[plug] What's "too much work at interrupt"?

Bernd Felsche bernie at innovative.iinet.net.au
Tue Sep 7 19:47:06 WST 2004


On Tuesday 07 September 2004 15:27, Craig Ringer wrote:
> On Mon, 2004-09-06 at 11:41, Bernd Felsche wrote:
> > I'm getting kernel messages when trying to copy files over a network:
> >
> > 	eth1: Too much work at interrupt, IntrStatus=0x0001.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0001.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0040.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0040.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0040.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0040.
> > 	eth1: Too much work at interrupt, IntrStatus=0x0041.
> >
> > What could be causing those?

> I've seen that before with a faulty NIC ; a replacement NIC did the
> trick nicely. Of course, that does not mean that's the only possible
> reason.

> It seems to me that you've omitted some rather important information,
> like what hardware eth1 is, what the driver and version is, what the
> kernel version is, and what the main system hardware is.

It started happening "suddenly". I've upgraded to the latest
available kernel but the problem persists.

Kernel is 2.4.21-241-athlon (SuSE Linux 8.1)

8139too Fast Ethernet driver 0.9.26
eth1: RealTek RTL8139 Fast Ethernet at 0xe4c520 00, 00:20:ed:45:fc:b6, IRQ 11
eth1:  Identified 8139 chip type 'RTL-8139C'
eth1: Setting 100mbps full-duplex based on auto -negotiated partner ability 45e1.

It's on the mainboard; Gigabyte GA-7N400Pro (IIRC!)

There's another NIC plugged in that gets to be eth0:

de4x5.c:V0.546 2001/02/22 davies at maniac.ultranet.com 
eth0: DE450-CA at 0xd400 (PCI bus 0, device 10) , h/w address 00:00:f8:00:73:1b,
eth0: media is BNC.

I hope that the mainboard hasn't been zapped.





More information about the plug mailing list