Understanding VirtualLab

[RumataRus]

Hi, colleagues!

As we can see at p. 106 of "Veeam Backup & Replication 5.0 User Guide October 2010" there are two productive subnets: 192.168.1.0/24 and 172.16.0.0/16.
Is it correct that:
1. Proxy Appliance must have IP-address in the same subnet as Veeam Backup Server?
2. 192.168.1.7 and 172.16.10.20 are IP-addresses of default gateways in corresponding subnets 192.168.1.0/24 and 172.16.0.0/16?
3. Veeam Backup Server should not be in the same subnet as productive VMs?

[Vitaliy S.]

Hi Dmitry,

1. Not really, as you may have different subnets, but Proxy Appliance should be able to reach Veeam Backup Server in anyway.

2. These addresses are IP-addresses of a Proxy Appliance in the isolated network. They are acting as gateways for isolated networks and typically are the same as in the corresponding production networks.

3. You can have Veeam Backup server and production VMs in the same subnet.

In addition, you can also check out our blog post covering some of the basic configuration question you may have further:
Five important points of Virtual Lab networking in Veeam Backup and Replication v5

Hope it helps!

[RumataRus]

What kind of address is assigned on tab "Proxy" in section "Production Network Connection"?
Proxy Appliance has two addresses 192.168.1.7 and 172.16.10.20 already. Is this the third one? For what?
What subnet this address must belong to?
If I have Veeam Backup Server in subnet 192.168.1.0/24 and Production VMs in subnet 172.16.0.0/16 what address I should assign to Proxy Appliance in section "Production Network Connection"?

[Vitaliy S.]

What kind of address is assigned on tab "Proxy" in section "Production Network Connection"?

In the Production network connection section you need to select a network where the proxy appliance should be created, specify its IP address and settings of DNS server to be used.

Proxy Appliance has two addresses 192.168.1.7 and 172.16.10.20 already. Is this the third one? For what?

The example we have in User Guide shows situation when a tested VM and its dependencies are located in two logical networks, these two networks are recreated in the virtual lab and mapped to corresponding production networks.

What subnet this address must belong to?

This subnet should be similar to your production subnet.

If I have Veeam Backup Server in subnet 192.168.1.0/24 and Production VMs in subnet 172.16.0.0/16 what address I should assign to Proxy Appliance in section "Production Network Connection"?

You should assign any IP address from your production network (172.16.0.0/16). To get more information you can continue reading our User Guide on page 111, there is a fairly good description of this wizard.

[RumataRus]

Hm...
I have Veeam Backup Server and production VMs in different subnets.
When I assign to Proxy Appliance in section "Production Network Connection" address from production network then there is no route from Proxy Appliance to Veeam Backup Server.
As a result Proxy Appliance does not start.
When I assign to Proxy Appliance address from network where Veeam Backup Server is then Proxy Appliance starts but Ping test failed. Is that correct or I should ask support?

[Alexey D.]

Dmitry,

Can you ping any VM in production subnet from any VM on 'B&R' subnet?

BTW, for generic verification purposes you may not want to put appliance to actual production network. The main principle is that appliance should be reachable from B&R installation, so scenario 2 (from your post) is fine. The only thing, when are you seeing this error? While VMs are being powered on? What is the exact message?

[RumataRus]

Dmitry,
Can you ping any VM in production subnet from any VM on 'B&R' subnet?

I cannot.
The management network (where vCenter, Veeam Backup etc are) is separated from the production network.
There are no persistent routes between them by design.
There is access from production network to management network from restricted list of PCs only.

Dmitry,
The only thing, when are you seeing this error? While VMs are being powered on? What is the exact message?

There is fragment of the protocol:
07.02.2011 17:46:17 TestVM2: Powering on (IP address: 192.168.254.205, network: 192.168.254.0, mask 255.255.255.0, gateway: 192.168.10.222)
07.02.2011 17:46:19 TestVM2: Heartbeat test
07.02.2011 17:49:59 Fail TestVM2: Pinging (Ping failed)
07.02.2011 17:46:24 Fail TestVM2: Pinging (IP address: 192.168.5.205, result: 'TimedOut')
...
07.02.2011 17:49:54 Fail TestVM2: Pinging (IP address: 192.168.5.205, result: 'TimedOut')
07.02.2011 17:49:59 Error TestVM2: Application initialization (No test scripts defined)
07.02.2011 17:50:04 TestVM2: Powering off

It is obviously to me that address 192.168.5.205 cannot be pinged from subnet 192.168.10.0 because there is no routes between them.
But I thought that Veeam can do "Ping Test" with proprietary means.
If I create persistent route on Veeam Backup Server to subnet 192.168.5.0 is it will resolve my "Ping"-problem?

[Alexey D.]

So, if no routes, it's not an option for you to have appliance and B&R sitting on different non-interroutable segments.
You may want to use the approach I described in the previous post.

To access VMs in isolated network, B&R adds entry to routing table. Most probably, the error you are getting has a different cause. Please contact our support team for investigation. Thanks!

[RumataRus]

So, if no routes, it's not an option for you to have appliance and B&R sitting on different non-interroutable segments.

I do agree.

You may want to use the approach I described in previous post.

I see... I still think about it.

Most probably, the error you are getting has different cause. Please contact our support team for investigation.

I've already done it.

Vitaliy, Alexey, many thanks to you!
Best regards, Dmitry.

[RumataRus]

Hello, Alexey!
Ping failed because Inbound Rule "File and Printer Sharing (Echo Request - ICMPv4-In)" in Windows Firewall on the tested VM (Windows Server 2008 R2) was disabled.
Regards, Dmitry.

[Alexey D.]

Ah... I see. Thanks for sharing this!