Backup to B&R - Seed backup copy job

[frankive]

If we backup a workstation to a local B6R server and then use backup copy job to a cloud connect provider. Can we seed this backup like we usually do with other backup copy jobs?

[Dima P.]

Frank,

There should be no limitations for regular Veeam B&R behavior

[danielchung]

I'm testing the same thing that VEB->VBR repository->Backup copy to offsite. It works by using Backup Copy to copy backup offsite, but there is no "map backup" function for an Endpoint Backup Copy job that allows putting the VBK file as seed.

Looks like it is a missing function or any possible workaround can be provided?

[Mike Resseler]

Daniel,

I'm afraid I don't understand your question completely. If you take a backup with VEB to a VBR repository, and then use that backup (from the repository) to see it offsite, that should work? Or am I misunderstanding you here...

Mike

[danielchung]

I have two repositories, one at headquarter and one in local office. In local office, we'd use VEB to backup endpoints' to the local repository. This simple VEB-> VBR works. Further, we'd like to use backup copy to offsite the backup from local repository to the repository at headquarter.
To avoid heavy load on WAN link for copying the first VBK file, we want to manually transfer the VBK file from local to HQ and make it as seed for Endpoint backup copy job. But during the setup, I don't see there is "map backup" function similar that we used to do in VMware backup copy job.

[Mike Resseler]

Hi Daniel,

Could you follow the procedure in this KB: http://www.veeam.com/kb1856 and see if it works?

As far as I know that should do the trick but I can be mistaken. If not, I will follow-up further with the teams

Thanks

Mike

[danielchung]

Thanks Mike.

Have read that KB and actually I've been using the backup copy job for VMware for a while. However, when creating backup copy job for Endpoint, in the target tab, there is no "Map Backup" option allowing to go further to select seed.

Seems like seeding is missing in VEB-VBR integration.

[Mike Resseler]

Daniel,

I can't test this in my setup so I will try to contact the teams and see what they can tell

Thanks

Mike

[HMwkr]

I have just hit exactly the same issue. Map Backup does not appear for a Backup Copy Job for an Endpoint Backup.

Which, as the backup is about a terabyte, and is on a repository at the other end of a WAN link which runs at just 5Mbps, is a bit of a bummer !

Case #00917915 refers.

[Mike Resseler]

Gentleman,

I received word that this is not supported by design

Sorry

Mike

[HMwkr]

What?!

Good grief.

Well that's thrown a major roadblock in front of all of my plans for reorganising our backup resources over the next month.

Not happy.

[HMwkr]

Veeam's response: " it may be best to run the backup copy to another local repository if at all possible"

My response: This renders offsite backups of endpoints effectively unworkable, with endpoint integration. I'll have to revert to backing up endpoints to a separate Windows share, then backing that up to a local repository, then running a copy job to a remote repository. So all the time and effort I have expended in setting up endpoint integration has been *completely* wasted !

Have I missed something? How are the rest of you making offsite backups of your endpoints?

[Mike Resseler]

Hey,

One question, do you have Enterprise plus? In that case I would advise you to use the WAN acceleration and use the populate cache functionality
This will be very effective for your bandwidth

Cheers,

Mike

[HMwkr]

Thanks. Yes, I am already using WAN acceleration, and the cache is populated.

It would still take several days to backup to a new remote repository location, during which time our users would be severely inconvenienced. And I can't even seed a new location. So I am stuck with this endpoint backup on a specific repository.

[Mike Resseler]

I think I start to believe I am misunderstanding you (if so, I deeply apologize and blame jetlag for it )

What do you exactly try to achieve? You take a backup of your endpoint, store it to a local repository, and then do a backup copy job to a secondary repository on a remote location... Is that correct? Or do you want to do something else?

Let me know (and again sorry if I am confused here...)

Mike

[HMwkr]

You understand correctly. Your jetlag is causing self-doubt, not confusion

[Mike Resseler]

Lol

In that case, your users won't suffer from it. They will do their day-to-day backups to that "local" repository and the backup copy job will not interfere with that, at all. Now it is difficult to guess what your ratio will be, but I ask you to don't believe me and test out for yourself (maybe with your own PC) and do this (make sure you use the populate cache feature!). The results will be fantastic (did it myself and got 36x which is pretty cool in my book). And the results will only get better when there are more PCs (think about it, a lot of your PC backups will be exactly the same...).

I know it isn't seeding, but you will be surprised by the results. But again, don't believe me... See for yourself

Mike

[HMwkr]

Um, that's not the problem.

Endpoint backups run to local repository X at night. All good. Backup copy job runs later at night copying to remote repository Y. I need to reorganise remote repositories, so I moved the remote backup files to repository Z then found I couldn't map the backup copy job to that repository Z. So The next time the backup copy job runs it will copy the entire backup across the WAN, which even with acceleration and caching, will hammer our internet connection so users will be severely impacted for days.

I could run the first backup copy job to repository Z at a weekend - but it wouldn't fit into one weekend, and the other main data backups run at the weekend also soaking the WAN.

Is that clearer?

[Mike Resseler]

OK,

Now I understand it completely. In that case I have not much more to add... (besides sorry and it is a v1 product, standalone, but we take all that feedback into account...)

Mike

[HMwkr]

And Yes, I should have read the Release Notes

[Mike Resseler]

Yes you should (not that you are the only one that hasn't read them )

Actually that is something I wished more people would do... Our R&D team puts a massive amount of time and effort in that and this is one of the reasons why working for Veeam was so attractive for me. Documentation is not seen as a waste of R&D time but seen as something very important. Funny enough we almost never get feedback on that

[fafa24]

I just came to this thread from a google search I have exactly the same problem. I like to seed a backup copy job from Endpoint and realized there is no mapping feature. I have other backup copy job to seed from VMware and they work just fine. I hope for Endpoint backup copy job mapping a seed will be a feature in version 9.

[Dima P.]

I can’t say “yes it will” but I am allowed to say “yes it might”

[Kuhnster]

Please ADD this functionality. We are a Veeam Cloud Connect partner and there are times when our clients have physical servers to back up. I have a 7.6TB server that I'm backing up with Endpoint, but I can't seed it. Please consider this in the near future. Thank you!

[mr_iang]

I found a solution which has worked for this. Here's a brief outline.

Get your backups running from Veeam Endpoint to the client's on-premise B&R server.
Create a copy of the backup files -ie your seed, on to a portable medium.
Back up the B&R server configuration to the poratble medium.

Install a B&R server on a network local (or with fast connectivity) to your CloudConnect server.
Restore the configuration from your client's B&R server, along with the seed files on to this temporary server.
Set up a backup copy job to your CloudConnect server, let that run.
Once the first job completes, create a configuration of your temporary B&R server.

Send that config backup to your client's on-premise server, restore it over the existing configuration.
Voila, run a backup, it will copy to your CC server keeping the seeded data as your base iamge.

Obviously this isn't a detailed how-to, but I can tell you that this solution works. We seeded a couple TB of data from a client on a connection that would have required weeks to seed, in the matter of 2 days.

Iang

[bolish]

Just to chip IN. It's 2018 and I am just wondering if this feature has been added to Veeam Agent for Windows backup. I mean ability to take the Local backup of a Physical server with Veeam Agent, copy it to a physical location and transfer it to the cloud repository, so it only take incremental backups rather than the full backup.

[Dima P.]

ability to take the Local backup of a Physical server with Veeam Agent, copy it to a physical location and transfer it to the cloud repository, so it only take incremental backups rather than the full backup

Yes, agent job mapping is supported. Kindly take a look at this KB article. Thanks!