Rig total freeze, requires power reset

Hi. I have same issue, rig freezes in random times, only hard restart helps. Sometimes working 12h, sometimes 8h. My cpu is RYZEN 5 1600, and B450 MSI mobo. I have solved this by adding little cpu overclocking ( CPU loadline calibration control - mode 3 and underclocking my RAM from 2400 to 2133 (i have 2400 memory) ) Now works great. Maybe it will helps for you.

Same issue here. Sometimes it works 24 hours straight and other times 5-6 hours before it freezes. No error message, it just freezes. Maybe it will be interesting to post our rig builds in order to find a common issue:

  • MSI X470 GAMING PLUS MAX
  • Athlon 3000G
  • Kingston HyperX Fury Black 8GB DDR4 3200Mhz PC-25600 CL16
  • 2 x PSU Corsair RM750
  • Risers: VER009
  • SSD installation
  • System version: 0.6-203@210424
  • Miners: T-REX for the 3070 and Teamredminer for the amds.
  • Cards and OCs:

Hey I have same issue, suddenly my system started freezing, 2h -4h -6h very random, and need to turn off the system and turn it on back again, how to do CPU loadline calibration? can you please guide?

My system config is:
Ryzen 5 1600X
MSI GA-AB350-Gaming 3
8 GB 2666 ( today I have set it to 1866 let see how it goes.)
128 SSD
Miner: GMiner 2.53

mix rig 5700XT + 580XT, 1060 x3

CPU Temp stays around: 60 degree.

In your bios should be CPU overclocking settings, it should be there in cpu power mode or something like that. Just go to bios and try to find it. That helps for me, but i donā€™t guarantee that will helps for you.

Iā€™ve been having random crashes ever 2-4 days, I setup a telegraph bot, so I can reset the power remotely with a smart plug.
Every now an then it would turn off just as I went to bed. So a good 6-7 hours before Iā€™ve notice that itā€™s not running anymore.
I ended up created a small javascript page, that pulls the rig data, and if the rig is offline for more than 3 minutes, commands the smart plug to reset via IFTTT

3 Likes

After the last update my rig has been mining non-stop for two days. Thatā€™s a record for me.
Maybe the problem has been solved??

Hello, before trying anything else, I would reinstall everything from scratch on a ā€œknown to work SSD/HDDā€ and use the cards with no flashed memory timings and with no overclock. I would maybe only do a slight undervolt so that they donā€™t heat too much during the testing. If the freeze keeps happening, THEN I would spend time investigating.

In my case, I think that diminishing the overlock and memory timings on 2 of the 6 cards helped with stability, as well as a fresh reinstall.

Hi! How can I check, which card freezed? My computer is freezed and after restarting I canā€™t see which card not worked.The miner restarting and itā€™s log file rewrited.

The miner never reported one card with errors or bad submitted blocks? If so, then you have to try one by one over a few days

I canā€™t see miner report because when I see that itā€™s already restarted. I have 9 cards in the rig. One by one solution is very expensive.

Hello @eavmarshall.
Is there a difference with your script and hashrate watchdog proposed by HiveOS?

Does not work with watchdog :frowning: ā€¦
@eavmarshall => If you can share you script, would be fine :wink:

Guys my rig keep on freezing on random times and goes offline. Iā€™ve tried different profiles for my gpus but the freezing still persist. Last profile was - 500, 2200, 130.
Now Iā€™m running on default stock without any overclocking or power limit. I have a double fan 3070, a triple fan 3070 and a 3060ti.

Iā€™ve noticed that the double fan 3070 is running at very high fan percentage. Could i be facing a problem with this gpu leading to failure and freezing

It eventually crashed so the update was not the solutionā€¦
I changed the risers and so far so good. Maybe that was the problem.

My rig randomly freezing after underclock, imo & in my case, its not the motherboard or the bios, it is fine if not underclocking, but itā€™s high power required & smooth running as always, it had to be hard reset, and its little bit annoying, still finding best and stable setting with the underclock

Hello, I have the same issue since the update.

Before it was running stable with Hive 0.6-190 for a few days.
Since the update to 0.6-203 it keeps on crashing.

My System:

Ryzen 5 1600
MSI B450 Gaming plus max
16 GB 3200 ( already reduced to 2333)
16GB USB Stick
Miner: GMiner 2.53 / T-Rex 20.1

I tried already chaning from USB Sticks and SSD installing HIVE from scratch. ā†’ still freezing.
I tried reducing the DRAM freq ā†’ Still freezing
CPU Temp is always at 35 degress, so it should not be an issue.

My overclocks and the risers should be ok, since I donā€™t get any errors on the GPU or from the Miner.
I only get the notification that HiveOS is Offline.
Also the logs do not show any error or occurance, it simply freezes.

I have a work around with power monitoring on a smart plug, each time if power is less then threshold for more then a minute it switches off the plug and switches on, but that should not be the long term solution.

Would be glad about any other hints or proposals.

Seriously, try different risers. I was with the same problem FOR MONTHS and I tried everything. Sometimes I get reboots every 2 hours, sometimes 2 days but at the end it crashed.
A few days ago I changed the risers and no more reboots since then.

1 Like

Thanks Juan, I actually have good risers and they were working well without freeze on hive 0.6-190.
Just 0.6-203 is making that issue.
And usually the single GPU crashes if bad risers are the root cause.
I will keep on changing after each crash:

Nvidia driver: 460 vs 455
Different Miner version: T-Rex 20.3 /20.1, Gminer 2.49 /2.54

OC with absolute clock (1075) vs delta clock (-500) with power limit

And downgrade back to Hive to 0.6-190.
In last step Iā€™ll simply try another OS if no improvement will be found.
Even on Windows it was running a week without errors on the same set up.

Any other hints and tips are welcome.

1 Like

Whatā€™s your ssd capacity?

Itā€™s a Kingston SSD 120GB