Zurarczurx wrote:
I was asking VikingOS. Judging by their answers on these boards they seem to know what they are doing so I was just curious about their setup. I'm a gross amateur when it comes to network stuff and thought that maybe I could learn something.
If you’re running a home lab and want to learn more about networking, Ubiquiti Unifi gear can help you learn about a whole lot of networking. The Unifi gear has a reasonable user interface, but is very capable, and extendible into sensors, monitoring, storage, access control, and cameras.
I'm curious as to why you don't use the router's binding to assign the static address so it sticks even if you cycle power on the router.
There are various ways to assign IP (IPv4) addresses.
- from the pool of IPv4 addresses available to the local DHCP server.
- from the DHCP server, but with a fixed IPv4 address. The client is set to use DHCP to request its address, but always gets the same address. This uses the MAC address (media access control, and not the Apple brand) to map the client to the IP address. (I suspect this is what you are referring to by “binding”. And “BIND” has a whole ‘nother networking connotation, too.)
- with a manually-assigned fixed IPv4 address, but the assignment is uncoordinated. Network IT groups don’t like this.
- with a manually-assigned IP address, coordinated with the IT group. IT groups still don’t like this, but they usually don’t like this less than they don’t like the uncoordinated IP address assignment strategy.
I use it to assign static IP addresses to peripherals like NAS and printers in an IP address range
which I have firewalled off from the internet using the router's firewall.
Pretty much any NAT-based firewall-gateway-router-NAT box will use one of the IPv4 private-address blocks:
10.0.0.0/8: 10.0.0.0 - 10.255.255.255
172.16.0.0/12: 172.16.0.0 - 172.31.255.255
192.168.0.0/16: 192.168.0.0 - 192.168.255.255
These blocks are used all over the place, but NAT means these IP4 addresses don’t (usually) leak onto the public network.
NAT tends to be configured to block most incoming traffic at the firewall-gateway-router-NAT box, but NAT is its own fine mess. And IPv6 variously doesn’t involve NAT.
As for the original poster’s case here, a number of networks will either use public IPv4 and/or IPv6 addresses inside their network, or a mix of public and private addresses. Smaller networks tend to use private IPv4 addresses, and either private or public IPv6 addresses, with details depending on the ISP and the IT group.
The central “fun” for the original poster is getting a host name assigned for the IP address, and there are various ways to do that including a private DNS server, or multicast DNS (mDNS, Bonjour, Zeroconf), or both.
Private DNS servers can be integrated with the local DHCP server, which makes assigning a static IP address that much easier. The client then gets the same host name everywhere, and the routers, DHCP, and DNS servers deal with the name-to-address mapping.
mDNS is great scheme for small networks, but it does not scale well in moderate and larger networks. It’s usually configured segment local, in most multi-segment networks. mDNS may not be particularly reliable in the networks of some organizations for another reason: those networks that followed an ill-considered and since-rescinded Microsoft suggestion to squat in the .local DNS top-level domain.
Yeah, using a static IP address is nice, but (uncoordinated) use of static IP really doesn’t work all that well in even a moderately- complex network, and tends to get the local network IT group “cranky”.