Veeam different subnet

[jamerson]

Dear all.

I am trying to collect some information about my issue before opening a ticket.
Before we had veeam and esxi/vcenter st the same subnet.
Because of a busy network we decieddd to separate the backup traffic so we created a se-erste vlan for the esxi /vcenter.
I don’t know if this the cause but the backup is extremely slow 15MB/s.

The veeam has one nic and is a virtual machine I have tried the transport mode as appliance and automatic but non has helpend.Do I have to add second nic for the veeam to be at the same vlan as the vcenter ?

I hope someone can help me understand the cause
.

[Mildur]

What is the bottleneck in your jobs,
See the bottleneck in the job status and start there to troubleshoot.

veeam-backup-replication-f2/read-this-f ... tml#p95291

[jamerson]

Thank you for your answer.

the bottleneck shows source,
ive tried all of them but the speed is poor.

[jamerson]

now it chose proxy but speed still 16MB/S.
cannot seems to find the cause still. Downgrad / new vCenter / New VEEAM server.

non speed improvement at al.

[Mildur]

It‘s best to get veeam support involved.
If the speed was good before the upgrade, then there could be a bug.

Do you have updated to vsphere 7.0 U3?
Be aware, that it is not officially supported at the moment.

[jamerson]

thank you for your answer.
actually two things happens at the same time for this customer.
the vsphere is updated to the B release not the 3E and also the ESXI has been moved to different subnet.
i think the firewall rules are causing those slownless as those two are at diferent subnet.
i will move them back to the same subnet and see what happens.
we will get the support involved.

[jamerson]

We managed to fix the issue with vMware, its appear the network ( firewall was blocking packages on the it side).
VEEAM and vCenter /ESXI were at a different subnets. even we had firewall rules any/any somehow the packages were still being bloked.
one issue we cannot seems to delete the CDP job, its greyed out and still showing removing.

https://ibb.co/GRW744K
https://ibb.co/xDx0XTS

[rennerstefan]

Good to hear you found the issue. I was just about to write that usually routing and/or FW are the issues in such configurations.

Did the CDP job stop or is it still running? If so feel free to open a ticket at our support.

Thanks

[jamerson]

We managed to fix the issue with CDP too.
my latest question, when you have the VEEAM server and ESXI servers at separate network.
opening the ports on both interfaes seems not to be enought for this matter.
what else need to be done?

thank you for your answers

[veremin]

Might be a stupid question, but are all ports opened in accordance to this guide? Thanks!

[jamerson]

veremin thank you for your answer.
those ports are open between the two subnet.
we are using those as alias on the firewall. the speed comes back only when we use any to any rules
the ports we were using :

Code: Select all

Backup Server Connections
Veeam Backup & Replication Console Connections
Microsoft Windows Server Connections
Linux Server Connections
Backup Proxy Connections
CDP Connections
Backup Repository Connections
NFS Repository Connections

[rennerstefan]

Hi again,
not sure if your question is about CDP only now or more in general!?
I would, in addition to what @veremin has already shared, recommend to use the following tool: https://www.veeambp.com/ports/index.html
In the tool you can add every component you are using and will get a clean output on what has to be opended from A to B.
With that you should get everything you need specific for your environment.
Regards

[jamerson]

We have fixed the issue,
one was related to the vcenter and second one was related to Firewall blocking packages.
for the time beeing the vlan is allowing everything to the LAN.