This post is more than 5 years old
3 Posts
0
10026
Same static IP in internal NIC & in USB type C NIC docking WD15
I have a DELL e5480 and I found in BIOS the Pass Trought MAC Address, but how can I use in Microsoft Windows 10 and 7 ( no WinPE or PXE service ) .
We have 3000 Client with Static IP my intent is to setup a static IP address on the internal network card and use the same when I connect the laptop to via Type c connection. Is it possible?
jphughan
4 Operator
4 Operator
•
14K Posts
0
November 10th, 2017 11:00
If you have users moving between desks, then I'm afraid the problem may get even worse, because not only will Windows consider the dock's NIC as a separate device from the built-in NIC, but I believe that each dock's NIC will be treated as unique devices, so a static IP configuration configured while attached to one dock would not carry over to another dock's NIC when you switched. I haven't verified this, but you should definitely check.
The reason for the difference compared to the E-Dock is that the E-Dock used a completely different design. The E-Dock basically just had a bunch of physical connectors built into it, but the actual chips to run those connectors were all built into the laptop, and the E-Dock just routed every individual pin from every individual connector into the laptop (with some exceptions not relevant here). That's why the E-Dock connector had over 200 pins. The benefit to this approach is that you don't need to install any drivers for the E-Dock (like you do with the USB-C/Thunderbolt docks), since all of the hardware was already built into the laptop anyway. And in the case of the NIC, the same internal NIC chip was just rerouted to pay attention to the dock NIC pins rather than the built-in NIC pins (although this meant you couldn't use both simultaneously). The downside is of course that the system's dock connector and the dock itself are both proprietary, so you can't use Dell laptops in third-party docks or Dell docks with another vendor's laptop.
By comparison, the USB-C/Thunderbolt docks have actual audio and NIC chips built into them, and they all ultimately package their traffic as USB traffic to pass to the system (the TB16 dock packages USB traffic inside PCIe using a USB host controller also built into the dock). One benefit to this design is that it would allow a system that didn't have a built-in NIC chip to still use a wired LAN, since the dock has everything required built into it. It also allows for far greater interoperability between laptop models and dock models. But it does mean that the dock devices require their own drivers, and as you've found, that means devices are treated separately from any built-in equivalents.
Bottom line: If a consistent IP address for your systems is a requirement, you're really going to want to start planning to implement DHCP reservations, because even if all of the benefits I already described aren't enough to entice you, then the behavior I described at the beginning of this post will probably force you to anyway.
spiccolo
3 Posts
0
November 10th, 2017 05:00
Windows 7 Function witouth problem. The only problem ( for the company ) is use single ip adress in different position ( the users whit older docking station - change desk position witouth problem )
jphughan
4 Operator
4 Operator
•
14K Posts
0
November 10th, 2017 05:00
In addition to the above, I’m fairly certain the 5480 will not work with Windows 7 because the CPU is too new and Microsoft is no longer supporting new CPUs on Windows 7. You might also have trouble getting drivers.
jphughan
4 Operator
4 Operator
•
14K Posts
0
November 10th, 2017 05:00
Just to give you a few more things to consider, DHCP reservations would be a much better solution than static IPs configured on the laptops anyway for several reasons:
- It's faster and easier to assign IPs on a server than going to each laptop, and it's MUCH faster and easier to change IP assignments later if that's ever required since you don't have to track down that system.
- It's much easier to maintain a list of existing IP assignments since you can just check the DHCP server's reservation list rather than maintaining a spreadsheet or whatever.
- If you set a static IP on the laptop, the NIC wouldn’t be usable elsewhere unless the user switched it back to DHCP, although then they’d need to remember their static IP when they returned. And if the user isn’t an administrator on the laptop, they wouldn't be able to do that, so their NIC would become completely useless outside the office.
If you have 3000 clients, surely you have a highly available DHCP server infrastructure available? If not, it's not too difficult to deploy, although if you have lots of subnets/VLANs, you'd need either DHCP servers in each one or DHCP referral configurations on the routers/L3 switches that separate the network segments.
jphughan
4 Operator
4 Operator
•
14K Posts
1
November 10th, 2017 05:00
MAC Address Passthrough is just a BIOS-level feature that causes the system to read the built-in MAC and then use that to override the default MAC of the USB-C NIC, since user-defined MACs are a pretty widely supported feature. It's intended for to allow things like DHCP reservations for systems to continue working even when the system is docked. But the system still sees the built-in NIC as a physically separate device from the external NIC, which means Windows does too. In my testing on Win10 1709, Windows warns if you try to configure the same static IP on multiple NICs, and although it does allow you to save that configuration on the second NIC, it's blank when you come back to check (the first NIC is unaffected).
The only way to achieve a consistent IP address in this setup would be to use DHCP reservations for that single MAC address, since MAC passthrough means that will always be consistent.