Gotcha. That makes far more sense. Your Uncle is a very nice guy if he's letting you peg all those CPUs to 100% all the time. You should tell him to pick up a couple of those Scrypt blade miners and hook them into a single machine in his data center. You'd get greater hash rate, and he'd not have to worry about you burning out his machines. Well, those computers are used mainly for storage, so the CPU resources are never really used in any profitable way, that is why he granted me the logins. What are those scrypt blade miners, how do they work, and more importantly, how much do they cost? The Scrypt blade miners are ASIC hardware for mining coins like LTC, DOGE, etc. At present Gridseed manufactures them, along with their 5-chip dual miners. The miners are dual blades, with each blade having 40 chips, for a total of approximately 5.2 MH/s from the unit, all for about 140 watts. You can pick them up for about $1500 USD give or take. If you were to pick one up for yourself, you'd have to deal with VAT and customs fees (depending on where in Europe you are). I'm not sure what kind of tax/import fees there are in Brazil.
|
|
|
Oh, sorry, I never explained myself, I'm not mining SHA256, I'm mining Scrypt altcoins! Gotcha. That makes far more sense. Your Uncle is a very nice guy if he's letting you peg all those CPUs to 100% all the time. You should tell him to pick up a couple of those Scrypt blade miners and hook them into a single machine in his data center. You'd get greater hash rate, and he'd not have to worry about you burning out his machines.
|
|
|
You do remember it 297777 was mine.
|
|
|
You're using Putty. All you have to do is start your remote miners with the nohup command added to them. Example:
nohup ./bfgminer -o stratum+tcp://stratum.somepool.com:3333 -u workerName -p workerPassword &
If you were to use screen as was suggested earlier, you'd have to install it on the remote machine. Not sure what Linux flavor those remote machines are running, but if they're on a Debian-derived distro, you'd just do this:
sudo apt-get install screen
Then, you'd start your miners like this:
sudo screen -dmS MyMiners ./bfgminer -o stratum+tcp://stratum.somepool.com:3333 -u workerName -p workerPassword
This starts up a session in the background. To access it, you'd just do this:
sudo screen -dR MyMiners
Then, once you're in the screen, to get out of it, all you need to do is hold down the control key and hit a then d.
And with this method, can I just set it and forget it? Setup, install, run and never login to it again? Pretty much. Also, something to consider, as I mentioned earlier, my uncle has provided me access to its server infrastructure in Brazil, but he has given me this access under the form of 1500 DIFFERENT VPS Logins, which means that the 7 Mh/s that I mentioned earlier is only possible by being connected to 1500 INDIVIDUAL SSH sessions.
I had to MANUALLY install cpuminer in 1500 different sessions, which in of itself took me nearly 4 days.
Please tell me I'm not reading this correctly or that you've misrepresented your hash rate, or you're mining something other than BTC. 1500 sessions. 7MH/s??? I get far greater rate than that out of my MacBookPro running CPU miner.
|
|
|
You're using Putty. All you have to do is start your remote miners with the nohup command added to them. Example:
nohup ./bfgminer -o stratum+tcp://stratum.somepool.com:3333 -u workerName -p workerPassword &
If you were to use screen as was suggested earlier, you'd have to install it on the remote machine. Not sure what Linux flavor those remote machines are running, but if they're on a Debian-derived distro, you'd just do this:
sudo apt-get install screen
Then, you'd start your miners like this:
sudo screen -dmS MyMiners ./bfgminer -o stratum+tcp://stratum.somepool.com:3333 -u workerName -p workerPassword
This starts up a session in the background. To access it, you'd just do this:
sudo screen -dR MyMiners
Then, once you're in the screen, to get out of it, all you need to do is hold down the control key and hit a then d.
|
|
|
Here you go:
--- 162.242.245.151 ping statistics --- 20 packets transmitted, 20 packets received, 0.0% packet loss round-trip min/avg/max/stddev = 14.774/18.033/47.896/6.942 ms
I'm in the southern part of New Jersey...
Great start and nice work on the front end by the way.
Thanks for the ping, very low latency for you Their is a lot more I want to do with the front end, hopefully will have some time for it next week. I've also been playing around with the front end code, but haven't put as much time into it as you have. I've done some modifications to the javascript (for both the calculation of node/miner payouts as well as to the charting scripts to add in other information). Recently I've started playing around with the python to get more statistics, total BTC network hash rate, block difficulty, etc. Unfortunately my job gets in the way, and when I am home and have free time, the little woman prefers I spend it with her rather than going on about "that bitcoin thing you do". My friend's pool Local area network, this is the best efficiency. And share with you。 Your friend's node makes up about 10 to 15 percent of the p2pool network by itself. Impressive bit of hardware he's got online.
|
|
|
OK, Node is up and and stable at 120% efficiency currently (thanks to you guys!), there is still much do do on the front-end, but that will have to wait for next week. The node is hosted in a Virginia (US) data center with quad processors, 100% SSD Drives, and a huge pipe. I'd love to see some latencies from various places if anyone cares to give it a ping I'm in Connecticut and average 27ms right now. http://162.242.245.151/Here you go: --- 162.242.245.151 ping statistics --- 20 packets transmitted, 20 packets received, 0.0% packet loss round-trip min/avg/max/stddev = 14.774/18.033/47.896/6.942 ms I'm in the southern part of New Jersey... Great start and nice work on the front end by the way.
|
|
|
|