Hello Everyone,

I have VMM 2012 R2 configured and All the Hyper-v host Added to Fail-Over Cluster.

Most of the VMs are tag the VLAN ID and few of the VMs are without VLAN Tag. I have configured the Dynamic Optimization.

Every 10 minutes Dynamic Optimization running and base on the load on the host VMs are moving to different host.

Problem is Vlan Tag VMs are not live migration different host While running the Dynamic Optimization. Rest of the VMs (with out vlan id )are moving  to different host.

Vlan Tag VMs are live migrating through Fail-over cluster and Through normal VMM live migration. 

While running the Dynamic Optimization job VMs i'm getting below errors on VMM job.

Error (26857)

The VLAN ID (532) is not valid because the VM network (INTERNAL) does not include the VLAN ID in a network site accessible by the host.

Recommended Action

Either specify a valid VLAN ID, or do not specify a VLAN ID and a valid one will be assigned automatically.

Any one have any idea please let me know.

No updates, it means no one face similar  issues? 

Just had the same issue, although related to migrating storage for a VM and managed to fix it...

Did quite a few modifications in SCVMM, but the key things that seemed to get this working was that I made sure my VMs were connected to the right virtual network and also that they had a VLAN enabled.

This was done by right-clicking the VM in SCVMM, selecting 'Properties', then 'Hardware Configuration', then select the Network Adapter - 

Make sure the "Connected to a VM Network" has the right (for your setup) VM network selected, then check the "Enable VLAN" checkbox just below that and select a VLAN ID from the drop-down list.  I only had one VLAN configured on my VM Network so the selection was easy.

Once I did this the 26857 error went away and I could migrate storage.

Hope this helps !

• Edited by gje0371 Thursday, July 10, 2014 11:58 AM

Just had the same issue, although related to migrating storage for a VM and managed to fix it...

Did quite a few modifications in SCVMM, but the key things that seemed to get this working was that I made sure my VMs were connected to the right virtual network and also that they had a VLAN enabled.

This was done by right-clicking the VM in SCVMM, selecting 'Properties', then 'Hardware Configuration', then select the Network Adapter - 

Make sure the "Connected to a VM Network" has the right (for your setup) VM network selected, then check the "Enable VLAN" checkbox just below that and select a VLAN ID from the drop-down list.  I only had one VLAN configured on my VM Network so the selection was easy.

Once I did this the 26857 error went away and I could migrate storage.

Hope this helps !

• Edited by gje0371 Thursday, July 10, 2014 11:58 AM

Can you check in Fabric-->Logical Networks-->Hosts (from the ribbon).

there you should see if the logical switches and network configuration are identical across your cluster nodes. This is important when you are working with live migration and dynamic optimization. Since I don't know how you have configured your fabric, I recommend you to read this whitepaper in order to get a complete understanding of the networking setup in VMM: http://gallery.technet.microsoft.com/Hybrid-Cloud-with-NVGRE-aa6e1e9a

-kn

Hi,

we currently have the same going on here. VMM UR5 was installed by mistake (slipped through the mazes of the net, disguised as security update) and we needed to revert back by restoring the last good backup.
Strange thing is, when I run the optimizer manually there is no issue (green job log), when it's done automatically by VMM service it fails.

Logical Switches and network config is the same on both fabric nodes (this didnt change and it has been working fine previously)