I migrated my vCenter from 5.5 to 6.0 (hosts are still ESXi 5.5). This was done by spinning up vCenter 6.0 on a new VM, then migrating the hosts across.
However now one of my backup copy jobs (offsite job) that uses the WAN accelerator appears as if it is doing a full copy instead of an incremental, and not even sure if the WAN accelerator is working now.
Previously it would take about 5 hours, now its been running for 13 hours and still on the first VM.
The way you have done the migration, your VMs have been assigned new unique IDs, so Veeam treats them as completely new VMs. Which means it needs to perform full backup of these VMs, and transfer their entire images over WAN with Backup Copy job. Of course, this will take a few times longer than an incremental backup.
To avoid this in future, be sure to perform in-place vCenter upgrade.
I figure that the process won't take "as" long as the first time I did the copy because now the WAN accelerator on the target should be populated enough. Is that right?
Also if it takes too long, can I seed it by going to site and copying the most recent backup to the repository and Veeam will "realize" everything is up to date?
scidoner wrote:I figure that the process won't take "as" long as the first time I did the copy because now the WAN accelerator on the target should be populated enough. Is that right?
The time could be the same, but the amount of traffic should be less if similar blocks are met.
scidoner wrote:Also if it takes too long, can I seed it by going to site and copying the most recent backup to the repository and Veeam will "realize" everything is up to date?
Yes, you can do that, however please keep in mind that digest calculation will still be performed.
FYI this is actually over a DSL link (crazy, I know!). 9 VMs in the job and thanks to the bloody WAN accelerator it works REALLY WELL. Hooray for Veeam!
