v6 Offsite Replication : Seeding How To

[nags_lucas]

Version 5 provided an option to do the initial replication by chosing the "Perform initial relication over this removable storage" and select the removable disk, but now the V6 did not provide such features.

In the V6 user guide, it shows: "If you replicate a VM to a remote DR site, you can perform replica seeding. Instead of having full replication traffic go over WAN, you can create a backup of a replicated VM, save it to a removable drive and move the drive to the remote site.", so any idea we save it to a removable drive in V6 ? Please help!

[Gostev]

Just copy VBK to the hard drive

[nags_lucas]

From where we copy the VBK? You mean i have to do a local replication first? Thanks

[Gostev]

Just read your original post again, and looks like I misunderstood your question.
It looks like what you are saying is that you cannot find the replica seeding options.

On the very first step of the replication wizard, did you tell us that you want to seed the replica?

[nags_lucas]

I intend to do replication offsite but the remote site do not have any replicated VM, so intend to have a replicated VM done it locally (V5 provide such capability to save in removable disk but V6?) and then i can physically move it to remote site and start the offsite replication.

[Gostev]

I understand that. You did not answer my question. Anyway, v6 does provide replica seeding options, and they are much better than what v5 had. Kindly please read the VMware User Guide, page 111 and onwards for the description of this feature.

[nags_lucas]

Thank you, now i understood what you said, that means the seeding is actually same like the v5's "Perform initial replication over this removable storage" . For permanent replication, is this seeding is one off thing and can be removed at production site after moved to offsite?

[Vitaliy S.]

Yes, the logic behind v6 seeding functionality is the same as we had it in v5. If you replicate a VM to a remote DR site, instead of having full replication traffic go over WAN, you can create a backup of a replicated VM, save it to a removable drive and move the drive to the remote site.

[hagepat]

Sorry for my misunderstanding. But do you mean that I have to create a backup to the removable disk. And then ship the disk to DR?
Do I have to copy data from USB disk to datastore on ESXi as well?

[Vitaliy S.]

Yes, you should ship you backup files as you did it with v5. Please take a look at the detailed explanation in our User Guide (page 43) for more info. Thanks!

[hagepat]

Vitaliy,

I read it. But I don't understand this sentince: "you can create a backup of a replicated VM". For my Firts initial replica to work do I have to setup backup job? Or do I first have to replicate local and copy files from repostitory to USB disk? And what do you mean by "Copy the backup to a DR site repository and perform repository rescan." With replica I don't have a repository, only a datastore on ESXi host. How can I solve this one. Thanks

[Vitaliy S.]

For my Firts initial replica to work do I have to setup backup job? Or do I first have to replicate local and copy files from repostitory to USB disk? And what do you mean by "Copy the backup to a DR site repository and perform repository rescan." With replica I don't have a repository, only a datastore on ESXi host. How can I solve this one

Here is the procedure you should follow:

1. Configure a backup job for the VM you want to perform initial seeding for.
2. Create and choose any local repository for the backup files.
3. Once you have backup files created, move them to a DR site and place them on the DR site repository (you should create it beforehand).
4. Rescan DR site repository to populate backup files in the backup console.
5. After that just create a replication job with "Low connection bandwidth" checkbox enabled.
6. On the "Seeding" step of the replication job wizard specify DR site repository where you've placed VM backup files.
7. Finish configuration of replication job wizard and start replication job.

Hope this helps!

[Gostev]

It would really help us tremendously if everyone read the User Guide before posting questions on forums

[hagepat]

Gostev,

You are absoluty wright, but the manual wasn't clear enough espacially to my specific situation. I've worked a lot with v5 and creating a backup of replicated machines is new to me. The procedure as Vitaliy described helped me a lot. Have a nice day

[nags_lucas]

Yes I do agreed the manual is not clear enough, hagepat is 100% right therefore instead of explaining in words, it is better to do procedure steps on user guide such as the above steps created by Vitaliy, which make things easy! If all the steps showing in videos, then it will be much beneficial.

[ddores]

[merged]

Hi - I'm feeling a little dumb here. Just playing with 6 for the first weekend.

I've gotten across the WAN replication working with compression, and the across the wire is working fine. It's definately faster than 5.

I can't get Seeding or Replica Mapping working, no matter what I try.

My scenario: taking a drive with Veeam 5 backups to my remote location, and trying to setup either Seeding or Replica mapping in Version 6.

When I attempt to seed, I put the backup in the remote repository, and I can even 'import' it just fine. Rescan the Repository. Attempt replication job. Fails every time (on multiple VMs) with "vm not found in backup for initial sync". Veeam has no problem finding the remote repository - it just won't 'find' the backup.

When I attempt to restore the guest into my remote ESXi, register it, and do replica mapping - it 'attempts', but sits there at 0% forever. Now, it doesn't 'detect' my replica on the remote ESXi (which i'm thinking is a danger sign), so I manually map it. In the logs, it gives an error like 'can't find vm' also.

Am I missing something? I've read the docs, and it seems like I'm doing it right. The only thing that I can't think of is that I'm working with Version 5 VBK files. But, the docs say that 'should' work for Version 6 replica mapping.

If anyone sees anything glaring that I'm missing, please comment.

Thanks,

[Gostev]

Hi, you have to use v6 VBK files for replica seeding.
"Detect" will only work if replica VM was earlier created by v6.
Thanks.

[tsightler]

You can use V5 backup files if you manually restore the VMs from the V5 backup files at your remote site. As Anton stated the "detect" option won't work because that requires V6 replicas, but you can manually map to these replicas after you have restored them. I think that is what you are attempting to do in the second part of your post and this should work (I've personally talked customers through this and seen it work).

