Can't enter my D-link switch through the default settings

Hello there,

I have a DGS-1210-16 managed switch that I can't seem to get into using the default settings. The switch itself is working and letting traffic through our network at the moment. Online searches say to use the default ip http://10.90.90.90 or telenet 10.90.90.90 but I've done so without success.

I assumed the problem was in the subnet and ip range (my meager knowledge of networking). D-Link has an app that discovers the switch on the network. I used it and discovered the switch and was able to change it's ip and subnet which was 10.90.90.90 and 255.0.0.0 respectively, to the 192.168.x.x and 255.255.255.0 that I'm familiar with.

Doing this gave me access to the switch and it's' web gui, but something tells me this is not how the switch is supposed to work. I've always thought of switches as a sort of network environment manager of sorts, and by placing it in the "working" network itself I feel like my switch should be in a lower layer where other terminals without sufficient privileges can't reach it, like in it's default settings. However, in those default settings where the switch is in a different subnet and ip makeup, I can't reach it through any of our terminals unless I use the d-link app which only gives me the option o change the device ip and doesn't really give me access to the switch's gui/console to manage it's settings.

Right now our network is as follows: we have our modem which is in bridge mode that connects to a router that gives out a DHCP pool to the terminals in the network. The terminals are not connected directly to the router. The router hooks up to the switch, and our server and other terminals connect to the switch and get their IPs from the router. We also have an access point/extender that is on the switch to expand the wireless of the router.

We really want to make use of this switch's capabilities. Hope you can help us!

1 Like

Switches usually get DHCP addresses these days when you plug them into the network.

Well you're not wrong. In larger organizations, or those that have a strong networking team, it's typical to have all your networking devices on what's called a managed network. This is a VLAN subnet where all of your switches/routers/(maybe) management servers sit and has restricted access in and our of the subnet. When I developed the VLAN infrastructure at my old job I had wireless clients on 172.16.0.1/24, wired on 172.16.3.1/16, and my management network was 10.1.100.1/28.

It's not a bad thing to have all your managed devices on the same subnet though, just setup the ACLs on them to only allow access from specific workstations and give the users good passwords to access them. Best bet would be to have one virtual machine that you can access them from... I have a physical box in my office that has no internet access and that's where I do all my management on the network from.

First of all, thank you for taking your time to reply. It has shed some light somewhat and, the solution is just a limitation in my current understanding. If you don't mind, we've got a ton of questions.

A colleague of mine and I decided to play around with the network settings and ran into a couple of problems.

We were under the impression a switch can generate and distribute it's own DHCP addresses, because of a feature in the switch called "DHCP Auto Configuration". So we disabled DHCP from the router and promptly lost connection. So I'm assuming switches do not generate their own IPs. Sorry if we're sounding noob. I'm fairly new to this. The closest I can relate to this is back when we used to connect pcs in lan party with just a cross cable for 1v1s if not on a switch. The switch represents a bridge between pcs much like the cable, and that the IP addresses where what was set on the computers itself.

Instead of placing the switch onto the router's subnet, we placed the router in the switch's subnet instead to gain access. We kept the switch IP at 10.90.90.90, then set the router to 10.10.20.10, with a DHCP pool of 100-254.

Now every thing is fine and dandy until we checked the IP addresses on the clients. The IP addresses read 10.0.0.70 for some wired connections, 10.0.0.240 for wireless. Oh did we mention that our wireless station connects through an access point that is hooked up to the switch?

I would have expected the client IPs to read something lke 10.10.20.xxx, to conform to the gateway IP of the router being 10.10.20.10. Could you explain as to why the middle values turned to zero?

I'll do some research on it, but I'd also like to know how your example "172.16.0.1/24 & 172.16.3.1/16" translates to vlans (which I know nothing about). I'm only familiar with the basic 192.168.x.x:(insert port here). The "/" is alien to me. And why is it that the 3rd number afer .16. are different (0 and 3)? Is this a "layer" thing I hear about? My colleague did something similar with the access point and set it to 10.10.30.10. Seeing that your management address is in 10.100.1/28 clues me that the addresses can be different as long as the subnet for the entire network is the same.

What is the subnet mask you are using for 10.10.20.10?

the /** is the subnet mask btw, where as 255.255.255.0 stands for /24. I'm getting into subnetting here which is easy, but can be very overwhelming at first, but I'll explain it anyway for you. Each 255 is associated with 8 bits, where 255 is:
1 1 1 1 1 1 1 1 or 8 bits turned on. Each bit holds a number, starting from left to right:
1 2 4 8 16 32 64 128 so, 00000011 would be the number 129. In binary your subnet would look like 11111111 11111111 11111111 00000011 or 255.255.255.192 or /26

Right now we're using the subnet that came as default on the switch, 255.0.0.0. I was told today that this is where the 0s in the IPs on the DHCP pool came from.

I'm picking up a little on where you're going with the subnets. I decided to do some reading and it seems the 10.10.10.10 is for a class A network with millions of users, and that your example of 172.16.0.1 is a class B (I think). Both of these seem a little overkill for our office as we only have between about 20-40 devices on the network at any given time. So I decided to go with a class C network, back to the 192.0.0.0 iteration (if that's valid, to my understanding).

In your example you have your "management network to 10.1.100.1/28". It's different from 172.16.3.1/16 and 172.16.0.1/24. If we decide to go with a class C network with a 255.255.255.192 subnet, what static IP should we give the switch so we can access it? What is this management network?