VMware Cloud Community
BHagenSPI
Enthusiast
Enthusiast
Jump to solution

Basic (unicast) connectivity check (normal ping): Failed, VSAN 6.5

My VSAN 6.5 cluster continues to fail these health checks despite my best efforts; it won't even respond to "fists-full of hair" sacrifices. I think I'm missing something pretty basic here, but it is simply escaping me.

I've gone so far as starting completely over with my cluster: moved all 4 hosts out of the cluster, deleted the cluster, deleted the distributed vswitch and associated port groups / uplinks, etc...back to a vcenter, a datacenter, 4 hosts, and a standard switch with default management and vmnetwork Networks.

Then I created a new dvswitch, created a new cluster, enabled vsan, added the hosts to the cluster, created port groups, added uplinks, created vmkernal adapters on each host pointing to the vsan portgroup...and still can't ping between hosts on the vmk. Check the PuTTy screenshot for the failure at the CLI level.

vmk1 is the vmkernel adapter connected to the vsan port group that lives on the on the dvSwitch.

vmk0 is the vmkernel adapter connected to the Management Network port group that lives on the default, standard vswitch.

Help?

vsan2.jpg

Here's a shot of the vmkernel adapters on host 04, as an example. The other hosts are setup the same way:

vsan3.jpg

Reply
0 Kudos
1 Solution

Accepted Solutions
BHagenSPI
Enthusiast
Enthusiast
Jump to solution

Ok, I figured it would be something relatively simple. A weekend away and fresh eyes/brain cells helped!

I'm running a distributed vswitch. Turns out I missed the fact that the uplink port group I created is apparently automatically created as a trunk port.

So I simply changed my (physical) cisco switch ports (that are associated with that uplink port group) from a access to trunk ports, and now all the network tests pass and I have my first working vsan Datastore.

Whew...

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 60

switchport mode trunk

View solution in original post

Reply
0 Kudos
2 Replies
elerium
Hot Shot
Hot Shot
Jump to solution

Does the ping actually work if you try from SSH?

vmkping 10.108.60.10 (replace this IP with the VSAN vmk interface IP of other VSAN hosts)

If it does, then it's a bit of a puzzle why the health check fails. If it doesn't work, you have a network configuration issue (maybe IP,vlan or switch related) or something wrong with your dvswitch/portgroup config (vlan or uplinks maybe).

Reply
0 Kudos
BHagenSPI
Enthusiast
Enthusiast
Jump to solution

Ok, I figured it would be something relatively simple. A weekend away and fresh eyes/brain cells helped!

I'm running a distributed vswitch. Turns out I missed the fact that the uplink port group I created is apparently automatically created as a trunk port.

So I simply changed my (physical) cisco switch ports (that are associated with that uplink port group) from a access to trunk ports, and now all the network tests pass and I have my first working vsan Datastore.

Whew...

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 60

switchport mode trunk

Reply
0 Kudos