12 Card Rig Randomly Crashing - Need Help With Logs

[drfrag]

Hey guys, I’m in the process changing all of my 6 card rigs over to this new 12 card board that was recently released:

and I’ve been having some issues since going from 10 cards to 12 cards.

I don’t believe its a problem with the board…I think its an issue with one of the cards OC settings or something that ends up taking the whole system down.

I turned on Watchdog and that actually helped some but I can’t seem to keep this thing running for more then a couple of days.

All my rigs are in a commercial building a few miles away and whenever something goes hard down it means I have to get in the car so I’m trying to see if there’s a way to get a log of the crash causes.

I’ve looked through all the logs listed here : https://hiveos.farm/troubleshooting-logs/

But all I see is logs created since the last boot…

Is there a running system log that saves to the HD that will tell me historically what caused the last crash?

Also, I have a mix of NON-LHR and LHR cards and even one AMD card in the mix.

All of these cards run in my 6 card boards for days or weeks without a crash but since moving over to the 12 I’ve had some issues.

Thanks!

[Molock]

Hi.
Connect to your rig in SSH and type:
logs-on
and press enter.
Your logs will stay on your disk on next reboot.

[drfrag]

Awesome thanks doing it now.

Also, which log do I want to look at then after the next crash?

[Molock]

After a reboot, go to to your rig, in /var/log/miner/YOURMINER/
you will see the current log and the old logs (numbered from 1 to …xxxx) in tar.gz format (or .tar format)
Uncompress it and open the log inside the archive and it’s your logs before crash. For me it’s t-rex.1.log.gz for exemple.

[drfrag]

Awesome thanks, I will do that. Hopefully I can determine the cause…

[drfrag]

So this just happened…

I’ve looked it up and no one seems to have any idea what to do…

[Molock]

If you use ETHPILL or GDR5XPILL, put a 60 seconds delay before applying it.