I'm coming from networking where a router/switch has a separate interface for management. Therefore I want to understand how this is implemented for servers/virtualized servers.
I have one subnet for management (remote access to the VM via SSH) and another subnet for connecting the server to the service network. Subnet 0 for management Subnet 1 for webserver, ftp-server, syslog-server, radius ...
Is such a separation practical in server environment or is only a single subnet used to connect a server/virtualized server both for management and service access?
In my current setup, I have two bridges configured on the host and on the guest I have two eth-if (eth0 to br0 and eth1 to br1) but this seems to not be sufficient because the guest needs to know which interface has to be used depending on the application.
Maybe I'm totally wrong here and I'm making it too complicated. Any ideas or help would be great.