I'm having trouble with an onboard nvidia gigabit network, under times of heavy load on the network, I'm seeing this error logged:
"too many iterations (6) in > nv_nic_irq"
I'm running Hadoop DFS over these NICs and I see checksum errors build up until the whole thing just fails.
I'm running the 2.6.26-2-amd64 kernel, and my initial research seems to imply its a problem with the forcedeth driver.
Has anyone run into this problem before?
Thanks in advance if anyone can help! Ryan