When failing over to a new site using SRM - NIC Interfaces get renamed

[Darryl_R]

Hello All,

I am wondering if anyone else has seen this issue.

I am using SRM 4.0 with some Windows 2003 R2 VM's.   These VM's have 2 NIC interfaces.  One named MGMT and the other Public.

When I do a failover to a remote location, I have scripts that run on the machine to set routes and such based on the new IP's that are set during the SRM process (IP's are set using customization files created using the dr-ip-customizer.exe tool).  The scripts that run, are hardcoded to rely on the NIC interface name.

Unfortunately when I do a failover, on first boot (when the sysprep customization script runs) the interfaces are set properly,  on second boot, once the new IP's are set, it appears that the NIC interface names are being modified in an incremental fashion.

eg, MGMT now becomes MGMT(1),  Public now becomes Public(1).

If I do another fail over, it increments it again..   eg, MGMT(1) becomes MGMT(2) etc etc.

I setup another set of applications previously and this situation did not happen.  for some reason with these new VM's which are using the same templates as before, are renaming the interfaces.

Anyone know how to stop this?

Cheers!

[admin]

Why not on your protected site have the portgroup as MGMT_P and on the recovery MGMT_R.  Inever have the same portgroups over the two sites, and map then when configuring SRM

[Darryl_R]

Its not the portgroup I am talking about.

I am talking about the NIC within the Windows Guest OS.

during the sysprep process to change the IP of the VM, the process changes the name of the NIC within the guest OS.

[TimOudin]

I'll bet this is Windows registering a new device in the O/S.  If you set the flag to show non-present devices, open device manager and show hidden devices you will likely see multiples of the same network device type.  Removing the non-present devices will allow you to change the name on your interfaces.  If this is the case then this is not a VMware issue but a Windows one instead.

[mal_michael]

Tim

I am not sure this is a Windows issue. Virtual machine should see the same hardware after the failover. So for what reason Windows would detect a new NIC?

Darryl R

I will try to look what happens in my environment. Does this happen during recovery test also?

Michael.

[Darryl_R]

Its only after we use the dr-ip-customizer that this happens.

the mac address and hardware type of the NIC's stay the same.  It seems to be an issue with the sysprep tool that changes the IP somehow, I am just not sure what its doing.

I am going to look into the hidden devices just to be safe today to see whats going on.

We have another set of applications which we already have configured successfully for SRM.. they seem to be fine.  its just these new VM's we've put up.

I do not do "test" failovers, since I have the environmental ability to do "full" failovers it doesnt make sense to do test runs.

[Darryl_R]

So the good news is, we figured it out.

I didnt realize it was a SQL cluster.

the renaming was acutally due to the SQL cluster being re-ip'd.   simple solution is to run a netsh and rename the interface.