You might be able to help me with a noob question. Whenever I enable testnet=1 in my .conf file, I always end up with issues.
It always shows that it's out of sync, 2 weeks behind, and "no block source available".
What am I doing wrong?
I don't think you are doing anything wrong - it has another block chain on the test net and that needs to sync separately. I ran into some issues syncing with it with antivirus software (only for the first part). It might help to disable your firewall temporarily until you have part of the block chain and multiple block sources. Then again I'm a complete noob and might be goiing about this completely wrong - take what I say with a grain of salt. Ok, I'm glad it's not just me. I'll just hold tight then. |
|
|
|
Thanks for the explanation - I was just curious about settings since I'm using the default sieve size and sieve percentage and don't know if changing them helps at all right now - I saw people discussing different settings making a difference on the testnet and was wondering if there were any better settings that I could use right now.
Your best bet is to run some test w/your box on testnet to find your settings. For v7 my best are 2M/8%, 2M/14% did well too. However, no blocks w/v4-v7 for me on i7 860 @ 3.22. I do spend a lot of time on testnet though. I figured since I wasn't finding blocks I could at least find some good settings for any other i7 quad miners out there and I'll wait for the GPU miner. You might be able to help me with a noob question. Whenever I enable testnet=1 in my .conf file, I always end up with issues. It always shows that it's out of sync, 2 weeks behind, and "no block source available". What am I doing wrong? |
|
|
|
Still getting new blocks mined on my i7-970 Gulftown at 3.2GHz (2 today, 2 yesterday). Maturing seem to take at least 48 hours, so I'm guessing a lot of machines have dropped out of the network.
What settings are you using and how many primes per second are you getting? I'm mining on my 3930k and get anything from 4k to 5700 primes per sec and have only gotten one block in three days of mining. It really is hard to tell if/when you'll get a block. I don't mean that as in "I don't know how to calculate it", I mean it in the sense that "I'm not sure if it can be calculated". Prime numbers and prime chains are quite fickle and inconsistent things. With hash-based cryptocurrency, there's a certain uniformity and predictability with difficulty. With primes and prime chains, there's some predictability, but not enough to get you a simple formula. See the "Difficulty Adjustability of Proof-Of-Work" of the primecoin paper. With hash based coins, the question is simply "if I throw dice with X sides, what's the odds I'll get a number lower than [difficulty]". With primecoin, the question is "what's the odds that I'll find a chain of primes [difficulty] numbers long, with the next number not being prime and having a remainder greater than [difficulty]". So yeah. I have no idea (and neither does anybody else). P.S.: Just because you're finding X primes/sec doesn't mean you'll find them with the correct properties. |
|
|
|
24 July, 2013 - Major updates! See main post for additional info. If you'd like to remotely backup your wallet.dat modify the following segment of code... cat << "SCRIPT" > ~/notify-wallet
#!/bin/bash
# YOUR SCRIPT HERE so that it does something like... cat << "SCRIPT" > ~/notify-wallet
#!/bin/bash
cp ~/.primecoin/wallet.dat ~/${HOSTNAME}.bak
curl -T ~/${HOSTNAME}.bak ftp://your_remote_location --user your_ftpusername:your_ftppassword
SCRIPT The script triggers based on primecoind's -walletnotify="~/notify-wallet" flag. It should trigger whenever the wallet file changes, which includes when a new block is minted.
|
|
|
|
|
Does anybody have a recommendation on how to keep my keypools in sync? I just noticed that when a block is mined it makes a new address that doesn't show up in getaddressesbyaccount "".
Although, I can still do a dumpprivkey on it.
|
|
|
|
Complete scrub to this, should I be running the primecoin-qt or primecoind?
