Virtual UC environment

[thomuff]

How do you keep the Virtual UC Publisher and the Virtual UC Subscriber on different ESX hosts?

[Jaime Valencia · Accepted solution]

Are you asking about the affinity rules on ESXi or exactly what do you want to know??

HTH

java

If this helps, please rate

www.cisco.com/go/pdihelpdesk

[thomuff]

You answered it. How does that work? 

A funny thing:  I thought my vmware guy said Infinity

I found the doc with the recommendations to avoid all actives on the same server, chassis or site

http://docwiki.cisco.com/wiki/Unified_Communications_Virtualization_Sizing_Guidelines

Redundancy and Failover Considerations

Application-layer considerations (such as Unified CM Cluster over WAN or Unified CCE Remote Redundancy) are the same for virtualized (UC on UCS) or non-virtualized (MCS 7800) deployments.

However, since there is no longer a 1:1 relationship between hardware and application instances, "placement logic" must be taken into account to minimize the impact of hardware unavailability or unreachability:

• Avoid placing a primary VM and a backup VM on the same server, chassis or site
  
• For failover groups, avoid placing all actives on the same server, chassis or site
  
• Avoid placing all VMs of the same role on the same server, chassis or site
  

[Jaime Valencia]

The way this works is that you create the affinity rule to make sure the VMs are always together or you configure them in a way in which the VMs never are placed in the same HW.

You would use the alway together for example for a DB and a web server, the always on separate servers for a CUCM PUB/SUB.

You can read the VMWare explanation here:

http://pubs.vmware.com/vsphere-4-esx-vcenter/index.jsp?topic=/com.vmware.vsphere.resourcemanagement.doc_41/using_drs_clusters_to_manage_resources/c_using_vm_host_drs_rules.html

If the link breaks it should be under

vSphere 4.1 - ESX and vCenter > vSphere Resource Management Guide ->

Using DRS Clusters to Manage Resources

HTH

java

if this helps, please rate

www.cisco.com/go/pdihelpdesk

[thomuff]

You answered it. How does that work?

I found the doc with the recommendations to avoid all actives on the same server, chassis or site

Redundancy and Failover Considerations

Application-layer considerations (such as Unified CM Cluster over WAN or Unified CCE Remote Redundancy) are the same for virtualized (UC on UCS) or non-virtualized (MCS 7800) deployments.

However, since there is no longer a 1:1 relationship between hardware and application instances, "placement logic" must be taken into account to minimize the impact of hardware unavailability or unreachability:

• Avoid placing a primary VM and a backup VM on the same server, chassis or site
  
• For failover groups, avoid placing all actives on the same server, chassis or site
  
• Avoid placing all VMs of the same role on the same server, chassis or site