Error on second replica attempt after first full replica

[rizks_2000]

Am getting the following error. I performed a full replica about two weeks ago and have not attempted anything since until this morning. This has worked previously in testing. The replica does already exist on my target destination. But shouldn't it just perform an incremental replica job?

Finalizing target session Register VM failed, configFile "[Iomega-NAS] VeeamBackup/SOA-APPWORX-P(vm-3105)/SOA-APPWORX-P.vmx", name "SOA-APPWORX-P_replica", isTemplate "False", poolRef "resgroup-2646", hostRef "host-2647", folderRef "group-v2203" The name 'SOA-APPWORX-P_replica' already exists.

[vbussiro]

Didn't you powered up your replica with vsphere client ?

[rizks_2000]

When the full job completed, yes, I did power it up to see how things looked. I then powered it back down and it has been that way since. Should I have not powered it up?

[vbussiro]

No you shouldn't. Please use veeam backup gui to "perform failover to replica". Doing this, you won't break the replica "chaining", and be able to continue your replica job.

[rizks_2000]

Ok. What does this exactly do for me? I guess I am not clear.

[vbussiro]

By powering up the replica with vsphere client, veeam isn't aware of it, so its next job run won't be consistent, and fails.

You must power on the replica through veeam interface choosing "restore-> failover to replica". you'll be able to choose a previuos replica if needed, and then "fail back" to the original machine when it becomes available again. Doing this, you 'll be able to continue your further replications. Quite difficult to explain, but try it !

You can, in a panic scenario, power on the vm directly from vsphere client, but you'll lose ability to restore previous replicas and break the job.

[rizks_2000]

Ok, by doing this though, I am in no way affecting my source VM which I originally made a replica of, correct? I am not interested in even attempting something which will affect my production VM.

[Gostev]

No, source VM is not touched during failover and undo failover operations - only replica VM is affected.

Alternatively, if you'd like you can test replica manually using the following procedure in order to be able to continue replication:
http://www.veeam.com/forums/viewtopic.p ... ing#p10941

This allows to bring up replica VM on isolated network during testing.

[rizks_2000]

So I choose Restore, Failover to Replica, Perform Failover, Navigate out through the Job Name tree and choose my one and only replica I have listed there which is my original full that I can rollback to?

[Vitaliy S.]

That's correct.

[rizks_2000]

Ok, so I performed this and currently the replica is powered on. Do I now power it off and then try to run my original replica job again? Or do I run the replica job with it my "Replica" powered on?

[Gostev]

Hi Anthony, please clarify what exactly have you performed, as there were 2 options provided above?

[rizks_2000]

I performed the failover to replica as mentioned above. So as it stands now, my replica which sits on an ESX host over at my DR site is now powered on. When I performed the failover to replica, as directed too, it seemed to create a quick snapshat, reconfigure VM, and then powered on the Replica VM. Thats how it stands. I want to run my replica job again.

[Gostev]

OK, in that case you should just perform "Undo failover" using Veeam Backup UI. Then, you will be able to continue running your job.

[rizks_2000]

Ok Gostev...this is what I am now getting. I know it already exists. I can see the replica mounted (not powered on) on my DR site ESX host. I guess I am half tempted to kill the entire replica and start over fresh. Not sure what else to do. I'd performed the failover to replica. When I did that, it completed successfully and actually powered on the replica on the DR site ESX host. You had then said to Undo Failover and re-run my job. I did that and got the following:

Replica: APPWORX-P
Created at: 4/2/2010 1:25:52 PM Created by: ******\support
Session Details
Status Error Start time 4/22/2010 11:01:03 AM Details

Total VMs 1 End time 4/22/2010 11:53:07 AM
Processed VMs 1 Duration 0:52:03
Successful VMs 0 Total size 40.00 GB
Failed VMs 1 Processed size 40.00 GB
VMs in progress 0 Processing rate 13 MB/s

Processed Objects
VM name Status Start time End time Total files Processed files Total size Processed size Processing rate Duration Details
SOA-APPWORX-P Error 4/22/2010 11:01:16 AM 4/22/2010 11:53:07 AM 5 5 40.00 GB 40.00 GB 13 MB/s 0:51:51 Finalizing target session Register VM failed, configFile "[Iomega-NAS] VeeamBackup/SOA-APPWORX-P(vm-3105)/SOA-APPWORX-P.vmx", name "SOA-APPWORX-P_replica", isTemplate "False", poolRef "resgroup-2646", hostRef "host-2647", folderRef "group-v2203" The name 'SOA-APPWORX-P_replica' already exists.

[Gostev]

Anthony, yes sure you can recreate the job and start over fresh. But if you'd like and have time to nail down why this happens, please send full logs to support@veeam.com and we would have developers to take a look at this issue and why it happens for you. Thanks!

[rizks_2000]

Thanks. Spoke to one of your phone support techs (Damon) and he basically led me in the direction of starting fresh, so that is what I did.

[rizks_2000]

Gostev -

I run replica jobs only currently using version 4.1. I've had the product one month and am really finding out about all the features. My goal is to have replicas run on a schedule and if I need to go back and restore the the entire VM, I can do that by a Restore job. I also want to be able to restore at the file level as well. I believe I can do this by failing over to replica at a specific restore point. Once I do this, I can go find the file(s) I want, move them somewhere out on the network, and then undo the failover. Is this correct? Should I be performing Backup Job instead?

[Gostev]

Anthony, no you don't need to failover to perform file-level restore. Something unique about our product is that we provide file-level restores directly from offline replicas. There is a corresponding option in the Restore wizard, check it out.