What's happening during "Preparing target VM" in quick migration?

[pkelly_sts]

I'm testing a migration to production of a backup that I've done an instant VM recovery on. FWIW this is based on a backup of a physical server taken with the Veeam Agent but at this stage it's all VBR I believe hence why posting here rather than the agent forum.

The vm is ready (including P2V updates) after about 30 mins which is perfectly reasonable. However, even when choosing NOT to power on the VM, after selecting "Migrate to production" it has just sat on "Preparing target VM" for over an hour so far before I've given up & aborted.

When checking vCenter, a process "Reconfig VM" also appears to be running VERY slowly, having only reached 17% in the 1:07 hrs it's been running.

It's worth mentioning that the repository is based on a Synology NAS in a 5-disk SHR array, however throughout the whole process there is *no* data yet transferred from the NAS (via vPower) to the VMFS so it doesn't appear to me that, at this stage at least, the Synology is the bottleneck.

The whole reason for this test is because I wanted to see what kind of quick migration throughput I should expect from the synology, knowing the limitations.

I must admit that at this stage I don't yet have a proxy on the host so the proxy in use was another physical server (all resources in the same subnet on the same physical switch) which I know would be faster anyway, but I still wanted to understand how it would perform in this scenario, but I expected it to have at least *started* some data transfer within an hour!

Probably worth mentioning that the server/backup in question is around 1Tb.

Could it be that it's taking a long time to create a thick virtual disk for some reason?

Any thoughts?

[HannesK]

Hello,

The vm is ready (including P2V updates) after about 30 mins which is perfectly reasonable.

I think that everything in your environment is very slow. Troubleshooting that via a forum sounds hopeless for me. 30min for instant recovery is slow. Everything I read looks like there are some performance issues outside Veeam.

To me it looks like "expected for the design". I recommend to talk to a partner / service provider that can help to improve the overall speed of the environment.

Best regards,
Hannes

[pkelly_sts]

Oh I don't dispute that the environment is slow - I'm having to work with what I have to do a PoC of sorts to get the client to spend on a "proper" solution. The P2V is what's taking the time in the first stage and I kind of expect it to take longer than just spinning up an instantly recovered VM anyway, even though I pointed the cache to SSD.

I've built a few high-performing Veeam infrastructures in the past so I know what a "proper" environment looks like & how to get it going, I just don't have the budget available to me at this stage so have to work with what I have available for now.