Need help with network issues

Okay, this is not something I have done a lot so I apologize if I don’t get what you mean.

I used two computers, one with WiFI and one with Ethernet.
Both got the same results for pinging as yours did. Then I unplugged the cable and I pinged from that one again using only WiFi with the same result.

I am not sure what to make of this but this should mean that they are communicating in between the x.x.2.1 and the x.x.1.1. Right?

I also noticed that the computer using both ethernet and WiFi had two IP’s, one LAN and one for WiFi, which in hindsight is very obvious.

This is enough for you to help me further?

Hmm… This is hard to explain I guess.

Let me get some outside sources to help explain what I am getting at here. I will link them at the bottom, then I will try to explain as best I am able. Bear in mind, I am a better network admin than I am a teacher.

The kind of network you are mentioning is called a “Multihomed Network”. Most private networks will all live inside the same subnet. To understand what that means, you need to understand what a subnet mask is, and what it does, but for now, know that yours does not all live in the same subnet, which is why it is “Multihomed.” With that in mind, here is a rambling and horrible attempt at explaining subnets in a hurry.

In addition to your IP address, your network configuration dialog will want to know some other information. Specifically, it wants to know what your “Default Gateway”, and your “Subnet Mask” are. I will now explain what a subnet mask is, and how it works-- Later, I will explain what the default gateway is, and what it does.

First up-- IP network addressing. An IP network has 4 decimal separated numbers, between 0 and 255. The decimals are there for human convenience, as the number is really just a 32bit integer, expressed in binary-- as far as the computer is concerned. The values given between the decimal points is exactly equal to a value that can be stored in 8 bits, which is why each of these values is called an ‘Octet’. There are 4 numbers, and 4X8=32 Eg, a 32bit number. Each computer needs to have a unique number in order for other computers to send it data.

So, what is a subnet mask, and what does it do? Basically, it divides the IP address space up. Most home networks will use a subnet mask of 255.255.255.0 . What does that even mean?

Well-- It means that the full values of the first 3 octets, and none of the last octet, are used to define “Networks”, while the last octet is used to define the members of those networks.

EG, if we had a subnet mask of 255.255.0.0, then the first two octets are used for networks, and the last two octets are used for members of those networks.

Why is that important?

In order for a data packet to get to another “Network”, it has to be routed. This is accomplished with a device that has a connection to both networks, that is used as a middle man to hand that packet across. For devices within a single network, no router is needed. The subnet mask defines how large the local network is.

Now, you might ask why there isn’t just one enormous network, eg, a netmask of 0.0.0.0. (All of the space is used for devices)— The answer, is that when you have lots of devices in the network, the wire gets VERY VERY BUSY. Basically, a network is like an old fashioned telephone partyline. Every device in that network hears every message that is sent across that network, and each device has to wait its turn to send a message. (Or rather, in the case of Ethernet, it has to try to get a word in edgewise when the line is not super busy.) The more devices in the subnet, the busier that wire will be. To mitigate this, and keep traffic more local, subnetting is used so that packets from outer mongolia arent preventing you from using the network. This is why packets that originate in one subnet are not magically transported to all other subnets. They have to be specifically given a “Route” to get there, otherwise the message will never be delivered. This is by design, so that traffic from some distant part of your building does not become a nuisance elsewhere. (It also keeps your neighbor’s internet traffic from beating down on your own home network.) This isolation is very useful, and is the reason for subnet masks existing.

You can then imagine a router being like an old fashioned telephone operator. You have two “Party lines”, each chatting it up, but one of the people in the first party line wants to talk to somebody in the second, but they dont want all of the conversations in both lines to be forwarded, because then nobody could hear anything. So, the router listens in one both lines, listens for the message to the other line, then repeats that message on the other line. The router is actively listening to both networks at the same time, and serves as this seamless middle man to forward the messages.

To make effective use of this, the people in both party lines need to know who is the designated person that forwards their messages. This is what the “Default gateway” is.

Basically, there can be any number of designated routers attached to your network segment, but one of them is the “Default” route. If there is no specified route, and the destination is not inside your local network, the datagram gets sent to this “Default Gateway.” Again, in most home networks, the default gateway is going to be your home router, which sits between your home network, and the internet. The internet is a hugely interconnected mesh of privately owned networks, each designated with ICANN registered network numbers. When you ask for a website from your browser, your browser asks your router for a DNS entry for the web address you have entered, and then your router responds with a global (Internet) IP address. Your computer then sends a data packet to that address, requesting data. Since that address is not inside your local network, and no dedicated route from your network exists to that target network, your packet is forwarded to the default gateway, which then forwards it out to your ISP’s network— where it then gets forwarded to another router… then another router… then another router… until it reaches the target network.

Why is that applicable to you and your situation?

You have multiple networks inside your site. Unless there is a dedicated route between those two networks, then the “Default route” will be selected. However, your router cannot deliver the datagrams being sent to it, and so it discards them!

