Test Management NIC Failure - both Management IPs go down?

[scooter_]

I have the NICs/Network configured as Follows:

I have successful Pings to:

10.19.128.241 on vSwitch0

10.19.128.251 on vSwitch1

10.19.128.246 on vSwitch1

If I remove the Cable from vmnic1 or vmnic2 all is great.

If I remove both Cables from vmnic1 or vmnic2 I loose connection to 10.19.128.251 and 10.19.128.246 as expected.

If I remove the Cable from vmnic0, I loose connection to 10.19.128.241 AND 10.19.128.251

I'm not sure why I can no longer communicate with 10.19.128.251 when I remove the cable associted with the NIC assigned to vSwitch0 and 10.19.128.241?

Here are the Configs for the Management network on both vSwitches

Thanks!

Scott<-

Message was edited by: scooter! - Added vSwitch Images

[a_p_]

I think the issue may be caused by the two Management Networks being on the same subnet, causing issues in the VMKernel. If you want to have redundancy without connecting the Management Networks to two different subnets, I'd suggest you use only a single Management Port Group.

In case you want to dedicate a vmnic to the Management, you could add all three vmnics to a single vSwitch and configure the port groups with active and stand-by adapters. e.g. Management: vmnic0 active, vmnic1+2 stand-by and the Virtual Machine Network with vmnic0 stand-by and the other two as active.

André

[scooter_]

Hmmm..  So I only have the 6 NICs on the servers.  I would like to have Redundant NICs temed for SCADA-Network and PLC-Network.  So that's 4. I then need a NIC for a Internal Communication Channel for Software we are running. That leaves me with 1 NIC for Managment.

So I dont think I can move things around to get net multiple Physical NICs into the Management Network.

In Configuring the HA Cluster, it complained that I didn't have a Management network configured on a seperat vSwitch.

I could configure a Managemnt network on the PLC network and it would be in a seperate Subnet.