Replication between 2host is very slowly.... 10mb!!

[real.momix]

Hello everyone, sorry in advance for my English ..
we have an infrastructure vSphere 5.5 and veeam br 8update2.
here's how it is composed:
- 1 hp proliant DL360G7 with local storage 4xhd sas 7200 raid10, 1 cpu 4core + 32gb ram, 512mb cache controller, 4 nic card

- 1 hp proliant dl380g7 with local storage 4xhd sas 7200 raid10, 1 cpu 4core + 32gb ram, 512mb cache controller, 4 nic card

we have a virtual proxy server on each host with 4GB of ram and 2vcpu.
do not we start job competitors, but only one at a time.

the network is divided into a gigabit vlan:
vlan01 = = management esxi 2nic card etherchannel (LACP static) per host
vlan10 = vm is the network with all the vm to replicate = 2nic card etherchannel (LACP static) per host

vmware license: 5.5 essential kit

the problem is only slowly during the replicas because the backup job is done on a nas qnap 259pro with 2 hard drives 2tb and speed is about 30mb (I'm not complaining)

there is something that I did wrong in the configuration? or is the speed that must have this infrastructure ??

I thank everyone for the help.

[Gostev]

What is the bottleneck stats for the replication job?

[real.momix]

Gostev hello , thank you for the answer , the bottleneck is indicated TARGET ...
Also during the processing it comes indicated : No available proxies are running on ESX ( i) management interface subnet . Using proxies from a different subnet , performance may be impacted .
"
This message is normal ? 2 proxy that I have on vlan10 192.168.10.0/24 , while network management is on 192.168.1.0/24 vla01 .. I was right to separate the two networks ?
so sorry but I'm not an expert ...

[Gostev]

OK. To resolve your performance issue, you should specify a hot add capable proxy (basically, any VM running on the target ESXi) as the target proxy in the replication job settings. Thanks!

[real.momix]

Hello Gostev , I leave the selection of proxy automatically ... it is wrong ? I have to manually force the proxy ?

[Shestakov]

You need to grant a proxy role to one of your VMs on the target host.Once you add a new proxy to your VBR console, during the job run it will be picked automatically. You also can choose it explicitly on the Data Transfer Settings step.
Thanks!

[real.momix]

Thanks, I had already configured 1 proxy vm on each host dedicated to datamover leaving automatics selection ..ma performance is around 14mb !! any idea??

[PTide]

Hi momix,

Please check if picking the proxy manually will help.

Thank you.

[Shestakov]

Momo, please check if the created proxy was chosen for the replication job. It can be seen in the job`s "Action" list for replicaped VM. It should be:
"Using target proxy Proxy name for disk Hard disk 1 [hotadd]"
If the proxy was chosen correctly, there is no need to set it explicitly. Please provide your new full bottleneck statistics. Thanks!

[real.momix]

thank you so much guys for your precious help, the proxy is chosen correctly in an automatic way.. I have attached the photo with the state of the replicas . I note that in a replica FULL speed is very bad , while during an incremental replication speed is around 30MB / s .. look your advice .. and thanks again .

[Gostev]

Yep, all looks good. From the bottleneck statistics, it does look like the issue here is your target host's storage speed. If you don't agree with this conclusion, please open a support case for further troubleshooting, and include case ID in this thread. Thanks!

[real.momix]

thanks thanks thanks

[Vitaliy S.]

Out of curiosity - what bottleneck stats and job performance do you have if you run your replication job and force target proxy server to use NBD mode?

[real.momix]

I have not tried , but it just got a bit of time I try and let you know .. thanks