Rig keeps crashing/restarting after sometime, please help!

[Kent99]

Please! I need help, it’s been 2 weeks of losses and searching…

-The problem is my rig keeps rebooting after sometime (maybe 10 mins or 8 hours!)
Using h110-d3a motherboard connected to two GPUs (inno3d 3070 and gigabyte 1660S) with evga psu 850
I get no miner errors on hiveos platform

I tried to
1- Lower my oc and still nothing! (check OC in the photo)
2- Changing the usb to different model with new flash
3- reconnected everything and cleaned the dust
4- tested the motherboard alone and worked no issues…
5- tested the 3070 alone and still restarting after hours of working…

Where could be the problem?
Maybe motherboard settings? The mining mode? Internal Graphics enabled!!
RAM? since it is used?!

[Smining570]

Just to confirm, there is no error code or nothing on web qui after Crash? And there is no errors in miner logs or tail log about Crash event?

[Kent99]

@Smining570

thing on web qui after Crash? A

I cant find any red alerts anywhere, I’m newbie!

[Smining570]

hey newbie, time to stuby then youtube
Allso Knowledge base havE good info about logs and how to read them. Good luck

[birdbot]

I also have a similar issue. No errors in HiveOS. Sometimes runs for 30 hours, sometime just 30 minutes. Swapped PSU, motherboard, changed overclocks, moved from USB to SSD, done clean OS multiple times, tried updating, rolling back, nothing.

I have a 6 card 3070 rig. It has got to be a riser or something at this point, right? Not even sure how to figure out which riser it could be.

OS just hangs and miner goes offline, not submitting shares or anything. Next time it crashes, I’m going to check the logs again and see if I can spot anything, but… riser issue, right? Has to be.

What I don’t understand is how it could run for 30 hours if it’s a riser issue. Seems so odd to me for it to run perfectly for so long and then it still be a riser issue.

[Smining570]

if riser, ther is often dead gpu, autofan unreal temp 115 error… etc error. next…are logs on commands applied?.. so u can check later what gpu is causing these issues… most off the time its too high oc anyways…

[Kent99]

the same problem and same question…if there is a failure in somehting (riser or psu)
HOW IT CAN RUN for more than 12 hours as we mentioned…

I’m so tired and disappointed…been like that since 2 weeks…

[birdbot]

Logs are on - is there a way to export them easily?

I run a somewhat mild OC on the cards. 1100 for absolute core clock, +1600 on memory. Temps all below 51C, pulling b/w 106 through 116 watts.

[birdbot]

I don’t know, been trying to trouble shoot the issue for a while - cost me a lot of downtime, swapping parts, most of all stress.

For what it’s worth, your memory OC is really high on your 3070.

[Kent99]

feels sorry…

I tried 2400 and still the same problem? 2400 is still high?

[birdbot]

I’d try something like 1600 and then work upwards from there.

[Smining570]

knowledge base- working with logs

[birdbot]

What you’re referring to says nothing about exporting the logs from the rig itself.

[Smining570]

On miner icon u can get lates miner log.
use command: message file “syslog” -f=/var/log/lastlog.log
allso if u enter mc from the shell, there u will fing you miner and its last logs, or u can input; (teamredminer=miner u use)
message file “teamredminer.log” -f=/var/log/miner/teamredminer/teamredminer.log

[pitnick]

Hello, what is the version of your risers ? Do you mix several versions of risers ? I have similar problem on 2 rigs. On my rigs, I had two different version of risers (006 and 009S). V006 seem to be unstable. I put only 009S risers and now it is fine. For the other rig. I tried to use 006 risers but same problem. Rig freeze after 10 - 20 hours. (disconnected under Hiveos) . I ordered V009S risers ans I hope it will be fine on this rig too. Regards, Nicolas.

[cstvorka]

Thanks $mining570 and pitnick
Had as well few times in last 2 days error dead gpu + restart. Will change them, have 006 model. hopfuly it helps.

[birdbot]

I have 009S risers. I’ve been swapping a new riser in every time the rig goes down. I don’t want to waste time trying to isolate the riser by running just one card at a time.

I ordered more risers as well, hopefully I have better luck with them.

[birdbot]

I checked my logs - they don’t say anything, just show the normal operation of the rig without any errors being generated.

[val66]

Hello do you have new about your problem ?

[vip]

I had a similar issue. The power supply wire going into one of my cards seemed to be the problem. I replaced all of them and it seems to be fine since. Been running it for about 12hrs now.