Error event - ip: Invalid ARP Target: 0.0.0.0 on w.x.y.z

[Neil_77]

I have received some error events on all the HP Procurve 4000/8000 switches on the LAN which read:

"ip: Invalid ARP Target: 0.0.0.0 on w.x.y.z"

where w.x.y.z is the IP address of the switch itself.

Now, I understand that this message means the ARP packet is being dropped because the target IP address is invalid. All well and good.

But, can anyone give any pointers as to WHY this should be happening on the network?

(My main reason for concern is that I saw a number of similar messages during the "broadcast storm" that I posted about recently so now that I have seen these messages appearing again, I want to determine the cause, in case it leads to a similar chain of events!)

Any help would be appreciated.

Thanks.

[Ron Kinner]

Thought to be a request for a bootp server.

See:

http://groups-beta.google.com/group/comp.dcom.sys.wellfleet/browse_thread/thread/71190ed9f7f623a8/166005a6ffa53e6d?q=%22ip:+Invalid+ARP+Target:+0.0.0.0%22&_done=%2Fgroups%3Fq%3D%22ip:+Invalid+ARP+Target:+0.0.0.0%22%26num%3D100%26hl%3Den%26lr%3D%26safe%3Doff%26sa%3DN%26tab%3Dwg%26&_doneTitle=Back+to+Search&&d#166005a6ffa53e6d

Ron

[Les Ligetfalvy]

Yes, normally 0.0.0.0 is used for bootp or DHCP. Is your switch a routing switch and deploying the DHCP helper?

[Neil_77]

No, the switches these errors appeared on aren't doing any routing - they do connect to some Cisco routers which ARE running IP Helper config though. However, there are no errors reported on the routers.

[Les Ligetfalvy]

Did you disable DHCP Relay agent then on those switches? I think HP enables it by default.

config
no dhcp-relay
write mem
show config

[Neil_77]

Hi Les,

don't think these 4000/8000M switches have any features like DHCP Relay available - they are just bog standard, no frills, Layer 2 switches. (Certainly, in the way we have them configured on our network, at least!)

What I'm still a little confused about is what the original error message is actually meaning? Is it saying that the switch is trying to boot over the network, and since it doesn't have a bootp server configured, it sends the request to 0.0.0.0? (All the IP config on these switches has been configured manually)

If so, then why should it be doing this?

[Andrew_291]

I have the same message on my 8000m when damaged netagent in Powerware UPS was plugged in switch . When I unplug them - none of that messages was issued.