Did anyone find some information on how to restore a Cisco Nexus 1000V?
The license is based on a fingerprint of the host-id from the VSM. In case we loose the VM with the VSM or a ESX host server needs to be reinstalled, this fingerprint is different. So what would mean the licensekey need's to be regenerated.
Has anyone found some information on this?
The hostid would most deffinitely change on a reinstall. Which means you would need to rehost license. There is a procedure for this, but it requires that you work with licensing agents on www.cisco.com.
The process is detailed in the License configuration guide on the www.cisco.com website.
Here is a link to the document.
The 1000V host-id is tied just to the 1000V. It is not tied to VMware software.
If the VSM is changed/replaced a new license key can be sought as follows:
Step 2: Be prepared with following information: Old Hostid, or Original PAK, or Sales Order and Preferred contact method
Step 3: Receive email from Cisco GLO with new license file within 48 hours.
You say that vMotion will be fully supported to the 1000v VSM. Does that also mean that the VSM VM can run on the same VEM and ESX cluster that it is managing? If so, that's different from early information I received.
(Sorry to hijack this thread)
first of al thx for the information about the license and the host-id fingerprint.
I'm just doing a PoC with the Cisco 1000V and had some issues:
1. when the service console of the ESX server which is hosting the VSM get's migrated into the VEM, the connection get's lost and never come's up until the ESX and the VSM get's rebootet
2. it's not possible to connect the VSM Mgmt NIC onto a VEM - connection get's lost and never comes up again - reload of VSM doesn't help
rebooting the ESX Server end's up with a isolated host and no connection to the SC on the VEM can be established, even the VSM is up and running; actually without any connection to the "outside" world. It looks like the VEM is loosing his configuation. The only way to get the ESX server back to "life" is t remove the vswif0 and recreate it on a vSwitch. After that the VSM can be removed from the VEM and reconnect to the vSwitch - everthing is fine again.
So this means - VSM on VEM yes; but this ESX Server can not be rebootet as long the VSM is hosted on it.
There must be some misconfiguration. I'm able to migrate my VSM to VEM and migrate service console port with no issues. The VSM is not in the datapath. So if the VSM goes away the VEM will continue to foward packets based on the last known configuration from the VSM.
Now you can't migrate the VSM at the same time you add a host to the VEM. That creates a catch-22 condition. The VEM needs to talk to the VSM , but yet you are migrating the VSM so it has no network connection.
If you have two nics create another vswitch and stick the VSM connections on that vswitch. Migrate your service console and vmk interfaces from the initial vswitch and then create port-profiles for control, packet, and management. Once the VEM is connected and the port-profiles are created you can migrate the VSM to the VEM.