My networking guys wanted to verify that it is not possible to have a situation where traffic coming in to a vswitch on two separate rails would not end up in a looping situation. I found in the network concepts pdf that this is nearly impossible unless you specifically setup a guest machine to cause it. Can anyone verify that you should never have looping cause by a vswitch or has anyone actually seen it happen? This is on ESX 3.x
Thanks
The vSwitch will not cause a network loop by itself. You can configure a VM in such a manner that it (not the vSwitch) causes a loop.
vSwitches do not participate in spanning tree, so there is no possibility of a vSwitch causing a spanning tree issue....
Ken Cline
Technical Director, Virtualization
VMware Communities User Moderator
THis depends on how you have setup your nics on the Virtual switch. If you have loadbalancing on your nics connected to the physical switch then incoming traffic is controlled by the load balancing policy of the physical switch. THe multiple nics of a vswitch will not be acting independently.
I seem VMs that where seriously misconfigured and caused loops.. it had to vnics to 2 different vswitches connected to same hardware switch. On VM the nics where teamed..
Caused all sorts of weird and wonderfull problems.. removed 1 nics and problems went away.
BY the way on ESX 3.5 network is significantly improved from ESX 3.0X in fact you can enable Cisco Discovery Protocol on them. GRanted does not prevent loops but will defintelly help find them.
The vSwitch will not cause a network loop by itself. You can configure a VM in such a manner that it (not the vSwitch) causes a loop.
vSwitches do not participate in spanning tree, so there is no possibility of a vSwitch causing a spanning tree issue....
Ken Cline
Technical Director, Virtualization
VMware Communities User Moderator
Thanks guys