If you're running it and want visual interaction, you want to run primecoin-qt. If you want to control it via command line or are using a headless system, primecoind. You should get the same performance with each. |
|
|
|
The script has been updated.Here's the relevant section: echo "Downloading and building primecoin..."
cat << "SCRIPT" > ~/build-primecoin
#!/bin/bash
[[ -f ~/stop-primecoind ]] && ./stop-primecoind
rm ~/primecoin*.tar.bz2
rm -rf ~/primecoin*
wget http://sourceforge.net/projects/primecoin-hp/files/latest/download -O ~/primecoin.tar.bz2
tar xjvf ~/primecoin.tar.bz2
cd ~/primecoin*/src
sed -i -e 's/$(OPENSSL_INCLUDE_PATH))/$(OPENSSL_INCLUDE_PATH) \/usr\/local\/include)/' makefile.unix
sed -i -e 's/$(OPENSSL_LIB_PATH))/$(OPENSSL_LIB_PATH) \/usr\/local\/lib)/' makefile.unix
sed -i -e 's/$(LDHARDENING) $(LDFLAGS)/$(LDHARDENING) -Wl,-rpath,\/usr\/local\/lib $(LDFLAGS)/' makefile.unix
make -f makefile.unix USE_UPNP=-
sudo cp primecoind /usr/local/bin/
cd
SCRIPT
chmod +x ~/build-primecoin
~/build-primecoin It will automatically download and compile the newest version of primecoin-hp from SourceForge (as to not overload git).
|
|
|
|
anything over 2000000 and mine crashes. x64 bit launches and stops closes window.
You sure you're on the newest version (hp7)? I just tested 10mil on x64 and it worked fine. What kind of CPU are you using? I have an i7 3820, and I have been reporting this problem since at least hp4. I used to be able to go up to 4M just fine, and anything over that crashed - but with hp7, anything much over 2M crashes. Core i5-3210, 2.50GHz I've had mine running at ~4M for the last few hours without a hiccup. I ran 10M on accident (had an extra zero) for about half an hour. |
|
|
|
anything over 2000000 and mine crashes. x64 bit launches and stops closes window.
You sure you're on the newest version (hp7)? I just tested 10mil on x64 and it worked fine. |
|
|
|
Good catch. I've modified the script so that it only clones once and uses git pull to catch up.
The initial pull will be ~15MB, but after that just a few KB. Much lighter and quicker than ~15MB each time.
I would still prefer you didn't clone it at all. SourceForge has tons of mirrors which can handle load much more effectively. My concern with that is that the path of the newest release can change without notice. I don't want to put your bitbucket account in jeopardy though. I'll start working on a way to dynamically pull from SourceForge, regardless of how hackish it turns out. Once it's finished and tested, I'll update the setup script. |
|
|
|
Nice initiative. However, I would prefer that people wouldn't post instructions that automatically clone the git repository.
Good catch. I've modified the script so that it only clones once and uses git pull to catch up. The initial pull will be ~15MB, but after that just a few KB. Much lighter and quicker than ~15MB each time. |
|
|
|
What is approximate amount of primecoins you can mine in a day by using 512MB/1core ?
0 or less. In a week or month you may get some. However if you ran it for just 1 day i would predict zero coins. Per day, you're probably technically correct. Then again, your statement also applies to just about any non-cluster setup.
Then wat is the point in paying 5$ for this ?  With the way DigitalOcean works, you pay by the hour regardless of what configuration you've selected. What that means is that you can run one 512mb/1core for a month or you can run four of them for a week. You can have up to five droplets per account, but you can open a ticket to raise the cap. Using the setup script, I managed to get a handful of droplets running in under 30 minutes and can update them with a single command. Worst case scenario, I'm out $5. |
|
|
|
After reading mikaelh's Primecoin setup guide I decided to automate as much of the process as possible to make it noob friendly. The result is this! Setting up a Primecoin Miner on VPS for Absolute NoobsDonations welcome! BTC: 1E2egHUcLDAmcxcqZqpL18TPLx9Xj1akcV Ψ: AWHJbwoM67Ez12SHH4pH5DnJKPoMSdvLz2 Before You Begin- Download PuTTY (putty.exe) so you can connect to your VPS when the time comes.
- Sign up for a DigitalOcean or Linode account.
- During the creation process, be sure to create a node running Ubuntu 13.04 x64 (not x86)
- If you signed up for DigitalOcean, check your email for your root password. If you signed up for Linode, your password will be what you specified.
- Run PuTTY and paste the node's IP address into the "Host Name" slot, then hit "Open" at the bottom. (figure 1)
- PuTTY will ask you to confirm the SSH key, press Yes (it will only ask once) (figure 2)
- Type or paste your password into PuTTY and press enter. To paste, simply right click.
