I had recently upgraded my vcsa from 6.5 to 6.7 and it was working fine-- I decided to migrate it to another host I added which had more capacity. the migration went fine but when I went to log on to vcsa running on the new host the next day it had reverted back to version 6.5 and all the other hosts and vms are disconnected. I tried going in to maintenance mode on the new host and rebooting but this did not fix it. do I need to unregister and reregister the vcsa vm on the new host ? TIA
but when I went to log on to vcsa running on the new host the next day it had reverted back to version 6.5
Not sure exactly what's happened here, but this just isn't possible.
do I need to unregister and reregister the vcsa vm on the new host ?
At this point, we need more information about what's going on in order to give advice.
quite right -- it is not possible unless you goof up the autostart-- turns out the old host had been rebooted and the old vcsa (preupgrade 6.5)vm copy that was left over after the upgrade to 6.7 restarted automatically :smileysilly:--my apologies for the mistake.
Another question that did arise is the the old host VM Network had VLAN 200 and the new host VM network did not (ie vlan 0)
Now that i killed the old vcsa vm and got the new one running its IP is the same as before but I cannot access it unless i add VLAN 200 to the VM Network vswitch on the new host. Does VCSA retain the vlan properties of the initial network it is created on ? (i.e. I am not trying to change the IP address of the vcsa just the properties of the network it is connected to).
Normaly the VLANs are set on Portgroup level and not inside the Guest OS. So go directly to the use by using your browser https://esxi-ip/ and modify the port group or choose a the right portgroup for the VM.
Regards
Joerg
I thought that was the case, however when I change the VM Network port group on the esxi host page to zero I still cannot access the vcsa vm (except via console)
is it potentially a MAC address problem because that would change during migration correct?
tried deleting the port group and recreating it with VLAN 0
still unable to connect to vcsa unless i change VM Network to vlan 200--can't even ping the address of vcsa
essentially I can ping the esxi host address 192.168.200.12 (management network on vlan 200)
but cannot ping vcsa vm address 192.168.200.5 (VM network on vlan 0)
clearly a network issue but would appreciate advice on troubleshooting
How is your switch configured where the ESXi uplinks for this port group? We'd need to start at the bottom and work our way up.
ignore that vcsa10 has connection to t200 i have been trying different scenarios
t200 is a vlan 200 port group i created to test some new VMs to pinpoint issue
VM network is main portgroup that cant get vcsa10(my latest vcsa instance) to connect when on VLAN0
You need to isolate your testing. Create a vSwitch with a single uplink. Create the necessary port group for vCSA and attach the VM. Test network connectivity. At that point, it's either your VLAN ID on the port group is wrong, or you have a physical switch configuration issue.
ok I traced it to the new unifi controller
I think anything on the unifi switch that uses the 192.168.200.1 subnet is being forced to vlan 200
the reason I can ping/log in to the esxi host is that it uses management network (vlan 200) but the vcsa uses vm network--so if vm network doesn't have vlan 200 set the unifi is not allowing access. I assume this is correct behavior but I have a ticket in with their tech support to understand options to create separate subnet without vlan tagging in the switch. I appreciate your suggestions to isolate this.