replication jobs taking too long

[markwragge]

Hi, I am evalutaing veeam replication at the moment as a replacement for our current backup product which is replicated offsite for DR. I have setup a replication job containing 4 Vms and they are replicated to an ESX server at the DR site. The link between sites is a 100megabit link. The first replication job has completed to the DR site.

I set the repliction job to update once per hour but the replication appears to take longer than that. The statistics for the replication job state that there is 466gb per replication job. THis is the total size of the 4 VMs.

How does the replication job determine what changes are required for each hourly job. I expect that only the changes to each server would be replicated. How does replication determine what has to be replicated?

For example; the replication job for the exchagne server is taking 40 minutes for 70gb every time it runs. The first replication took 60 minutes. There are no significant changes on this server to justify this length of updates.
Another serrver which has minimal changes is 100gb and took 50 minutes for initial replication and now takes 20 minutes.

Thanks for any advice.

[Vitaliy S.]

Hello Mark,

Could you tell me what replication job mode are you using? Have you specified service console credentials for the remote ESX host?

Here is a good explanation of how CBT works, please look through it for more information:
http://www.veeam.com/forums/viewtopic.p ... 717#p11171

Thanks!

[markwragge]

Hi, I have created a replication job. I am using Vmware vstorage API mode (san only). The data is being replicated across the network to the esx server at the dr site.
The DR esx server is in the same virtual center server as the source so i have not been ask to enter any credentials for login when creating the replication job.

Thanks

[Vitaliy S.]

Mark,

Please specify service console credentials for your target ESX host using Veeam console, you can configure this in the connection settings of ESX host's properties. Using service console agents (which provides network traffic compression, and other benefits), should give you better performance for the replication job.

Thanks!