VSphere esxi 4.1 - Network best practises for failover

[navaneedak]

Hello,

I have a ESXi 4.1 single node with 4 Nics. I want to configure network redundancy for this node. In case of any exteranl link or switch fails i want the ESXi to be stable without any loss of netowrk connectivity.Presently i configured the below .

2 Nics - service console - Vswitch 0

2 Nics - VM network - Vswitch 1

I enabled Teaming on eatch switch with the below settings

"Default load balancing = route based on originating port ID"
"Failback = No"

Is the above configuration is OK ??? Can i achieve network redundancy with the above config ??

I have seen some articles mentioning like creating 2 different service console network and eatch with different vswitch... is this an advantage ??

[mittim12]

I've read where people have put an extra service console on their storage network for HA purposes.  It doesn't look like you have a storage network nor do you have a lot of physical nics to spare.    I'd make sure that the two nics in each vswitch go to different physical switches and you should be ok with that setup. 

Once you have it setup just put some test VM's on the server and do failover testing.   That should make you comfortable with the setup.    You can also check out this great series of articles on the vSwitch.  http://kensvirtualreality.wordpress.com/2009/03/29/the-great-vswitch-debate-part-1/

[GavinY]

The best practice for network failover with two NICs in teaming is both NICs should connect to different switches. So if a switch connection fails the nic connected to the available switch take over. You have not posted your "Network Failover Detection" setting on NIC teaming yet. It usually is set to  "Link Status only". "Notify Swiches" should be set to "Yes". Your other two settings for "Load Balancing" and "Failback" are looking fine.

[AndreTheGiant]

See also: http://communities.vmware.com/docs/DOC-10985

Andre