I'm sure it's the way I'm presenting the questions and information, but my case isn't getting me answers.
First question, does Veeam offer a product or service that would allow client-end desktop traffic reach the VBR servers without giving the clients direct open ports to the VBR servers? We are wanting to explore backing up sensitive desktop computers of some employees and reading over the documentation it seems like those desktop agents require a lot of firewall ports to be opened directly to various VBR servers. I understand that, but I would not like to do that. Is there a proxy role or another Veeam product that would allow us to funnel such traffic through it as a middle-man between users and VBR?
Regarding the firewalld zones question, we have an RHEL9 server running both the proxy and repository roles. Since it's RHEL9, I'm trying to work within firewalld zones. I've got a "veeam" zone that I've added the source IPs of the VBR server, other proxies and repos, as well as the SQL servers for transaction logs). Per https://helpcenter.veeam.com/docs/backu ... ml?ver=120 and https://community.veeam.com/script-libr ... zones-5064 I've specified "FirewalldZones=veeam" in the following files:
- /opt/veeam/transport/VeeamTransportConfig
- /etc/VeeamNetConfig
Code: Select all
# grep -B12 FirewalldZones /var/log/VeeamBackup/VeeamEnvironmentSvc.log
[12.09.2023 22:45:48.318] <139687746580928> tpl | Loading transport options.
[12.09.2023 22:45:48.318] <139687746580928> | /opt/veeam/transport/VeeamTransportConfig was found.
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [AgentOrphanTimeoutSec] : [1200]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [AgentHangTimeoutSec] : [3600]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [AgentShutdownTimeoutSec] : [60]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [FipsMode] : [true]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [MaxLogSize] : [15728640]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [MaxLogCount] : [10]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [ServicePort] : [6162]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [ServiceUser] : [root]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [BaseLogDirectory] : [/var/log/VeeamBackup]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [IptablesAutoupdate] : [false]
[12.09.2023 22:45:48.318] <139687746580928> tpl | Non-default option [FirewalldZones] : [veeam]
I'm trying to understand the behavior, because my expectation is that specifying the "veeam" zone in the files above it should limit where Veeam opens ports.
Here's the "veeam" zone without the IPs (6160/tcp 6162/tcp 2501/tcp 2504/tcp were added dynamically by the Veeam services):
Code: Select all
veeam (active)
target: DROP
icmp-block-inversion: yes
interfaces:
sources: [removed 5 IPs]
services: ssh
ports: 6160/tcp 6162/tcp 2501/tcp 2504/tcp
protocols:
forward: no
masquerade: no
forward-ports:
source-ports:
icmp-blocks: echo-reply echo-request
rich rules:
Code: Select all
management (active)
target: default
icmp-block-inversion: yes
interfaces:
sources: [removed IPs]
services: cockpit ssh
ports: 6160/tcp 6162/tcp
protocols:
forward: no
masquerade: no
forward-ports:
source-ports:
icmp-blocks: echo-reply echo-request
rich rules: