Rig back on, the card that wouldn’t show up had a different riser than the rest. Riser wouldn’t work with the 6 pin, changed it to molex and it is recognized now. Temps showed up, changed mem to 950 for the 6600 and back to drive error, no temps.
Making progress though. Thanks for the assists so far.
Also noticed I’m not on the latest OS version
lower your core to 900, you dont need any more than that for 6600/xt to get full hashrate. if you’re continuing to get errors reduce memory a lot, increase voltage some too, leave core at 900 as that wont be the culprit if you get crashes.
All 7 recognized, got temps initially, now back to the error. Will make the changes suggested to mem etc in the overclocks.
do any of the gpu driver errors show the pcie bus location of the card crashing?
so gpu in pcie bus 03:00 is your problem gpu, maybe try that one on another rig to make sure everything works correctly.
does it work without overclock?
Just took the over clock off, still not working.
Just took that GPU out and it’s working. I knew exactly which one it was. Will swap out the riser tomorrow when the new ones arrive.
Have the same cards and I’ve seen the same issues.
-Take a screenshot of your current settings.
-Go to overclock tab.
-Click clean all the overclock and only after that change/remove/move GPUs.
-After reboot do not apply any OC settings. Check if it works. If it does run just fine for 10 min or so apply OC without “pushing too far”. If this won’t help - this is not connected to your settings and is more likely to be GPU/raiser/hardware issues.
P.s. here’s another tip - make sure your Autofan settings are in place because 6600/XT has some weird fan issues.
I have been having same issue with one card…sold it, bought another 6600 XT, problem solved. Some just don’t like to run on Linux.
It may be the card but I think it’s the riser. Greeting a new batch on Tuesday, will update then.
I think it was a hardware issue with either the gpu or the riser. I wiped the overclocks clean and I’m getting 30mh with the 6600xt and 27mh with the 6600.
I am running 20x 6600XTs across 3 rigs all on HiveOS. I use TRM for the miner. I have run into the same issue when starting up a new rig, or modifying an existing rig in any way, for example, adding a new 6600 XT GPU.
Here is my current solution. This applies only to the 6600 XTs I have been working with. Before starting up a new rig or making any change, remove ALL OC/UV settings for the 6600 XTs, in other words reset to the AMD defaults. Bring up the 6600 XT GPUs on the defaults and let the system stabilize for a few minutes. The rig should come up this way without the “No Temps” or “Drivers installed with errors” problems. Don’t wait more than a few minutes because you don’t want you GPUs to overheat. Next, while the rig is still up and running, go into HiveOS → Farm → Worker → GPU settings dial and set you OC/UV values. The OC/UV settings will take while the rig is up and running and you should be good to go.
Also note, if your settings are too aggressive, then when your rig has a reboot you will run into one of the same two errors listed above. If so you may need to back off on the OC settings, or just know that you will need to go through the same sequence if there is ever a reboot. However, the “No Temps” error is a HARD Error – the rig will not come back to life without a power off / power on reset. So I set the rig BIOS to restart at Power On, and run the rig on a smart PDU so I can power off / on remotely.
Hope some of this helps.
Looks like my new risers will arrive today so I’m going to add the 7th and 8th card to this rig. Hopefully I don’t have any issues. Thanks for the advice.
Wipe the overclock setting on all the cards before you turn it off. You’ll thank me later 
6600xt are terrible under HiveOS. After 3 months of trying everything I could found on internet, I realized I’m wasting my time and hashes/shares. Installed Windows and everything is working perfectly. I’m so mad at myself for not going strait with Windows.