- Note: You will not see the password characters! This is normal.
  Getting Down to Business- If you chose DigitalOcean, it's time to change the root password! Type passwd, press enter, then enter a new password. Just like before, the password characters will not show up. (figure 3)
- Copy this automated setup script to notepad. (highlighted version)
- At the top of the script, there are fields starting with "PRIMECOIN_". Fill these out with your desired info. Be sure to keep the quotes!
- PRIMECOIN_USERNAME is your primecoind username. It's required for primecoind to work properly!
- PRIMECOIN_SIEVESIZE, PRIMECOIN_SIEVEPCT, PRIMECOIN_SIEVEEXTENSIONS are parameters used by mikaelh's high-performance primecoin client (see link for info)
- Copy the entire modified script from notepad and paste it into your PuTTY window (right click), then hit enter.
- The setup is now running. It will take approximately 10 minutes to complete.
- Once the script completes, you will see a brief help message and list of commands. (figure 4)
Congratulations! You're now mining for primecoins! It may take a few minutes before you actually begin mining. This is normal. Primecoin needs to download the blockchain before it can begin doing work. After it's caught up, you should be able to run ./peek and see that it's generating primes.If you'd like to maximize your mining potential, please see my mini-guide on How to Clone DigitalOcean Droplets.
What Now?- If you haven't already, download the latest version of the Primecoin client.
- To "merge" your primecoin accounts into one, you'll need to gather the private keys from your VPS nodes
- Log into your droplet like you did at the beginning of the setup. This time your password will be what you set it to.
- Run ./peek to verify that primecoind is running. If it isn't, run ./start-primecoind and wait a few moments.
- Run ./myinfo to dump your addresses and private keys values. (figure 5)
- Copy the private key value (the random text after "PrivKey:") by highlighting it with your cursor.
- On your main computer, open Primecoin and import the private key.
- From the main Primecoin screen, click Help -> Debug Window
- In the new popup, select Console
- Type importprivkey YOUR_PRIVKEY SOME_LABEL and wait a few seconds. YOUR_PRIVKEY is what you copied from PuTTY and SOME_LABEL is what you want the address to appear as in your address book. (figure 6)
- Repeat this for as many private keys you wish to import. Once finished, close and re-run the Primecoin client. You should see the new addresses under "Receive".
- Note: if an instance of primecoind mines a block, it may not be under the address you exported. Wait for the block to be confirmed, then dump your addresses again. Alternately, modify the notify-wallet script to FTP a backup of your wallet.dat file.
- Occasionally you may wish to update your primecoin miner when new versions or patches are released. To do this...
- Log into your DigitalOcean droplet
- Run ./stop-primecoind to stop the primecoind process
- Run ./build-primecoin to update and rebuild primecoind
- Run ./start-primecoind to start your primecoind process again
- Wait a few minutes and run ./peek to make sure it's running correctly
- If you'd like to remotely backup your wallet.dat file whenever you mint a new block, check this short guide.
Changelog- 27 August, 2013
- Cron reboot script now works properly. (Damn you cron and your newlines!)
- ~/peek no longer shows orphaned blocks. Balance will only include transactions with one or more confirmations.
- ~/myinfo now shows all addresses, not just those in the last ~100 transactions. The pattern matching has been made more generic, so it works with just about any coin on main/testnet. It also only does a lookup on addresses that generated a block or received currency.
- PRIMECOIN_ROUNDSIEVEPERCENTAGE has been removed, replaced with PRIMECOIN_SIEVEEXTENSIONS
- Now includes a ~/restart-primecoind script
- 20 August, 2013
- Tweaked apt-get package names to grab most compatible package for the given install setup
- Package list now includes fail2ban by default for some semblance of security
- Package list now includes nano as a text editor. Why some installs don't come with this is beyond me.
- Swapfile now persists after reboot.
- Auto reboot on kernel panic
- 15 August, 2013 - Now Amazon EC2 compliant and primecoin runs at startup.
