Quote from: nitrobg on February 19, 2018, 10:01:48 AM

Quote from: greatauror28 on February 19, 2018, 04:08:59 AM

Quote from: ANewbieMiner on February 19, 2018, 03:59:21 AM

Hi,
Can anyone help me please and give me some info on this error:

NVML API: failed to get device handle 4, error 15.
NVML API: failed to get device handle 5, error 15.
NVML API: failed to get device handle 4, error 15.
NVML API: failed to get device handle 5, error 15.
NVML API: failed to get device handle 4, error 15.
NVML API: failed to get device handle 5, error 15.
NVML API: failed to get device handle 4, error 15.
NVML API: failed to get device handle 5, error 15.

There are 4 1050 cards, and 2 1060 cards on the miner. Latest NVIDIA driver which for both cards is the same. MB: ASUS Z170-A ( I know a piece of crap!). PSU: 1050W Green. Win10 X64. MSI Afterburner for OC. 80 power limit, CC:+100. MC: +700. Runs at about 50 degree.

I've got lots of miners running stable. But this one works for about 3 days, then the 1060 cards fail, miner restarts, and the 2 1060 cards are gone. When it happens, the 1060 cards look fine physically, and both are fine in Device Manager, but in Afterburner, both cards are frozen without being able to do anything on them. And they are no giving any Hash either obviously. System need to be restarted for them to get fixed.

I'm hosting the system for a third party, the hardware is theirs, but I clean installed the OS, and did everything by the book. Only software issue that I can think of is, that after installing the OS and trying to run it for the first time, Claymore wouldn't start at all which was related to the OpenCL.dll file which for some reason was missing from sys32. I copied the .dll file from an identical system and put it in the miner's sys32 file.

The whole Claymore's log from when the issue seems to start:

18:41:37:870   cc4   ETH - Total Speed: 99.571 Mh/s, Total Shares: 1173, Rejected: 0, Time: 34:22
18:41:37:870   cc4   ETH: GPU0 13.016 Mh/s, GPU1 12.974 Mh/s, GPU2 13.037 Mh/s, GPU3 13.019 Mh/s, GPU4 23.765 Mh/s, GPU5 23.760 Mh/s
18:41:38:355   1738   srv_thr cnt: 1, IP: 192.168.XX.XX

WTF?

I know, right? How could people copy that huge wall of text?

A) Devil is in the details. I'm running over 25 miners with 6-8 GPUs each almost all alone by myself. Build from scratch. many different models. RX. R9. GTX, ... And I had no exp in these stuff 9 months ago. I managed to do all of this, get a perfectly stable system with little to none problems, only by paying a great deal of attention to details. So I do the courtesy to others who may want to help me, by giving them "the option" to see and read the details.

B) If you get off your high horse, you'll see that I put the important parts short and sweet in the beginning. Giving the line of error which I think is the main thing. System spec. And few lines on what's up. THEN. Only then, I wrote "the whole log file from which the problem starts". So the post is not in anyway or from defined by the log file. You can simply ignore it. You don't need to quote it. And it will take 2 sec to scroll past it.

So if you don't have anything to say, stop talking. It's not Facebook or twitter, or text messages with your bff. It's a forum. You can expect time to time a lengthy post. And as long as it's not messy, it's perfectly fine.