Messages to santaclaus type situation, as far as your router is concerned.

On a windows box, you can see what the current routing table for your network looks like by issuing

route print

It will give you output that looks like this:

===========================================================================
Interface List
11...d8 cb 8a c0 72 cd ......Realtek PCIe GBE Family Controller
1...........................Software Loopback Interface 1
12...00 00 00 00 00 00 00 e0 Microsoft ISATAP Adapter
13...00 00 00 00 00 00 00 e0 Teredo Tunneling Pseudo-Interface
===========================================================================

IPv4 Route Table
===========================================================================
Active Routes:
Network Destination        Netmask          Gateway       Interface  Metric
      0.0.0.0          0.0.0.0      192.168.0.1     192.168.0.20     20
    127.0.0.0        255.0.0.0         On-link         127.0.0.1    306
    127.0.0.1  255.255.255.255         On-link         127.0.0.1    306
127.255.255.255  255.255.255.255       On-link         127.0.0.1    306
  192.168.0.0    255.255.255.0         On-link      192.168.0.20    276
 192.168.0.20  255.255.255.255         On-link      192.168.0.20    276
192.168.0.255  255.255.255.255         On-link      192.168.0.20    276
    224.0.0.0        240.0.0.0         On-link         127.0.0.1    306
    224.0.0.0        240.0.0.0         On-link      192.168.0.20    276
255.255.255.255  255.255.255.255         On-link         127.0.0.1    306
255.255.255.255  255.255.255.255         On-link      192.168.0.20    276
===========================================================================
Persistent Routes:
None

It takes getting used to to read these tables, and ordinary people usually dont need to bother with them. Only people that need to actually administer a network need to mess with these things, and most people are not network administrators, so it is OK if this confuses the bejeebus out of you. That said, I will try to explain what this table says.

First, is the global context. Here is where the default gateway rules the roost. It says "For any network, ANYWHERE (netmask of 0.0.0.0), not specifically defined in this list, forward the datagram to the default gateway. (In this case, 192.168.0.1) Since I am using a windows workstation to view the local routing table (of that computer!), the interface it uses to talk to the default gateway is listed (192,168.0.20), which is the IP address of that workstation.

Next up, are the 3 " global Loopback" routes. These are special. You can basically ignore them-- they are used for diagnostic purposes and local services, and basically are just an alias for “This computer”, or the “local host”. Anything you send to that network never makes it onto the wire, and instead just gets delivered right back to you instantly.

After that, we start getting into actual routes.

The first one, is the default rule for any packet destined for the 192.168.0.X network. It has a network mask of 255.255.255.0, which sets the size of the network portion of the segment, and any message being sent to a machine in that network will go out on the interface at 192.168.0.20. The next two are the “Local Loopback” and “local broadcast” routes

Then you have another set of loopback routes after that. (note how they reference 127.0.0.0, and the current system’s IP address!)

And then you have the global broadcast routes at the bottom.

You might notice that I do not have any routes pointing to another private network. This is because I do not have a multihomed configuration.

Since you DO have a multihomed configuration, your router needs to have routes between these networks, in order to deliver messages.

These would be defined similarly to the default gateway, but would instead look something like this:

  192.168.0.0    255.255.255.0         On-link      192.168.0.20    276
  192.168.1.0    255.255.255.0         On-link      192.168.1.20    276

This would be for a router that has interfaces at both 192.168.0.20 and at 192.168.1.20. That is two network cards, each with its own IP address, attached to the appropriate networks.

Since such a route is defined, when a packet destined for one of those networks is presented to the router, it knows what interface to forward it out of, and all is golden with the world.

Now, like I said earlier in the previous post-- Some corporate networks PURPOSEFULLY keep segments from sending messages to each other, while still having a route out to the internet from both. This is used to keep say-- People waiting in the hospital lobby, looking at youtube videos, from being able to see the medical records computers in the hospital’s primary network, and vice-versa, while both can still reach the internet. Since both are private networks, there is no “return route” from the internet side back into the other private network, so packets can never get across. :stuck_out_tongue: Like I said, this is often PURPOSEFUL. Medium to large enterprises will have a special facility, called the Network Operations Center, or NOC, that handles all these routes, router configurations, IP address assignments, etc inside the enterprise-- They get REALLY REALLY mad if you go behind their back and start plugging things in, or making routes between networks they want isolated. You did not specify if this is a multihomed HOME network, or if this is a multihomed CORPORATE network. If it is the latter, you MUST go through your NOC, and more than lilely, they are gonna want to know why you are plugging in a strange server that they do not know about. :stuck_out_tongue: Trust me, it is best to stay on their good side.

If you are working with a multihomed HOME network, then I presume that you get the glorious task of being the “NOC” yourself.

You might find the following article on multihomed networks useful.

I know I am the one who needs help here but I am still sorry for the late reply. I have had some other things on my mind.
Thank you so much for your lenghty reply but like you say it, it is quite a lot to take in!

