Hi All,
Let me explain the topology to let you guys better understand:
2 switches with ip routing enable
2 switches without ip routing enable inside a blade chassis
6 ESXi 5.5 hosts
50 VMs
On all hosts the vswitch0 is configured with 2 nics, both active, with the management network on vlan 10 and vmotion network on vlan 1000 with one nic not used and one active for every vmotion vmk.
The problem is that "sometimes" (because is does not happen always and it fixes by itself) the VMs inside let's say host 1 are unable to ping the host BUT all the traffic to and from the VMs works like a charm.
What I understood is that by unassigning a nic from the vswitch0 fix the problem.
Another thing I found is that the vMotion network, with one nic assigned, was unable to ping the ip of a vmk on another host.
From the physical switch side, all the port are tagged with the vlan 10, 1000 and the other vlans i configured.
The vlan 1000 does not have an ip address and it's not routed like the iscsi one.
Any idea?
Thanks!
