Fixing bad plot size errors in Linux



  • DOESN'T WORK ON OPTIMIZED PLOTS

    So you have a plot file that was generated but gives an error in the miner due to an incorrect size?

    Fix it the linux CLI!

    Step 1: Locate and verify your plotfile is non 0 in size
    Step 2: Run the "optimize" program from the DCCT tools or Mjminer tools on the file

    It'll say file is expected to be X but "actually" is X

    Good!
    "Expected" is how big the file was supposed to be
    "Actually" is how much we have to work with, if it's close to the expected number you'll get most of the nonces.

    Step 3 Get the expected size BELOW the actual size by adjusting the NAME of the file.
    Keep renaming the file changing the nonce number (DON'T EDIT THE STAGGER SIZE) to match if it's already optimized until the expected number is just BELOW the actual number. Divide the difference by 256000 to estimate how many nonces you need to deduct. Normally in my experience it's only a few dozen on a plot file of millions!

    Then using the computer calculator or a pocket calculator, find the expected minus the actual. You now have $SizeDifference

    Step 4
    Use the following command with the $SizeDifference being the number you calculated against the plotfile, this will trim the excess bad nonces.
    truncate --size=-$SizeDifference $plotfile

    Voila! If you followed this guide, your file will be shrunk just enough to make all of the nonces valid and the filesize will match the filename!

    Step 5 OPTIONAL
    Run the "optimize" program against it again, it'll optimize it if it's not already optimized, or report that it's already optimized.

    Happy mining!

    Be advised that truncating in this manner will reduce the nonces available to mine on, and it should only be considered a temporary measure until you have time to go back and replot it properly.



  • After following this Now I get
    "Nonces not a multiple of stagger. Quitting"
    how can I fix this?

    This is what I've done

    user@vmi:~/mdcct$ ./optimize ~/burstplot/XXX_0_3814010_44489 
    File not complete. Expected size: 999819837440, actual size: 1000944013312
    user@vmi:~/mdcct$ truncate --size=-1124175872 ~/burstplot/XXX_0_3814010_44489 
    user@vmi:~/mdcct$ ./optimize ~/burstplot/XXX_0_3814010_44489 
    Nonces not a multiple of stagger. Quitting.
    
    

  • Mod

    @captinkid
    Not sure about truncate for optimized plots...
    Plotter splitting nonce for different places of file



  • @bugduino

    Your stagger doesn't match your nonce number, that's the first problem. The nonce number needs to be a round number multiple of the stagger size

    It's currently, not a round number, so you'll get the stagger multiple error

    Can you set the stagger number back to the original and change the nonce number only to reduce the expected size?

    Rename the file to change the nonce to be an even multiple of the stagger and try again, don't change the stagger number, change the nonce number only. You should be able to rename it back to your original stagger size and start again without changing the stagger.



  • @Blago All of mine are optimized and it's not getting read errors. Although perhaps if the truncated scoops or even nonces come up I might not get any deadline from them. Replotting the truncated files may be the best however in the long run, even if this works as a bandage.

    Normally I only have to trim something around .0001% even if it's multiplied along all of the scoops seeing fewer nonces, it's still less than 1% lost for a plot that failed after several days of plotting.

    I wonder if adding bytes to the file instead with a truncate + command might be the better solution? Assuming the hard drive has sufficient space that should repair the file without the need to remove any nonces. I'll do that on my next error file and test it out.



  • I finally managed to fix the error by changing the expected size and set it to be a multiple of the stagger size.
    I can now start the miner (creepMiner) but I get the following error

    00:49:25: using config file mining.conf
    00:49:25: Total plots size: 923.23 GB
    00:49:25: Submission Max Retry : 3
    00:49:25: Buffer Size : 128 MB
    00:49:25: Pool Host : http://pool.burstcoin.eu:80 (216.165.179.115)
    00:49:25: Mininginfo URL : http://pool.burstcoin.eu:80 (216.165.179.115)
    00:49:25: Wallet URL : https://wallet.burst-team.us:8128 (216.165.179.44)
    00:49:25: Server URL : http://192.168.0.100:8080 (0.0.0.0)
    00:49:25: Aborting program due to exceptional state: locale::facet::_S_create_c_locale name not valid
    
    

  • Mod



  • Ok I just found the solution here https://stackoverflow.com/questions/19100708/mongodb-mongorestore-failure-localefacet-s-create-c-locale-name-not-valid it turns out that I need to do the following

    export LC_ALL="en_US.UTF-8"
    

Log in to reply
 

Looks like your connection to Burst - Efficient HDD Mining was lost, please wait while we try to reconnect.