v6 Offsite Replication : Seeding How To

[MoifMurphy]

[merged]

Hello,

We've researched this as best we can and have read various blogs but we are none the wiser as to how this works. Could someone please explain the ins and outs to us?

This is our scenario. We have vCenter 5 on our main site and on our DR site. The two sites are on a 100mb WAN on the same IP range. We currently run in excess of 40 servers with 5 being at least 300GB. Even with proxies in place and the concurrent jobs set to more than we are comfortable with we are finding that the replication is taking far too long. In some cases we are looking at over 20 hours. Obviously this is causing a bottleneck and we aren't happy with what we are seeing.

We've looked at Hyper IP but that won't help as we're on the same IP range. So we're trying to find some information regarding seeding. The instructions state that we should create local replication, take the replicas down to the DR site and then configure seeding. My question is if we've already created an initial replica on the DR site over the WAN then can we still use seeding?

And what exactly is seeding? Is it differential changes only? Block level?

[foggy]

Seeding is the initial replication without having to transfer all of the VM data across the sites. It is described in detail in the user guide and the step-by-step procedure is given in this topic above. If you already have the replica VM in DR, you need not to perform seeding anymore as only changes are transferred during each consequent replication job run (block level). Thanks.

[slidek9]

I have a replication seeding issue here;

I've used a NAS box to seed my replicas at the DR site, the replicas have been chugging along (seeding completed weeks ago but NAS just still connected) now for 2 weeks and *boom* the NAS box dies.

The only data on the NAS was the initial seed data - the replica VM's and their meta data are stored elsewhere - however, none of the jobs now run as they are looking for the NAS box!

I cannot remote the NAS box from the repository as it says that the NAS is still in use by the jobs - so I removed the 'Seeding' step fro the jobs (and double checked repositry locations) to make sure the NAS is not being used in any way, and I still cannot run the replication jobs.

Support has suggested that I create a new job and 'Map' my replicas to that, well I have ran that in test and it seems that I lose all of my snapshots as the job 'reinitialise'; this is not really good as I don't want to wait on 1/2TB of snaps to consolidate manually.

So the (long drawn out) question: Why is the seeding repository still needed even after seeding is complete - is there any way I can fix my current situation gracefully?

Please help!

[foggy]

Leon, try to re-enable the 'Seeding' step and remove the reference to the NAS on it.

[slidek9]

Yes! That has fixed it! Thanks Alex Could you raise a feature request, 'when Low connection bandwidth (enable replica seeding) is de-selected, seeding repositorys are automatially removed from the job'. To clear up any confusion for others that experience the same issue?

[foggy]

Not actually removed but just not required to be up and running. This is already on the list, thanks.

[undecided]

[merged]

Hi all, I'm new to Veeam and need some help, I have a primary site and DR site connected with a 10mb/s link. I want to begin replication, and as I understand it I need to make a backup of my primary site VM's. I have now done so and have copied the backup file to my DR repository, imported the backup and can see the backup job and the backed up VM's in Veeam. Now I select to create a replication job, point it to the repository where the backup file is contained(I've also tried the local Veeam servers repository) and the job fails with "vm not found in backup for initial sync".

My question is, do I need to actually restore the VM files to the repository?

thanks

[undecided]

In my case the backup is not of a replicated VM, it's a backup of the parent VM. Is there a way to use this for initial seeding? Or do I have to go back to the primary location and create a replica, then backup this replica and bring it to my DR site? Is there no way I can use the backup of the parent VM for this purpose?

[foggy]

Please read through this topic carefully, it provides the step by step seeding procedure. You need to use the source ("parent") VM backup for that.

[undecided]

Ok, thanks Foggy, I see it now, I have to restore the VM's and then manually map to the VM as if it was a replica. I'm trying this now and will revert if it does not work.

[foggy]

Actually, you do not even need to restore the VM, just use its backup for seeding.

[undecided]

I'm failing miserably here, I've tried as suggested, just to be clear my backup is of the parent VM, it's not a backup of the replica at the primary site. When I follow the steps outlined in your link I get "VM not found in backup for initial seed" even though I've scanned the repository, it finds the backup file, i can see the backup file in Veeam. Then I tried restoring the VM files to the datastore and mapped the replication to the VM, this fails with "A snapshot operation cannot be performed.
Error: A snapshot operation cannot be performed."

I'm struggling here, any idea?

[Gostev]

Suggest you open a support case and have them guide you through the process over webex.

[undecided]

I've manually mapped to the replica after I've restored it. "Get seed data from the following backup repository" is NOT checked. Also, on the "Job Settings" screen, I've manually specified a proxy for the source and destination sides.

I don't believe I'm doing anything wrong here, besides opening a case and waiting for what may be potentially days, does anyone else have an idea that I may try?

[Gostev]

Can you snapshot the replica VM with vSphere Client?