Anyway here is the route print:
routeprint

I sort of understand what I see but I am not sure what to make of it.

Just to make it clear, for all intents and purposes this is a home network.
It’s a small company and this network is at the office at a private property. Usually there are 3-5 people at the office and at most 10.
Every computer is a windows laptop. Some are connected via cable and some are WiFi.
Ethernet is at 192.168.1.x and WiFi is at 192.168.2.x.

From doing the ping test it looked to me like the ethernet and WiFi were definitely communicating.

Where do I go from here?

To be sure that the networks are actually routing packets between them, you need to attempt to ping from a machine connected to only one of the networks, to the other network.

EG, attempt to ping a wired machine, from a wireless laptop (that is not connected to the wired network), and vice-versa.

I mentioned that the results of “route print” are the ‘local’ routing table for that machine. For the most part, windows uses edge discovery protocol to communicate with the default gateway to get the gateway’s routing table. It then makes local amendments, based on the local presence of network devices, to create that computer’s routing table. This table is how the machine knows to send packets out of its interfaces, so that it does not try to reach the internet through a non-internet facing private lan, etc. It is useful to see if there is an inherited route (obtained through border discovery) that connects the two networks, as it will show up in the routing table.

The routing table you posted appears to be from a 192.168.2.XXX device. It does not have a route leading to the 192.168.1.XXX network defined. This means that such a packet would be forwarded to the primary gateway. This MIGHT work, if the primary gateway has an internal forwarding rule, but then again, it might not.

You are certain that your pings are getting responses under the conditions I specified? (Ping from wireless only host, to wired host— and vice-versa?)

I was certain that the pings worked both ways but apparently it didn’t, which just made me more confused.

What I want is for the computers using only WiFi to be able to map the MyCloud to a drive since right now it only works when using ethernet.

Using the WiFi with the IP “.2.XXX” i get reply pings from other computers using WiFi only, of course.
There are also responses from Everything using ethernet. One printer, the MyCloud, other computers. All on “.1.XXX” IP.
The computers using ethernet only do not get responses from computers on “.2.XXX”.

Again, I am very new to all of this so I apologise if it is messy and unclear.

I think you need to look to your router, and find out why it is using different subjects for wired and wireless connections.

Which is something previously indicated and alluded to in previous days…

As the OP has found out using separate IP subnets means one typically has to create routing between them and that is where they appear to be stuck. Of course the simple and easiest solution is to use a single IP subnet range for all devices, both Wifi and wired… :laughing:

I know. But the OP doesn’t seem to be getting anywhere with the clever stuff, so I’m suggesting the simpler option might be more successful…

Really, the OP needs to administer the router. A persistent route between the interfaces needs to be added. Then it will all be happy days again.

I have seen enough to 90+% be certain that there is not a valid route between these subnets.
At this point he needs to ssh into the router and add a persistent route. (The details of how to do that will be router specific though. Different manufacturers do it differently, so it’s always “What manufacturer? Ok, what model? Oh goodie-- here is the manual, do it that way.”)

The OP is not a network administrator though, so hmm…

1 Like

Well, most routers provide a web-based administration interface, so, hopefully, SSH won’t be needed…

But the router user manual will be useful. And, hopefully, useful…

Hmmm… :blush:

consumer routers have web interfaces.
Production routers typically do not.

You know, things like managed fabric switches, or cisco production routers…
See for instance, the quick configuration guide for a Cisco 4000 series integrated services router…

BANG-- straight up “You use the management port to SSH into the router to administer stuff.”
No, “Oh yeah, use the web portal to do stuff, its so easy!” No no no. It’s “Hey, you are an administrator. We dont hold your hand here. Use SSH like the professional you are.”

I guess you can see which way I am used to doing. :stuck_out_tongue:

Again, each manufacturer “Does it their way”. You have to get to know their product to administer them effectively.

I was going to ask “Do you really think someone buying a consumer product for use in the home is likely to have a Cisco network router…?”

but then I went and had another look at the OP, and this reminded me it was an office environment.

However, he does appear to be using an Asus RT-AC66U router. Not a Cisco… He’s posted a screenshot of the web interface for the Asus router he’s using.

I think the OP needs to go and speak to his corporate IT admin, and get them to sort out the router configuration. And, whilst he’s there, ask them if they’re aware he’s plugging a consumer NAS onto their network…

Oh, I told him that he needed to contact his network operations center, and ask if the missing route was purposeful or not, and that they would likely not take kindly to him plugging in strange servers. (It was in the really big post I made about how IP routing works) I mentioned that it was very important to not make them unhappy. :stuck_out_tongue:

I presume that this setting is more “Small to mid size business”, where they dont have a NOC, but have a hamfisted set of inexperienced people trying to use shoestring budget hardware (inappropriately deployed consumer grade stuff), to get stuff done, and the lack of budget means they have suboptimal talent, or try to do it themselves.

In which case, this guy would be our “admin”.

1 Like