Recommended Posts

Hi,

This is my very first post in here, so please go easy on me...

PROBLEM:

Anyway, I have several embedded linux systems, All connected to a window 7 system. The Window system has a default IP address of 169.254.0.100 and has a DHCP server running on it, which allocates addresses to the embedded linux blades in the 169.254.0.x subnet.

The linux systems are separated by sets of 3 or 5 CPUs, and we do not want to allow one of these( in each set) to talk to the Window 7 server. But all of the blades need to talk to each other. So, we do have a separate subnet 169.254.1.x for all of the blades. This way, they can all talk to each other, but only the ones that are on the 169.254.0.x can communicate with the Windows7 pc. The single blade which we do not want to communicate with the windows PC will always have the same IP address of 169.254.1.199. So if all of these were connected to the same switch, we could potentially end up with multiple nodes with the IP address of 169.254.1.199 on our network.

MY SETUP (using a Netgear GS724T switch and a Netgear FVS318N router):

So, here is what I have as far as a setup goes:

I have put each of these sets of 3 or 5 CUPs on their own separate VLAN connected to the  Netgear GS724T switch. so I have the following (I am using sets of 3 CPUs here for simplification):

               IP of 169.254.0.1 and 169.254.1.1 << same system but two different IP addresses ( port 3 of the switch)

               IP of 169.254.0.2 and 169.254.1.2 and  << same system but two different IP addresses for (port 4 of the switch)

               IP of 169.254.1.199  << single CPU that we do not want to connect to the WIN7 PC, but want it to talk to the other 2 cpus above. (port 5 of the switch)

all on the same VLAN, say VLAN4

and another set of 3 CPUS on its own VLAN, say VLAN5 (as follow):

              IP of 169.254.0.3 and 169.254.1.3 << same system on two separate subnets (port 8 of the switch)

              IP of 169.254.0.4 and 169.254.1.4 << same system but on two separate subnets (port 9 of the switch)

              IP of 169.254.0.199  << single CPU with hard coded address that we do not want it to connect to the WIn7 PC, but can talk to the above 2 CPUS. (port 10 of the switch)

And I have the WIN7 PC on it's own VLAN, say VLAN1 (or the default VLAN of the switch) on port 2 of the switch. 

This setup does provide the isolation that I am trying to get keeping each of the CPU sets connected together, but keeping them from connecting to the other set of CPUs, BUT, it does not let me connect to my Windows PC, which is on a separate VLAN(VLAN6). So I need to have the traffic on VLAN4 and 5, to get to VLAN6.

I have followed the instructions outlined by Netgear in this article from Netgear support , which sets up a trunk line to a router (Netgear FVS318N) and connects all of the incoming VLANs to the WAN port. But I am not trying to get these out to the WAN port, rather, I need to get the packets from one VLAN to be routed to my VLAN1 of the switch.  (The Trunk line is set-up on port 1 of the switch connected to port 1 of the router. )

QUESTIONS:

Is this possible with what I have? and if so, how can I do it?

I have configured each VLAN with it's own routing IP address on the switch, But the router does not allow me to set-up routes to route packets from one VLAN to another. How is the Inter-vlan routing done with the Netgear equipments, their documention is horrible and is pretty much non-existent when it comes to inter-vlan routing. 

Any suggestions, comments, alternate approaches, pointer to documentations and etc would be appreciated.  

Thanks in advance for any help.

 

 

  Quote

blades in the 169.254.0.x subnet.

That is APIPA and not routable.. Why would you pick that??  Just use something in actual rfc1918 space... 10.x.x.x, 192.168.x.x or 172.16-31.x.x

There is no point in reading the rest of your issue, you can not route that address space - pretty much every device that routes will not route them...

But in general you do not have to add routes to a router for its own connected networks, if will know that it has network a.b.c.d connect and a.b.c.e and will pass traffic between them without having to actually create a route.. But again that space is not routable!!!

This topic is now closed to further replies.
  • Posts

    • Google open-sources zero-knowledge proof code for enhanced online privacy by Paul Hill When you go out to a physical store and attempt to buy alcohol or cigarettes, you’ll get asked to show some ID; online, when you want to visit adult sites, you don’t currently need to prove your age, but lawmakers in various countries are looking to change this. One technology that has been developed to address the age checks needed is Zero-Knowledge Proof, where you prove your age without having to reveal other data like birth date or ID. To help companies develop their own Zero-Knowledge Proof (ZKP) technologies, Google has released its ZKP libraries as open-source, meaning they can be taken and used in other projects or adapted. This will make it easier for third-party developers to create privacy-enhancing age verification applications. We heard a bit about ZKP from Google earlier this year when the company said that it was integrating the technology into Google Wallet to help disconnect your age from your identity. It also said at the time that it would be using it in other Google products and partner with apps like Bumble to help with verification. It also said at the time that it would open-source ZKP; that promise has now been fulfilled. As libraries, Google’s newly open-sourced software can be integrated into a whole range of applications. As mentioned, Google is using it in Google Wallet, and it could also be used to verify the age of people visiting adult websites without needing to reveal their identity as part of the verification process. Google has also said that the European Union’s eIDAS regulation encourages EU member states to integrate technologies like ZKP into the European Digital Identity Wallet (EUDI Wallet). The open-sourcing of ZKP could help with the acceleration of the development of these EUDI Wallets. The open-sourcing of the ZKP libraries by Google will benefit various groups of people. We’ve mentioned that developers will benefit as they can use the libraries in a variety of apps. Businesses will also benefit by being able to meet privacy needs more easily. Another group to benefit will be researchers who can use this “more efficient and performant ZKP implementation” to help create new applications and uses of technology. Finally, users will benefit from more private and secure digital ecosystems. Now it remains to be seen how much adoption these ZKP libraries will get, given the growing need for such technology. Image via Depositphotos.com
    • I Have a batch file that opens 2 websites dealing with money. I want to add to it with the first step to remind me to start a vpn extension that I use with chrome.  If  it is easier to use powershell that is fine too.  thx   Hope this is enough info.
    • IMO, Vivaldi feels overdesigned at this point.
    • You have to look at the value proposition. Any product can be good or bad at either the right or wrong price. At $110, the GTX 1050 was considered a good GPU for the price, and again at $160 the GTX 1650 was extremely popular and considered one of the best value GPUs ever. The problem here is this RTX 5050 has a performance uplift over the 4050 so small it is basically lost in the margin of error, yet they still price it at $250!! That is bad value. At $200 it is at least in line with the other overpriced GPUs, at $175, it would actually be a good product!
    • I built my first computer with a low profile GTX 750 Ti OC and it was a great 720p card for its time. And that was a $160 card at the time. The 50 models used to be a good mid-range for price to performance. Not anymore.
  • Recent Achievements

    • Week One Done
      956400 earned a badge
      Week One Done
    • First Post
      loose_observer earned a badge
      First Post
    • Week One Done
      BeeJay_Balu earned a badge
      Week One Done
    • Week One Done
      filminutz earned a badge
      Week One Done
    • Reacting Well
      SteveJaye earned a badge
      Reacting Well
  • Popular Contributors

    1. 1
      +primortal
      446
    2. 2
      ATLien_0
      157
    3. 3
      +FloatingFatMan
      148
    4. 4
      Nick H.
      65
    5. 5
      +thexfile
      62
  • Tell a friend

    Love Neowin? Tell a friend!