- 12 August, 2013
- Updated default parameters to match the newer hp9 client (1mil / 10% / 70%)
- Runs apt-get update before installing things
- ~/start-primecoind should no longer accidentally kill itself
- Initial package install includes bc so ~/peek correctly shows all balances
- Fixed haveged not running after setup. There's been some debate on the usefulness haveged, but until it's definitively proven to be harmful, it will remain.
- Post-install help screen now uses absolute paths instead of relative
- Minor formatting chagnes
- 01 August, 2013
- ./peek now includes the chains/day metric (still in development). It also shows "Immature Balance" instead of "Unconfirmed" to more accurately match the Primecoin-Qt client.
- ./myinfo now correctly dumps all keys, not just confirmed
- ./notify-wallet now has example headings
- Install includes HAVEGED for increased entropy pool (thank you blackboy)
- 29 July, 2013 - Lots of new stuff!
- Added option: PRIMECOIN_ROUNDSIEVEPCT
- PRIMECOIN_USERNAME is now required (I'm an idiot), but PRIMECOIN_PASSWORD is auto-generated.
- Installs newest version of GMP before starting
- notify-wallet now includes example on how to FTP a wallet backup
- peek now dumps averages of "primemeter" stats. This method provides much more accurate results than a single getmininginfo. See an example here.
- start-primecoind shouldn't kill itself anymore
- 24 July, 2013
- ./peek now shows confirmed/unconfirmed balance and pending "immature" transactions
- ./config now reboots primecoind after config changes have been made
- ./stop-primecoind now pauses after killing primecoind process to prevent running ./start-primecoind too early. It will also force-kill primecoind if it's misbehaving.
- When building primecoin, intermediate .tar.bz2 file is now removed
- RPC username/password config removed. They're not exactly "noob" features.
- New Feature: run script when blockchain is updated or when wallet is updated. By default it make a backup of wallet.dat whenever the wallet updates (usually from mining a block)
- New Feature: whenever primecoind crashes, the time of crash is logged in ~/crash.log
- 23 July, 2013
- When updating primecoin, pull from SourceForge instead of bitbucket.
If you have any questions or comments, feel free to speak up. |
|
|
|
I just went from v5 to v7 and now my prime rate is halved. Anyone else having this? Is it because of the optomizations made (ie. more efficient/accurate)?
Also, I have set: -sievesize=4000000
but when I view stats with getmininginfo it always says its on 1M not 4M?
Doublecheck that you don't have sievesize set in your primecoin.conf file? |
|
|
|
I've been consistently finding blocks once per day with my Core 2 Quad 9400 OC'ed to 3.5GHz on the second windows client. My pps is usually around 2500, so I'm either very lucky or something is wrong on your end.
I can attest to you being lucky. I've been running a constant ~2700 PPS for almost 48 hours and haven't found squat. Running Sieve% 10, SieveSize 1000000 |
|
|
|
|
I think I may have found the culprit: my USB hub's AC adapter.
Every test I've run has had both block erupters connected to the hub even if I was just using one of them for testing.
Unplugging either of the erupters makes the other work correctly.
The AC adapter is rated at 5v 1000ma, but given how bulk produced they are, it wouldn't surprise me if it's actually much less.
It would explain why it works on my desktop but not my RasPi: the desktop can actually deliver the missing 500ma via the USB port, the RasPi cannot.
SOLVED: Yep, it's the AC adapter. Just hooked it up to my trusty voltmeter. The AC adapter is rated at 5v 1000ma but puts out 5v 350ma.
|
|
|
|
I feel like I'm taking crazy pills. I did a fresh install of Raspbian and ran updates. Followed the adafruit guide from here. Ran cgminer...  ONE. One share in ~5 minutes. When I restarted and tried again, no shares. Plugged it back into my Windows 7 desktop, got "U" of ~4/m. I'm slowly, very slowly going insane. |
|
|
|
The Eupter LEDs blink 3 times, then stay solid when plugged in. I've hacked together the following to make sure I pick up every ttyUSB#: usb_list=$(find /dev -name 'ttyUSB*' | awk '{ printf "-S " $1 " " }')
cgminer --config ~/cgminer/cgminer.con $usb_listI have also manually listed out the -S /dev/ttyUSB0, etc. I feel like I'm reaching the point where the next sane option is to backup my RasPi, fresh install and see if that works. |
|
|
|
|
Show Posts
