MAP REPLICAS or SEEDING OVER WAN NOT WORKING

[cpitserve]

I am trying to replicate across WAN, 2 virtual machines. I have first performed a local backup on external hardisk and then attached the external hard disk on my WAN location on the target esxi host, and then imported the backup on the veeam that is running on the Virtual machine at the remote site. Then I create a new replication job and configured to “Get the seed from the backup repository”

However the jobs seem to start running without any seeding from the backup repository. It is starting the replication from the very beginning without noticing the already imported backed up vms from the repository.

I tried many times to recreate the jobs, read the forums, guide manuals but I don’t seem to manage to get to replicate only the changes from the backup repository.

I have restored the backups to the esxi target host and then created a new replication with low connection bandwidth and mapped the replicas with the existing vms which i have restored from backup.

Still the same issue. The jobs start backing up from the very beggining without any indication that is only doing a backup of the changes.

Any helpful hints would be much appreciated

Thanks
George

[Vitaliy S.]

George,

Can you please clarify how do did you determine that incremental pass is not running and the replication job began from scratch?

Thanks!

[cpitserve]

Hello,

For example i am now replicating one VM which is 300GB. Is being doing replication for 18:27 hours and has only replicated 33GB with 600KB average internet speed. Before i do that i have clicked on detect replicas and found an existing replication successfully for mapping.(that is the one i have restored from the backup from the external hdd). This map replica is actually 10 days old but is it normal to take so much time?

Thanks

[foggy]

This map replica is actually 10 days old but is it normal to take so much time?

Depends on the amount of changes generated inside the original VM during that period... Do you have proxy servers on both sides within the remote Veeam B&R instance that is responsible for this replication job?

[cpitserve]

Actually i had the same problem even from the first day i had restored the backups from the external hdd. Yes i have proxy servers on both sites and and i use automatic selection for both source and target.

Is there anything under the logs folder or the action's window that will indicate to me that the replication is actually reading from the map replicas? At least this will let me know if i am doing something wrong.
For example when initiating a new job, it takes about 3 hours to calculate the digest hard disk. Does that means that is calculating the changes from the source replica and the mapped replica or is it something completely different?

[foggy]

Yes i have proxy servers on both sites and and i use automatic selection for both source and target.

Make sure the correct proxies are selected for processing (this info is available in the job session window).

For example when initiating a new job, it takes about 3 hours to calculate the digest hard disk. Does that means that is calculating the changes from the source replica and the mapped replica or is it something completely different?

Yes,this is exactly what tells you that replication mapping is being performed.

[Vitaliy S.]

BTW, what is the connection link between these two sites?

[cpitserve]

The connection is 6Mbits/6mbits down/upload from the the source side and 8Mbits/1Mbits download/upload from the target side.

So i can assume that everything is correctly configured. I will then wait for the replication to finish as many hours or days as it takes and then perform an immediate replication to compare the timing difference since its going to replicate less changes.

Thanks

[cpitserve]

BTW how do i suppose to know which are the correct proxies? Do you mean that i shouldnt use the automatic selection?

For example i have a file server and an exchange server at the source site. I have created 2 backup proxies, one on the file server and one on the exchange and there is the veeam backup proxy on the local windows vm that veeam is installed on the target site.

I have defined 2 replication jobs, one for the file server and one for the exchange server. What should be the correct selection for the proxy servers for each case? For example should i define as source proxy the file server and the target the veeam backup proxy if i want to replicate the file server only?

Thanks

[foggy]

What I mean here is that a pair of proxy servers residing at different locations should be selected for optimal data transfer performance between them. Even with automatic proxy selection, sometimes selection criteria can point to the proxy on the wrong side, depending on the specific environment (particularly, its network topology).

[cpitserve]

Ok i will wait for the replication to finish and then see how it goes

Thanks

[Vitaliy S.]

BTW how do i suppose to know which are the correct proxies? Do you mean that i shouldnt use the automatic selection?

If automatic selection does not work for some reason, you can explicitly select which proxy servers to use.

The connection is 6Mbits/6mbits down/upload from the the source side and 8Mbits/1Mbits download/upload from the target side.

Thanks! This is explains why digest calculation takes that much time.