I haven't seen this before and it's causing me all kinds of headache.

I have a Cisco 3560G with
* some local VLAN interfaces and RIP2
* a router-on-a-stick and RIP2 (on a trunk port)
* a Xen server on a trunk port (VMs can be on one or more of three
different VLANs)
* some hosts on access ports

RTR is the Router-on-a-stick
HostA is a Virtual Machine on the Xen server.
HostB is a physical server on an access port.
HostA and HostB are on the same subnet.
RTR is the default gateway for the subnet of HostA and HostB.
RTR can ping both HostA and HostB
HostA can ping other hosts on access ports and hosts on other networks.
HostB can ping other hosts that are VMs and hosts on other networks.

HostA can NOT ping HostB (or other VMs on that network) and HostB can
NOT ping HostA (or other physical servers on that network).

I can see ARP requests on RTR from both HostA and HostB, but the hosts
themselves never see the ARP requests from the other host.

If I manually add the hosts to each other's ARP table then HostA can
ping HostB and visa versa.

So ... What the heck is going on? Why isn't the switch forwarding the
ARP requests to all ports on the same VLAN?

I think the trouble started when I enabled RIP2 on the switch. But
neither I nor Google can figure out why that would matter or how to
fix it.

-- 
Perfection is just a word I use occasionally with mustard.
--Atom Powers--
_______________________________________________
Tech mailing list
[email protected]
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to