- Posts: 1
- Liked: never
- Joined: Dec 05, 2017 4:18 pm
- Full Name: Alberto Moriconi
I already read a few threads about similar problem but I cannot manage to solve my issues.
We have 2 sites: site A (main site) and site B (DR).
A Veeam B&R server has been deployed both at site A and B.
We do have local VM Backups at main site A.
Then we do Backup Copy to a remote repository at site B (DR).
Finally we do replicate VM at the remote site B using remote site B repository as source.
The replication job is handled by Veeam B&R at site A.
This last step does not seems very efficient. All data will be moved at site B being orchestrated from site A.
This way the replication job requires nearly the same time as the backup copy which seems a bit odd because data has already been trasferred on the slow WAN link.
Thus we would like to handle this using Veeam B&R server at site B.
So we defined a replication job at site B to replicate a VM from site A to site B using as source the site B repository.
The result is always "No restore points found".
In the log, the only interesting line is:
"Physical host volume was not found for path ... on host ..."
We did rescan the repository and of course the data is there with all restore points.
I can even instant recovery the VMs. But this seems invisible to local jobs.
So I am wondering:
- Is the replication job at site B able to access backup copy data, in its own repository, coming form site A jobs ?
- Do I need a a dummy backup job and to script something (I guess I need some pointer to further documentation in this case) ?
- What am I doing wrong ?
- Veeam Software
- Posts: 4320
- Liked: 355 times
- Joined: May 19, 2015 1:46 pm
- Posts: 153
- Liked: 23 times
- Joined: May 21, 2013 9:08 pm
- Full Name: Alan Wells
Here is what I ended up with and this works great for me. I am very happy with the performance of all of my jobs now.
I have an HP Apollo Storage Server in Prod. I have a DL380 Server with lots of external disk array storage in DR. Both are FC attached back to their respective SANs to pull backups over fiber.
I have a single VM in Prod that acts as my Enterprise Manager. It is replicated to DR.
I have a single VM in DR that acts as my job server. It uses the 2 physical servers as proxies. It has all of my repositories setup, all the backup, replication and backup copy jobs setup. I replicate this VM back to prod just in case.
First I created a new VM in DR which will be my jobs server. I chose DR since that is where we would be doing things in the event of a disaster so it cuts a step of failing over my backup VM in that case.
I restored the DB from my original Prod server to that new VM since the vast majority of my jobs were in my old server.
I then recreated all of my replication jobs on that new server. I chained the replication jobs to run after the backup jobs complete and I set the replication sources as the production backup repositories. That cuts down on snapshots on prod VMs. I then chain a second backup job off the replication job which does a backup of the replica in DR. Yea that is overkill but it is required by our owner so that is what we do. Then my tape jobs are set to run after backups complete each night. It works great.
In my original configuration I had backup jobs in DR running on the physical server in DR so I have to re-catalog all of those tapes back into the new primary VM so that they are all in a central location. I am actually still working on that but for now the software is still installed on that old server so if I need to find tapes in DR I can always reference that old setup there.
Users browsing this forum: No registered users and 31 guests