We have noticed with newer appliances (standard cluster members, not VSX or Maestro) that we do not get a response to "arping" requests as we do with older appliances, open servers, and CG IaaS on VMware. The issue is present for both standard and bond interfaces. We are definitely getting a response as running a tcpdump in another window for ARP traffic for the address in the arping command shows the reply.
We use arping extensively for simple connectivity testing (ping is often blocked by firewalls), has anyone else seen this or have any workarounds? The tcpdump approach is a pretty ugly workaround as it needs another command (in another window or as a background task) for each arping.
Cheers