Mining issue with GTX 1070 mining rig: NEED HELP PLEASE!

[hustleman]

I have a 12 gpu  EVGA 1070 rig that has been working fine for over a year.  Recently i've been using NBMiner 25.5 to dual mine ETH and CBK, utilizing MSIAfterburner for overclocking. (77 powlim, +40 core, +610 memory) no problems. I switch miners to Phoenix 4.7c to mine strictly eth.  I set the command line in PHX4,7c for the overclocking (same thresholds listed above) but after about 3-4 minutes of mining a memory error stopped 2 cards and shut the miner down.  I went back to NBMiner and MSIafterburner to see if it would resume mining with all 12 cards, but with no luck.  It would get to the 12 card and throw (-err 02) and that card would drop off MSIafterburner's clocking controls, fan would spin at 100% and the card would not be mining at all.  I disconnected this card resumed MSIafterburner and NBminer with 11 cards.  After about 10 minutes of mining the same thing happened and a second card dropped off MSIafterburners clocking control, stopped mining and fan at 100%.  I disconnected this card, rebooted and started MSIafterburner and NBminer with only 10 cards.  Now the rig mines good with only 10 cards and I can not achieve the 12 cards I had running for over a year.  

Any ideas whats going on or how to fix this???

P.S. After several attempts at first I DDU uninstalled drivers, updated to newest drivers and newest MSiafterburner, but still had same results.  I also disconnected every gpu except the one losing the clocking controls, and GPU would work perfectly mining by itself, but as soon as I added other cards it would do the same thing of losing MSIafterburners clocking controls, stop mining and fan 100%.  

Rig:
Windows 10 os
SSD 120 gb
RAM 16gb
2 psu (1200 and 850)
virtual memory set at 48000mb start and 86000mb

[ayiphelmy]

I have a 12 gpu  EVGA 1070 rig that has been working fine for over a year.  Recently i've been using NBMiner 25.5 to dual mine ETH and CBK, utilizing MSIAfterburner for overclocking. (77 powlim, +40 core, +610 memory) no problems. I switch miners to Phoenix 4.7c to mine strictly eth.  I set the command line in PHX4,7c for the overclocking (same thresholds listed above) but after about 3-4 minutes of mining a memory error stopped 2 cards and shut the miner down.  I went back to NBMiner and MSIafterburner to see if it would resume mining with all 12 cards, but with no luck.  It would get to the 12 card and throw (-err 02) and that card would drop off MSIafterburner's clocking controls, fan would spin at 100% and the card would not be mining at all.  I disconnected this card resumed MSIafterburner and NBminer with 11 cards.  After about 10 minutes of mining the same thing happened and a second card dropped off MSIafterburners clocking control, stopped mining and fan at 100%.  I disconnected this card, rebooted and started MSIafterburner and NBminer with only 10 cards.  Now the rig mines good with only 10 cards and I can not achieve the 12 cards I had running for over a year.  

Any ideas whats going on or how to fix this???

P.S. After several attempts at first I DDU uninstalled drivers, updated to newest drivers and newest MSiafterburner, but still had same results.  I also disconnected every gpu except the one losing the clocking controls, and GPU would work perfectly mining by itself, but as soon as I added other cards it would do the same thing of losing MSIafterburners clocking controls, stop mining and fan 100%.  

Rig:
Windows 10 os
SSD 120 gb
RAM 16gb
2 psu (1200 and 850)
virtual memory set at 48000mb start and 86000mb

try to change the riser maybe that solve your problem

[frazier34567a]

• After checking risers and power connectors on each riser.   
• Next I would run at stock with no OC or power settings on all 12 cards. with and without MSIafterburner running.
• If still an issue, I would check the BIOS and see if it is up to date.  There was a memory issue back in 2016 with some 1070 cards.   https://forums.evga.com/EVGA-GeForce-GTX-1070-BIOS-Update-v8604500070-m2565056.aspx
• Go back to an earlier version of the miner and MSIafterburner, sometimes newer versions have bugs in them.
• Next I would check the cards of oil seepage.  I had one card that the pads they used to help prevent the overheating issues leaks oil and oil got onto the pci card pins and caused issues.  I cleaned it per EVGA. 
• I would then check power usage of the miners.  I had an issue once where an update for greatly improved performance but it drew twice as much wattage per card. could be that you don't have enough power to run all of them.
• Try cards in different computer see if it causes issues.
• Contact EVGA and ask for their help on a memory issue on one of your 3 cards in SLI mode erroring out.
• RMA the card

[hustleman]

• After checking risers and power connectors on each riser.   
• Next I would run at stock with no OC or power settings on all 12 cards. with and without MSIafterburner running.
• If still an issue, I would check the BIOS and see if it is up to date.  There was a memory issue back in 2016 with some 1070 cards.   https://forums.evga.com/EVGA-GeForce-GTX-1070-BIOS-Update-v8604500070-m2565056.aspx
• Go back to an earlier version of the miner and MSIafterburner, sometimes newer versions have bugs in them.
• Next I would check the cards of oil seepage.  I had one card that the pads they used to help prevent the overheating issues leaks oil and oil got onto the pci card pins and caused issues.  I cleaned it per EVGA. 
• I would then check power usage of the miners.  I had an issue once where an update for greatly improved performance but it drew twice as much wattage per card. could be that you don't have enough power to run all of them.
• Try cards in different computer see if it causes issues.
• Contact EVGA and ask for their help on a memory issue on one of your 3 cards in SLI mode erroring out.
• RMA the card

Thanks for the advise.  It will be a while before I get to dedicate some time to it.  But will update if find the issue.