not VMNetwork neither DVS

[asaraca]

I have a little bit the same problem. In my case since upgrading to V6 replication doesn't work at all. It always fails with the following error message:

Processing configuration Error: Unknown network: not VMNetwork neither DVS.
Error: Unknown network: not VMNetwork neither DVS.

I opened a support ticket (no: 5157824) on december 1st but I didn't hear anything from veeam support since december 2nd, it's now almost 5 days.

Antonio

[Gostev]

Hi Antonio, this is absolutely different problem than OP described, please do not hi-jack topics in future but create your own.

Your issue is likely caused by bad configuration of source VM (one of the vNICs is not connected to any virtual network). Edit the source VM's configuration, and try again - replication should work.

Thanks!

[officechris]

Same issue here, the NIC is bound to a correct network, same network as all my other VM's that are backing up without issue. I had no issues backing up this VM /w V5, only V6 produces the error.

I just tried removing the NIC & re-adding it, we'll see if that helps.

[-Gertjan-]

Same issue here. (case ID 5159968)
1 vm fails with replication on the "not VMNetwork neither DVS" error.

[allanwwi]

I´m having the same problem, anyone fixed it, or find a solution?

Regards

[foggy]

Please read the sticky known issues topic.

[kellison]

I am also having the same problem with all my replications. Has anyone found a fix for this yet?

[foggy]

Ken, this issue is listed among the v6 currently known issues in the corresponding sticky topic. As Anton explained above, you can edit your source VM configuration to workaround it. Thanks.

[kellison]

Alex,

I saw that as well and verified that my VM only has the one Distributed switch. I even went into the VM (Windows 2003 Server), ran the "set devmgr_show_nonpresent_devices=1" and then went in the Device Manager and found an old NIC. I removed that and tried a replication and again and it still failed. I then shutdown the VM, removed the one NIC and then readded it. Tried to run the same replication and it failed.

[tsightler]

It's not a problem with the VM, it's an issue with the VM's configuration. Either the VM has a NIC that is not assigned to an existing network, or there's an orphan in the VMX file. Clean up the VMX file, or assign the NIC to a valid network, and it will work.

[kellison]

I am looking at the VM's via my vCenter - if I "Edit Settings" and look at the list of devices I only see my one Distributed Switch. I have even tried to remove that one distributed switch and then add it back with the same results. Strange that I have replications the carried over from my Veeam 5.x to when I upgraded to ver 6 and they are still working. It is when I tried to create a new replication that I get this error. I have even tried changing the replication to run via the ESX hosts vice the vCenter and that did not make a difference. When you mentioned about "cleaning up the VMX file how would you recommend I do that?

[kellison]

I've been running some test. What I just found is it appears to be "picking on" my Windows 2003 DataCenter servers. I switched to a Windows 2008 VM on the same ESX host and it ran fine. I have tried a replication via 3 different VM's - all Windows 2003 DataCenter but also on different ESX hosts and they all fail. They are all using the same Distribution switch. So no idea what could be different with Windows 2003 DataCenter VM's but that appears to be the common denominator.

[tsightler]

"Cleaning up the VMX" simply means editing the VMX file to remove any left over entries that might refer to previous NIC configurations. For example, I recently helped a client that had an entry like this (this is not exact, I don't remember the exact line):

ethernet1.virtualDev = "e1000"

However, there were no other references to the "ethernet1" device in the VMX file and it didn't show in "Edit Settings" on the VM, only "ethernet0" was fully described in the vmx. Simply removing this "orphaned" entry from the VMX file manually resolved the problem with replication.

That being said, editing the VMX file is a little like editing the registry, you can cause yourself some serious issues if your delete the wrong things. If you are not comfortable with this you should contact support. Support will be able to assist with this issue and may have better workarounds or more insight.

[kellison]

Oh man, you have got to be kidding.

[tbblakemore]

An easier way than editing the VMX file is to simply add a new NIC, save it, and then remove it again. This will edit the VMX file for you, and fix the reference to the invalid ethernet entry. I have done this to fix multiple machines that had this problem.

As a note - In my case, I found that all the machines that were having this problem were P2V (used to be physical, and converted to VMs). Most had at least 2 NICs when physical, but typically only 1 was in use, so the P2V process only converted 1 NIC, but left a reference in the VMX file to a second NIC.

[kellison]

Thanks tbblakemore, I will give that a try and let you know if that works... Sure is a lot easier and safer to do that vice editing the VMX file... Thanks Ken

[tsightler]

Yep, good info. I thought about testing that, but haven't had time to spin it up in the lab. Glad to know that it works.

Also, several of the customers I worked with were also working with P2V'd machines, so I was suspicious of that as well.