A couple of things to check, if you are manually mapping replicas the "Get seed data from the following backup repository" should NOT be checked. This may seem obvious, but I've seen several customers make this mistake. Since you have already performed the restore manually you don't need the "seed" from a repository.

Also, on the "Job Settings" screen, make sure you have manually specified a proxy for the source and destination sides, and pick a repository that is on the source side to store the metadata.

[ddores]

Thanks for the tips! Worked great. Haven't got V6 VBKs to check the 'detect', but forcing the proxies to remote/local and the metadata force on the source side did the trick!!

Thanks again,

[mjwilde]

[merged

I am something of a newbie to both VMware and Veeam so please forgive me is this is a silly question.

My setup is ESXi 5 server at main site running a couple of VMs. Veeam v6 on separate physical server handling backups. ESXi 5 server on second site.

I want to set up a replication job for one of my VMs to replicate to an ESXi server on a second site. the backup file size is about 240GB so I obviously wanted to seed the replication to get it started.

In the prerequisites for using a seed with replication, it states:

1. Back up to a removable storage device VMs that you targeted for replication.
2. Move the device with the backup to a remote DR site.
3. In Veeam Backup & Replication, add a new backup repository and point it to the storage device.

I backed up to a removeable USB drive at the main site, but cannot see how to add this as a datastore at the DR site ESXi server. Having googled and looked on a couple of forums, there seems to be an issue with creating a datastore using a USB hard drive. Is this the case for ESXi v5? If so, what have others done in this situation? When it says 'removeable storage device', I'm sure most people would immediately think of USB hard drives.

I appreciate this is more of a VMware question, but I wanted to post here to see how people have got round this (if it is in fact a problem and not something stupid that I am doing).

Thanks in advance for your assistance!

Matthew

[foggy]

Matthew, you do not need to add your USB device as datastore to the host. Just create a new backup repository in Veeam B&R, specify the folder on this device as a repository path, and perform repository rescan. After that, when creating a replication job, select the 'Get seed from the following backup repository' check box and specify a path to the backed up VMs. Thanks.

[mjwilde]

Many thanks for pointing out my stupid mistake. I think I probably need some basic reading lessons!

I created the repository and set up a replication job to use a seed. However, 1.5 hours later, the job is still at 0%. Processed is 767MB, Read is 687MB and Transferred is at 270MB, with a processing rate of 137KB/s (which looks suspiciously like the upload speed of the remote network). How can I check that the job has actually used the seeding as this looks like it is trying to pull everything over the WAN link...

For information, the _replica VM was created at the remote site without what appeared to be any problems. In the job progress table, I have 2 entries:

CME-SBS [seed] 0%
CME-SBS Pending

so to me it looks like it is trying to use the seed but I can't believe it is as the transfer rate is so slow.

[tsightler]

Did you create a repository at the remote site? Did you select this remote repository when you selected the seeding option? If you picked your local repository it will simply attempt to seed from that which is probably not what you want.

[mjwilde]

Yes remote repository was set up and selected. However, I had left the proxies to automatically select instead of specifying them manually. Once I had done this, the job seems to have started fine and has a healthy (for USB) transfer rate of 10MB/s instead of the rather puny 150KB/s I was getting yesterday.

Hopefully I will have an up to date replica by the end of the day.

Thanks for your help guys.

[tabrez@ajmalksa.com]

Can you provide some additional input as my Veeam B&R 6 Server is VM. So how should i go about replicating this in first place. Kindly advice.

[Gostev]

The fact that Veeam B&R 6 Server is VM does not make any difference whatsoever. The process is still the same, as described on the previous page.

[tabrez@ajmalksa.com]

ok so i have already taken recent backup on my NAS box but in my remote site i have just installed vsphere 4.1 host and nothing else. Do i need to create physical machine for Veeam Backup server and start replicating from its console or is there any other option?

[Vitaliy S.]

Could you please tell me why don't you want to use your backup server on the main site to map and replicate itself using the instructions on the previous page? Installing additional backup server is not needed, as mapping across multiple backup servers is not supported. Thanks!

[tabrez@ajmalksa.com]

Is it possible if i get steps on how to go about it. As per my scenario it is as follows:

Main Site: 2 vSphere 4.1 Host with 6 production VM's and another vsphere 4.1 host for 3 VM's which include Veeam Backup Server also and i have NAS box acting as a secondary storage for backing up VM's
Remote Site: 1 vsphere 4.1 host only (NO datastore NO VM's)

I want to replicate my main site all 6 production VM's plus 3 other VM's including Veeam Backup Server. Kindly advice the steps if possible. Thanks

[Vitaliy S.]

Basically, if you have enough link bandwidth between your sites, then just configure replication jobs by adding all production VMs into it and then point these jobs to your remote site. Should more details on how to create replication jobs be required please take a look at our User Guide (page 100).

If you want to pre-seed these VMs, because of the poor connection link to the remote site, then use the procedure below to do that:

Here is the procedure you should follow:

1. Configure a backup job for the VM you want to perform initial seeding for.
2. Create and choose any local repository for the backup files.
3. Once you have backup files created, move them to a DR site and place them on the DR site repository (you should create it beforehand).
4. Scan DR site repository to populate backup files in the backup console.
5. After that just create a replication job with "Low connection bandwidth" checkbox enabled.
6. On the "Seeding" step of the replication job wizard specify DR site repository where you've placed VM backup files.
7. Finish configuration of replication job wizard and start replication job.

Hope this helps!