VMware Cloud Community
ThomW
Contributor
Contributor

host esx01.domain.com currently has no management network redundancy

I just upgraded vcenter from 2.0.2 to 2.5 with no problems at all. On my cluster i have a yellow triangle on it and when i goto the Summary tab is says:

Configuration issues:

Host esx01.domain.com currently has no management network redundancy.

Any thoughts at all on this?

JOE

Reply
0 Kudos
18 Replies
ThomW
Contributor
Contributor

I forgot to mention. My Service Console is shared with vmotion on all my ESX Servers

Reply
0 Kudos
jprior
Enthusiast
Enthusiast

I have this alert on my cluster also, all my hosts have a dedicated NIC port for the service console. Is there a way to turn off this alert?

Reply
0 Kudos
FredPeterson
Expert
Expert

You really should have redundancy to the service console NIC. Theres no reason not to and the benefits outweigh the extra port usage.

Reply
0 Kudos
jprior
Enthusiast
Enthusiast

So you don't know how to disable the alert then?

As far as whether I should, or have reason, I don't have the connectivity to spare unfortunately. Maybe in 9 months time when budget shortfall needs to be claimed I will purchase the extra hardware required to implement it.

Reply
0 Kudos
admin
Immortal
Immortal

There isn't a way to disable this alert.

Reply
0 Kudos
jprior
Enthusiast
Enthusiast

Thanks for the info :smileyblush:

Reply
0 Kudos
depping
Leadership
Leadership

I did not find a way to disable the message unfortunately. on the otherhand, I do think it's a usefull message. working with vlan tagging on the esx side could solve your "problem".

Duncan

Reply
0 Kudos
ThomW
Contributor
Contributor

so OK folks...what's the solution? Like i mentioned I have it shared with vmotion...could this be why the warning shows?

Reply
0 Kudos
jprior
Enthusiast
Enthusiast

Duncan-

It's useful in a nannying kinda of way - for me - as I know I only have 1 management port, I set it up :smileylaugh: - but yes, I see the reason for the feature. I'd just like to turn it off.

Regarding the 'problem' - you infer correctly that it could be solved with vSwitch level VLAN tagging, however my network group has ruled that out and insists only phyiscal ports on switches they own, manage and maintain be tagged, and by them. Inconvenient for me, but hey - I don't make the world, just live in it.

Joe -

The error message indicates that there is only one network path to managing the box. If you had two then you would have redundancy and therefore no error. It's interesting you have the management port and the vmotion port shared, I was told that wasn't a good idea - for best results keep the vmotion traffic on a seperate network (which in fact I do).

Reply
0 Kudos
SomeClown
Enthusiast
Enthusiast

I'm currently getting this error after having just upgraded to VC 2.5 and just to get rid of the nag, I added a second Service Console to each box and the error is still there. I even restarted VC and can't make it go away. Thoughts?

Reply
0 Kudos
admin
Immortal
Immortal

Did you reconfigure HA on that host?

Reply
0 Kudos
SomeClown
Enthusiast
Enthusiast

I did reconfigure HA and it got rid of one set of errors (relating to HA not being configured on one host) but left the network redundancy alert in place. I actually got some answers in a different thread (http://communities.vmware.com/message/832170) that seemed to point to having to have a redundant physical NIC for management. Interesting thing, however, and you can see it in that thread, is that when I unconfigure HA the alert goes away and I get back my ability to manually migrate my vms from host to host. Once I reconfigure HA again, back to the same problem--alert and no ability to migrate any VMs.

Reply
0 Kudos
smackthat
Enthusiast
Enthusiast

I also just upgraded from 2.0.2 to 2.5 and now have this alert. My vswitches have two NICs, my Svc Console and VMKernel port are separate, and only one of my hosts has this alert even though both are setup identically concerning port groups/vswitching.

Immediately following the upgrade, I had to manually reinstall the VC Agent on one of my ESX servers as it was invalid with a logon failed error. I had to copy the vpx-upgrade-esx-6-linux-32042 to the host and run it manually.

Anyways, not sure that is related, and it did not resolve the mgmt network redundancy alert. What is most concerning is that both hosts in the cluster are essentially identical, but only one has this alert.

Any ideas?

Reply
0 Kudos
smackthat
Enthusiast
Enthusiast

FYI....I put the afflicted host into Maint Mode and rebooted the host which resolved my issue...the alert is no longer present Smiley Happy

Reply
0 Kudos
sroth
Contributor
Contributor

I am gettint this error on the Cluster icon, I entered maintenance mode and rebotted and that did not make it go away

haalp!

Smiley Happy

Reply
0 Kudos
andranigian
Contributor
Contributor

This happened to me after I upgraded from 3.02 to 3.5. I was able to resolve this by removing HA from the cluster and then adding it back again. Once it reconfigures HA on the cluster the message will disappear.

Reply
0 Kudos
cp15
Contributor
Contributor

Ran into this after reconfiguring networking on 3 ESX hosts the other day. Disabling and re-enabling HA for the cluster solved the problem for us.

Thanks for the tip-

Chad P

Reply
0 Kudos
vmmeup
Expert
Expert

Maybe someone from vmware can chime in on this and let us know how these can be removed. If not maybe a patch can be released to allow for a checkbox in the VC config to turn these messages off. In case the developers forgot if ESX is installed on to blade servers there are only two nics total available and most buildout on blades have a vswitch for SC and vmkernel and a vswitch for host access allowing only 1 NIC per vswitch. I would think this is an oversite on these alerts and there is no reason to allow customers to disable them. So how can it be done.....................

Sid Smith ----- VCP, VTSP, CCNA, CCA(Xen Server), MCTS Hyper-V & SCVMM08 [http://www.dailyhypervisor.com] - Don't forget to award points for correct and helpful answers. :winking_face:
Reply
0 Kudos