>Two nights ago, my nic locked up. In /var/log/messages I saw this:
>Jul 27 00:53:01 sanctuary kernel: eth0: tx interrupt but no status
>I get these all the time on two systems that use the eexpress 16. It
>seems this is "normal" behaviour for this driver.
>Jul 27 00:53:11 sanctuary kernel: eth0: CU wedged, status 0040 0000, resetting...
>Jul 27 00:53:11 sanctuary kernel: eth0: i82586 reset timed out, kicking...
>Jul 27 00:53:11 sanctuary last message repeated 4 times
>Jul 27 00:53:11 sanctuary kernel: eth0: i82586 not responding, giving up.
>Jul 27 00:53:21 sanctuary kernel: eth0: transmit timed out, board on fire?
><4>eth0: i82586 startup timed out, status 0040, resetting...
>An ifconfig down, ifconfig up resolved the problem.
>The card was only recently put in the machine. It had been running
>fine under 2.2.10 for 5 days before this happened.
>Any clues? Anything else I could check to help diagnose the problem?
Yes, the problem is the NIC itself. ALL EtherExpress 16 cards
are flawed by design, and even Intel's own drivers randomly crash
under DOS, etc.. I had this problem before earlier this year,
and Alan Cox was kind enough to point this out to me.
The _ONLY_ solution to the problem is to get another Ethernet
card. For PCI, I recommend either a DEC Tulip based card (#1),
or an Intel EtherExpress Pro 100. Both are excellent and have
worked flawless for months on end. For ISA, stick with a cheapo
NE2000 card based on the RealTek 8029 chip.
-- Mike A. Harris Linux advocate GNU advocate Computer Consultant Open Source advocate
Tea, Earl Grey, Hot...
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to firstname.lastname@example.org Please read the FAQ at http://www.tux.org/lkml/