HiveOS version 210608 Team Red Miner Watchdog GPU dead error … but only some cards… XFX 5700XT Thick and MSI Gaming other RX 5700 run fine.
Yep, took days of troubleshooting my 5700 rig with its constant crashing and rebooting with TRM on 0.6-204@210608. Only 1 MSI Mech OC 5700 crashed repeatedly, the rest were fine.
Going to downgrade to 0.6-203@210604 to hopefully solve the issue.
On top of that, with 0.6-204@210608, the network connection for my R7 rig got all messed up. Couldn’t ping the rig and SSH did not work at all. Downgraded to 0.6-203@210604 and all is good now.
What a mess of an update, pulled out a lot of hair figuring out what the hell was up.
Hopefully the next update fixes these issues.
Well i learned from a Member of the Hive team the new release is much more pickie.
It gots better if you exact take for Core Voltage just this settings:
650 656 662 668 675 681 687 693 700 706 712 718 725 731 737 743 750 756 762 768 775 781 787 793 800 806 812 818 825 831 837 843 850 856 862 868 875 881 887 893 900
if you do not take the exact settings it can mess up.
VDDCI 750 or 800
MVDD 1293 or 1303
for the big cards (3 fans thick edition) the higher velues.
However i also go back to the old version cause one card keep crashing .
I also think they get hotter but my RX 5700 run an 57 MHs so a slide change can make problems.
So that the last card is not running could be my lag of knowledge about HiveOS and the new release.
I can confirm this issue. One of my 5600XT’s consantly crashing with this release. Rolled back to 210604 and it works again without reboots.
OK, so this is still an issue. Made a fresh image install, updated to latest 0.6-207@210812, still GPU 8 constantly crashing. Have tried with core voltage and MVDD / VDDCI settings above, but did not worked. GPU 8 dead all the time. It’s a Sapphire Pulse 5600XT with MSI unlocked & modded bios. GPU 6 and 7 same, but not crashing. Downgrading it to 210604 solves the issue. Any version after that causing the GPU8 to crash.