Im french so sorry for my english.
I has this problem long time ago, my rig going offline randomly and He stop mining but the rig continue to run.
I fix that on putting some fan on all the PCI E and mobo, i see after week of search the MOBO and PCI heat à lot and some fan fix the problem for me.
Thank you will give this a try.
what mb u have
I believed that this problem is due to the memory junction temp at the display cards too high. And your display cards are running at overclock. So it will get error very easy and then stop mining. The overheat is due to the back side heat sink too weak. It just an aluminum plate. The heat power cannot sink to the back side very fast. The solution are:
- don’t run at overclock or adjust to lower clock speed.
- to increase the flow speed. Add more fans as Betmxrc’s suggestion. Or add a back side fans with heatsink for display card.
- add a watchdog. A watchdog can reboot your miner automatically.
I do have a watchdog but that does not help. Can you share the exact settings for the watchdog? Also I have a blower type rtx 3070. I didn’t exactly understand what do you mean by back side fan for display card.
Did you use Autoo Fan? If yes what is your “Target Core TEMP”
I see your gigabyte 1660 Super’s fan running 97%?
And why all your 1660 super different core settings?
Any reboot message or notification?
Yes I did use Auto fan with 57 deg temp. One of the gigabyte gpu seems to perform differently than the others and hence I adjusted the OC settings accordingly. The same is the case with RTX 3070 Asus. One of them works differently as you can see gives me a lower hash than the others. When individually operated these work fine. Yes I do get Offline notification on discord and nothing other than that. The logs are also fine they show no error of any sort.
Let me reply the watchdog issue first. I am using the Chinese watchdog with device ID 5131:2007.
The setting are :
Regarding the back side fan for display card, it is a specical heat sink and fans which install on the display card back side. I had searched it at Taobao as below:
Based on my research, core below 70C is good. So my autofan setting is around 65C.
So fans can live longer (hope).
I cannot find these that can ship to my country but I am planning to put some additional fans around the rig to help with the cooling.
The fans reduce to about 70% on the gigabyte and about 60% on the RTX 3070 when I set the temperatures around 60C so that is what I have been using recently.
What is your miner?
t-rex miner mining on ethermine pool with exodus wallet
ever try phoenixminer?
Yes I have tried it. Its even worse on Phoenix.
I have been experiencing the same thing using Phoenixminer on 3 separate rigs; each with a single card.
I have the 2070S and 3090 underclocked/undervolted as they get hot at temp junction. All cards are running 100% fan speed as well.
Never had this problem before. But the fact it’s happening on all rigs at the same moment lead me to believe the problem may be with the pool. It’s odd that they’re fine all morning and then when the problem occurs, it occurs on all three rigs.
One other thing I considered is a problem with the electricity in my house. (All on same circuit)
Rolling back to see if that helps and will post results.
Same problem here, my rig go offline After few hours of mining. Surprise: on Windows system the Same rig work without problem. I bilieve that the problem Is in HiveOS. Many people facing this problem. Mg setup have 2x 3070 and 3x 1660super
It’s been 16hours since I rolled back to the previous HiveOS version.
Thank you to the people in here that suggested that.
I have had disconnections even after 24 hours sometimes.