Extremely Slow replication jobs - Network Bottleneck

[marvologaunt]

Hello, I have had Veeam case 01795117 open for quite a while with issues with my replication jobs going extremely slow for the last two months. The weird part is - they don't always go slow, and it's not always the same VMs that go slow. The same VM could run at 15 MB/s replication one day, then 250KB/s the next, then vice versa. It's hard to pinpoint the real cause due to this.

The following were what I thought to be potential causes of the issue:

Upgraded to Veeam v9
Added 10GB networking to ESXi hosts

I've done the following to try to resolve the issue:

Created a new physical target proxy server w/ 64 bit Server 2012 (per Veeam's instructions)
Reverted to Veeam v8
Tried a different source proxy server by using a VM instead of a physical (have since reverted back to the physical box due to no change)

Backup jobs run fine. Replications are always the issue with bottlenecks at: Source 0% > Proxy 1% > Network 99% > Target 0%

Obviously there must be some kind of routing issue between the local site and the replication site (which are connected via an ipsec tunnel). Prior to about two months ago, the jobs ran great. Only network change was the add of 10GB networking, though we still use the 1GB networking as well.

Any advice on this would be amazing, as I haven't found anyone else with jobs this slow, and right now the issue is with Veeam R&D, and we really need our replications working properly.

Thanks!

Ben

[Dima P.]

Hi Ben,

I see that T3 engineer was assigned to your case, let await for the investigation results.