Isolate HA heartbeat to second Service Console port (vswif1)

[Chamon]

I have two Service Consoles set on separate vSwitches each using different physical NICs. One is on the production network and the other is dedicated to the service console to prevent HA from shutting down the VMs during planed and unplanned network outages. Is there a way to direct the HA heartbeat to use only the dedicated Service console (vswif1) network and not load balance between the two. I would think that this is a per host setting but I am not sure how to accomplish this. Anyone have any ideas?

[langonej]

I still don't understand why people are attempting to use two service consoles in hopes of mitigating an HA issue. To me, proper virtual switch configuration and one vswif will do it just fine. To me, a second vswif for this purpose makes the environment more complicated than it needs to be, but that's just my thought.

[Chamon]

The reason for a second vswif is during planned network outages you have to reconfigure HA and in large environments it is a waist of time if the reconfig can be eliminated.

[langonej]

Why would you have to reconfigure HA?

If a vswif is attached to a port group that's attached to a Virtul Switch with multiple physical NICs dispersed on redundant ethernet switching networks, how does a network outage of one of those networks (e.g. A) affect overall connectivity?

The only way the entire LAN should go down is if all of your datacenters is blown up, God forbid.

[Chamon]

All vswif1’s are on separate vSwitches that connect to physical NIC’s that are on a dedicated network only for use by the Service Console2. It is on this redundant network to prevent HA from shutting down the servers in the event of planned or unplanned network outages.

What I am asking is if the heartbeat can be controlled to only use one network and not the other.

Does this heartbeat consume much bandwidth once the Cluster size is 20-30+ ESX servers? This is for future planned growth. Thanks.

[Chamon]

A little background may help. No one noticed that there were not redundant networks and during a planned network outage all VM’s were shut down by HA. The people on site freaked out. It is now set up with the separate networks to prevent having to reconfigure HA or have all of the VM’s shutdown during the complete network outages.

[Chamon]

Here is what I was looking for. It appears that it is a new function in Vcenter 2.0.2. Hope fully it will help some else. Here is the link to the KB and click on the link to the PDF at the bottom of the page. http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100208... . The configuration is on the 10th and 11th pages of the PDF. Good luck.