no management network redundancy

[vtechie]

setup info

2 ESX servers has 2 NIC each .

All 4 NIC ports connected to same physical switch.

One VM is created on a Datastore clustered on SAN seen by both ESX servers.

All NICs are 1 Gbps speed.

Able to migrate VM between hosts when powered off.

Configured Vmotion and also verified that it works fine.

Tried to put a host on maintenance mode while the VM on that host is powered on , it didnt seem to go to maintenance mode properly . Shutdown the VM and tried manual migration and then host went to maintenance mode. This message doesnt seem to go inspite of migration / Vmotion works fine.

After this in Cluster summary , there is this message " Hosts 1 .... currently has no management network redundacy" & " Hosts 2.... currently has no management network redundacy"

HA summary

Configured Failover capacity = 1 host

Current Failover capacity = 1 host

Any thoughts

[depping]

this normally means that the vswitch where the service console is located has a single uplink to a physical switch.

Duncan

My virtualisation blog:

[vtechie]

Thank you

What should be done to resolve this issue

[weinstein5 · Accepted solution]

typically I have seen this when HA is enabled - I do not know if there is a setting to disable this warning but the sure way is to add a second servoce console port

[vtechie]

i ve taken a screenshot and which has the n/w config info...

Im not sure where should additional service console is required.

Need some directions how to go about

Message was edited by: vtechie

pic attached

[MR-T]

What the other guy said is correct, it's a cosmetic message warning you that there's no management redundancy should the NIC fail.

It won't actually stop you doing anything.

To fix the issue (message) just add a second network card to the virtual switch with the service console connection.

Editied: Sorry, just noticed you've only got physical NICS in your server which limits the options.

You can either purchase a new network card, so you've got 3 in the machine. Best option.

or, configure the 2 existing cards to be used for everything. I wouldn't recommend this though, especially as you're using VMotion. And you'll most likely need to introduce VLAN tagging if you don't already use it.

So, it seems like you need to purchase a new card. I'd get a quad port NIC if I were you, then you can grow the solution.

[kjb007]

If you have the expansion slot, get a new multi-port NIC. If this is a blade, and you can't add anymore devices, I would use VLAN tagging (802.1q), and create a trunk out of both NICs, and team them together. This is what I do in an environment. Team the NICs, trunk the VLANs I need (it appears you have at least 2), and then create portgroups with the VLAN id's, so I can separate the traffic, but still use 2 NICs for redundancy.

-KjB

[depping]

I would buy atleast an additional 2 head nic for both servers and provide each vswitch with an addtional port for redundancy. set the vswitch on virtual port id load balancing!

or create a secondary service console on the vmkernel vswitch.

Duncan

My virtualisation blog:

[vtechie]

Thank you all.

I added a service console to the second NIC on both HOSTS.

Then unchecked HA.

Rebooted VC.

Reconfigured HA.

It worked !!!

Interestingly, the warning messages were not displayed even when i did not have this new configuration.Later when i was trying to put the host in maintenance mode to test vmotion, somehow this message got stuck.

Now things ve cleared up.

[max_inglis]

I have a similar problem, I'm using a HP DL585 with 2 on-board and 2 dual-port expansion NICS. I have the 2 ports for the management vswitch plugged into the 2 on-board NICS. I know VMware says not to mix NIC types in your service console, so I made the 2 on-board NICS the ones for the management network. I still get the "serverX has no management network redundancy". Any thoughts? The NICS are plugged into a separate switches, which I thought might have been the problem.

[kjb007]

Make sure your service console is set to use both NICs. Go into the vSwitch properties, edit the service console,and make sure both NICs are included, in either active/active or active/standby configuration.

-KjB

[max_inglis]

The SC has 2 physical NICs assigned yes. I took a suggestion from above and disabled/re-enabled HA on the cluster, and it went away.

Max.