I just finished upgrading (clean install) one of my hosts from 3.0.2 to 3.5U2. I've been configuring it and now there is a warning on the cluster saying there is no redunancy on the service console. Right now it is sharing the same vswitch and pnics as the VM network, which is 4 pNICS. Before you start blasting the config, I'm in the process of making changes, isolating etc... 1 step at a time.
Anyways there are 4 team'd NICs for the vSwitch and to me that is quite a bit of redunacy, is there anyway to clear that alarm?
I would see how the pNICs are setup on the vSwitch. But if it was me, I would ignore until all your changes are done. You may 'fix' the problem during your isolation phase.
Edward L. Haletky
VMware Communities User Moderator
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
CIO Virtualization Blog: http://www.cio.com/blog/index/topic/168354
As well as the Virtualization Wiki at http://www.astroarch.com/wiki/index.php/Virtualization
Agree with last post, however could be related to only having one service console port group? We have two service console port groups, one shares save vSwitch as VMKernel and one has its own vSwitch - as we dont see the error it could be the solution? Having two service console port groups protects you against all sorts of issues like VLAN shutdown (on switch), switch errors if configured over 1 x switch etc.. may be worth investigating if you plan to review your config anyway.
4 x VMNICs for SC and VMKernel could be split into 2 x VMNICs for SC1 and VMK then 2 x VMNICs for SC2 only? Just a thought.
It didn't make much sense to me after I posted this because the host is configured exactly the same as all the other ones. Only thing is this is the only 3.5 host... Still I think I joined the 3.5 host into the cluster too fast before I finished making all the config changes, including the networking side of things. I think the warning was a residual warning from when it was first brought into the cluster, configured for HA and only had 1 pNIC assigned to the vSwitch. I finished configuring the networking and had to reboot the server for the service console memory change anyways. When it came back up, reconfigured HA on the host and the error went away, at least for now.
And yes I know the NICs aren't allocated how they should and I'm in a R&D zone right now getting the configuration down to move the nics around to isolate mgmt(sc) / vmotion / lan / dmz traffic. Thanks for the helpful reminder that it wasn't configured correctly in the first place