\[fglrx\] ASIC hang happened
-
I have a strange error coming up with, as it seems now, just 1 GPU. Gigabyte R9 280x, Xubuntu 12.04, latest driver.
After a while cgminer stops responding and the kernel spits out the error: “[fglrx] ASIC hang happened”.
cgminer is still running but not responding to anything. Also a killall and restart of cgminer wont work, the rig has to be rebooted.Changing this card to another rig also gives the same error, sometimes after a day, sometimes just after an hour.
So at this point i think it is the card giving the problem or the driver in combo with this card.
Googling this error gives all similiar problems but no real answer tot the problem (yet)Anyone here had the same issues with this type of card ?
For now i’m just running a quick and dirty bashscript from cron to detect the error and reboot the rig.
-
Someone asked me if and how i did solve this matter.
Well, actually I didn’t but made a quick fix :-)
‘Asic hangs’ seems to be related to temperature issues with these R9 280X cards.
So I lowered the voltage to 1.075 and installed a check-for-crash-script like this in a cronjob:Add this into crontab (with sudo crontab -e)
[code]* * * * * root /usr/sbin/crashtest.sh
[/code]And create the crashtest.sh like this:
sudo nano /usr/sbin/crashtest.sh[code]#!/bin/bash
if ( dmesg | grep “ASIC hang happened” ); then
echo “system crashed”
/sbin/shutdown -r now
fi
[/code]