SureBackup Job VM's flipping to DHCP

[officechris]

When I restore some production VM's into a SureBackup job /w lab, all the VM's come up with DHCP enable on the NIC's. I thought VM's in SureBackup jobs should come up with the same static ip address that they had when they were backed up in prodcution?

I don't have any issues if I enable DHCP on the proxy appliance, but I was under the assumption that the Isolated network VM's should maintain the same static ip address that they had in production?

[Gostev]

This is a known Windows issue:
http://kb.vmware.com/kb/1020078

[jeepdude123]

So can you clarify, if you don't install the hotfix on all the servers running 2k8r2 and win7 (that you're trying to use surebackup with) and you are running those Vmware VMXNET3 adapters you will not get your static IP?

This is what I'm running into. If this is the case, why is this not a sticky or something? Seems like a pretty common issue that everyone will have if they're running VMXNET3 adapters and VMware, but yet I've farted around with the VLAB stuff for days thinking it was other problems. Frustrating. Guess I should have called Veeam support?

[Cokovic]

So can you clarify, if you don't install the hotfix on all the servers running 2k8r2 and win7 (that you're trying to use surebackup with) and you are running those Vmware VMXNET3 adapters you will not get your static IP?

This is what I'm running into. If this is the case, why is this not a sticky or something? Seems like a pretty common issue that everyone will have if they're running VMXNET3 adapters and VMware, but yet I've farted around with the VLAB stuff for days thinking it was other problems. Frustrating. Guess I should have called Veeam support?

Yes thats right. You will need the patch in order to get a Windows 2008 R2 or Windows 7 up and running with its originating IP within a lab in case you are using VMXNET3 NICs. It's because your VM within the lab will get a new NIC and therefore Server 2008 R2 (or Win7) will generate a new serial number for this interface. Thats why you are losing your IP address. A full VM restore is not affected by this behaviour (as far as you have unregistered or deleted the old VM before restoring it).

And its always a good idea to give Veeam a call or open a support ticket over the support portal. Or just ask here in the forums. Would have saved you alot of time researching

[pizang]

I have run into the same issue. What is surprising patch didn't resolve a problem. I had to replace VMXNET3 with older not paravirtualied E100 network card.

I do not like this idea so I have tried another options. I have just set on one VM MAC to static (Manual in vCenter). After backup I tired a SB job and it was ok - IP was retained. I find this idea better than changing network card or installing a patch.

[Cokovic]

If you have just a few VMs then this is a possible solution too. In our case we have 1500+ VMs so no way to handle MAC adresses manually

[VladV]

I have configured a SureBackup job consisting of:

1 application group with 3 domain controllers (DC1 and DC2 are on Windows Server 2008 R2 and DC3 is on 2008). Their IP addresses are 192.168.1.11-12-18
1 virtual lab with a proxy with IP obtained automatically from 192.168.1.x, vNic1 on isolated network with appliance IP 192.168.1.1 255.255.255.0 and masquerade IP 192.168.254.D NO DHCP because all 3 DCs have static IPs.

When booting up the DC1, it takes up an ip like 169.x.x.x because it does not find a DHCP which it shouldn't because the actual DCs have static IPs. If I setup the vNic 1 to have DHCP then the DC1 and DC2 gain 192.168.1.2 and 3 respectively and DC3 remains static (as it should) on 192.168.1.18.

I have two questions:

1. Why are DC1 and DC2 on DHCP when the actual VM are on static?
2. Is there any possibility to boot 2 machines simultaneously? It would be helpful in case of multiple DCs because it will reduce the bootup time significantly by letting the DCs find each other as soon as possible.

Thank you,
Vlad

[foggy]

Vlad, looks like the second post in this thread answers your first question.

Regarding the second one, you can use linked backup jobs instead of application group and configure the Process simultaneously up to ... VMs parameter to specify the maximum number of VMs that can be started at the same time. You can find some additional details in this thread.

Hope this helps.

[VladV]

This fixed my issue. Thank you.

For the second question, it seems that I cannot maintain the VMs up after the scripts are running for those included in the linked backup jobs.

[foggy]

Yes, this option is available for application group only. You can try another possible workaround described in the same topic, though (using Instant Recovery).

[dellock6]

I had the usual problem today at a customer with Win2008R2 and vmxnet3, we deployed the microsoft patch but even after a new backup, surebackup still created a new lan connection with all the related problems. Going to wait for this night's backup and try again tomorrow, maybe the patch did not applied correctly, or we found a situation where the patch does not work....
I'll update the post after tomorrow's tests.

Luca.

[dellock6]

Ok, did some further tests today, and two VMs out of 3 are working after the patch, one was also a domain controller, so I can confirm the patch works, and is a deployment problem of the patch itself.

Luca.

[jim3cantos]

This is a known Windows issue:
http://kb.vmware.com/kb/1020078

Second time in two days that I encounter myself with this "bug". Yesterday SureBackup Ping Test Timed Out, but only in a few vm's. Rest of them flipping to DHCP and again the same culprit.

[PaulNSW]

Just tried out surebackup for the first time, and it also changed my DC to DHCP instead of static. A little wary of installing hotfixes on DCs, but looks like there is no other choice!

It would be useful if the Surebackup wizard had some kind of pop-up or warning about this issue