I’m having a similar issue, after a long time maybe 20 hours or longer my miner, T-rex miner, just shuts down and does not reboot, my other miner gminer keeps going with no issue. Even with watchdog going T-rex does not restart. I found a work around for anyone else experiencing this. If you set the minimum watts in watchdog it will restart the rig if say one card drops off, or you lose a miner like my case
Can you show a picture?
Your PSU (Power Supply) is not holding up and it turns off. That’s why you don’t have an error message. If the PSU is compatible with the power of your boards, it is probably in trouble.
About rig logs for troubble shooting, in hive OS logs are default off cause they fill up and cause drive to “wear out”…if u have have flash drive make your self a favor, get ssd and use command logs-on and reboot to enable logs
This can be something inane, like a CPU fan with a shorting coil. Ask yourself what is new? what have you replaced recently (even a lowly fan perhaps) ? I swapped out the board and CPU before I thought of the brand new CPU cooler I’d recently installed - a reputable brand BTW.
The last time this happened to me though it was the PSU. It was fine until it came under load.
Hi Kent,
Where these OC parameters are coming from ?
Connect the first card only, maintain the power as it is and put 0 at the OC for both Core and MEM for this card.
Then let’s see if it is running like this and how long.
If it is working properly, connect the second card also and repeat the operation. Then we will see the next step.
Rgds
Hadès.
I had the same problem : the pc/ worker was repeatedly shutting down at 4-5 hours.
The GPUs were not overclocked , in fact there are sub-clocked relative to suggested profiles.
I realized that b350 chipset was hot, so I used an old cpu cooler to keep it cool .
No shutting down since than.
Configuration: AMD 1700 on Asus prime b350 plus, latest hiveos.
Try to do something like we did, change bios drivers of motherboard. It fixed my problem.
1 Like
I know this a bit late to answer, but maybe someone comes looking for this topic like I did.
My rig kept constantly restarting after I added 4 new cards. The rig had been stable beforehand for aprox 2 months.
The new cards I added was with Hynix memory and I was aware they couldnt be clocked as high as my other cards with Samsung memory. However, one of the cards couldnt overclock nearly as much as the other 3.
I found out via Shellinabox and saw that gpu #0 kept forcing a restart because of too high overclocks. So I simply set the overclock for that specific card very low and increased it bit by bit until it crashed it again. Then went back to the last working OC.
Now it’s stable and I have no issues.
And how can you see in Shellinabox that the 0 GPU kept forcing a restart??
It simply stated an error on GPU0 everytime the miner restarted. T-Rex then told me to lower OC’s and after I lowered my OC’s on GPU0 it’s been rock solid.
Been running for 5+ days ATM without any issues. I did lower my OC’s a lot tho.
This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.