FTLinx, Local Subnet Devices Thru EN2T

[cb423]

We have two, nearly identical machines.


One configuration utilizes a L33ERM and I am able to see and access the local subnet devices, such as troubleshooting drives via CCW.


The second instead utilizes a 1756-L72S, on this one I am unable to access the local subnet devices.


I spent about 3 hours on the phone with RA support and they were not able to come up with a solution but have said they will get back to me. The two long term solutions I see are, NATing the rest of the local subnet devices to have plant IP addresses, or adding a second EN2T to the Contrologix Rack


Is anyone aware of a better solution (hopefully one utilizing the existing hardware if possible)? See attached image






Thank you for your assistance

 

[Ken Roach]

Welcome to the PLCTalk forum community !

Can you describe what NAT device you are using and how it is configured ?

Exactly which software does that screenshot describe ?

NAT and router and VPN connections often prevent broadcast-based discovery protocols from working, and EtherNet/IP discovers devices using TCP broadcast messages. RA works around this by letting you manually add a device to the network schema.

That includes when you bridge through a 1756 backplane from one Ethernet module to another.

Another thing you might want to examine on the two similar systems is the Default Gateway setting in the IP configuration for the drives and POINT I/O. An absent Default Gateway will let devices work fine on a local network, but you would typically set Default Gateway to the LAN-facing IP address of that NAT router if you need to communicate from outside the LAN.

 

[cb423]

Welcome to the PLCTalk forum community !

Can you describe what NAT device you are using and how it is configured ?

Exactly which software does that screenshot describe ?

NAT and router and VPN connections often prevent broadcast-based discovery protocols from working, and EtherNet/IP discovers devices using TCP broadcast messages. RA works around this by letting you manually add a device to the network schema.

That includes when you bridge through a 1756 backplane from one Ethernet module to another.

Another thing you might want to examine on the two similar systems is the Default Gateway setting in the IP configuration for the drives and POINT I/O. An absent Default Gateway will let devices work fine on a local network, but you would typically set Default Gateway to the LAN-facing IP address of that NAT router if you need to communicate from outside the LAN.

Thank you for the welcome Ken!


The NAT is a 9300-ENA. The configuration is pretty simple, it is set up with a static Plant IP address and its local private address is 192.168.1.1. The PLC has its address translated such that it can be accessed over the plant network, thus that is set up in the 1:1 NAT tab. I think that is about it as far as the NAT configuration goes.


This particular screenshot is of FactoryTalk Linx Network Browser, however I am able to replicate the exact same screenshot within RSLinx Classic.


Unfortunately I cannot seem to manually add the devices in this case like I can on some of our other equipment that has multiple Network Cards. It just seems really silly to me to have two network cards plugged into the same switch, when the data is obviously accessible somehow since I am able to access the drive properties from within the Studio 5000 program (or I can even temporarily translate the Drive's IP address to a plant address and access it that way, however I am not trying to tie up that many IPs by giving every drive on every machine its own Plant IP).






The default gateway for the drives is set to be the NAT device, 192.168.1.1

 

[Ken Roach]

Thanks for the details; this is an interesting question.

I'm not sure why you're able to connect to the devices shown under the "A>" network port on the 1769-L33ERM. It might have something to do with it having the TCP/IP object directly at the same CIP port as the CPU, instead of using an actual backplane (like the 1756 controller) or a virtual backplane (like an older CompactLogix would use).

On the 1756-L72S, you have just the one 1756-EN2T, correct ? You aren't bridging into the controller on one -EN2T and out from another.

This might be something interesting about a CIP proxy object that is specific to the CompactLogix. Dang, now I'm curious.

 

[JeremyM]

If you right-click "Ethernet", you should be able to go into its properties and add IP addresses manually.

(in RSLinx)

 

[plvlce]

If you right-click "Ethernet", you should be able to go into its properties and add IP addresses manually.

(in RSLinx)

That picture was taken in Studio; if you examine the picture from the original post which was taken in Linx you will observe that there is no such option -- the EN2T does not show any ability to drill down further.

EDIT:

I'm not sure why you're able to connect to the devices shown under the "A>" network port on the 1769-L33ERM. It might have something to do with it having the TCP/IP object directly at the same CIP port as the CPU, instead of using an actual backplane (like the 1756 controller) or a virtual backplane (like an older CompactLogix would use).

On the 1756-L72S, you have just the one 1756-EN2T, correct ? You aren't bridging into the controller on one -EN2T and out from another.

This might be something interesting about a CIP proxy object that is specific to the CompactLogix. Dang, now I'm curious.

On 5380 CompactLogix with dual-IP setup you can access devices on the machine network through the controller while you are on the enterprise network in a similar fashion, seems like it might be related.

 

[JeremyM]

That photo was taken in Studio, yes. The concept I mentioned applies to Linx.

 

[cb423]

Thanks for the details; this is an interesting question.

I'm not sure why you're able to connect to the devices shown under the "A>" network port on the 1769-L33ERM. It might have something to do with it having the TCP/IP object directly at the same CIP port as the CPU, instead of using an actual backplane (like the 1756 controller) or a virtual backplane (like an older CompactLogix would use).

On the 1756-L72S, you have just the one 1756-EN2T, correct ? You aren't bridging into the controller on one -EN2T and out from another.

This might be something interesting about a CIP proxy object that is specific to the CompactLogix. Dang, now I'm curious.

That's correct, we just have the one EN2T on the L72 rack. Hence why as silly as it seems, the obvious solution to me is to put a second EN2T in the rack that plugs into the same switch as the first EN2T. I'm almost certain that would resolve the issue, but it seems so unnecessary.

That picture was taken in Studio; if you examine the picture from the original post which was taken in Linx you will observe that there is no such option -- the EN2T does not show any ability to drill down further.

EDIT:On 5380 CompactLogix with dual-IP setup you can access devices on the machine network through the controller while you are on the enterprise network in a similar fashion, seems like it might be related.

You're right, in Linx I have no ability to manually add devices. I am familiar with this practice on other machines that do have multiple network cards. I thought it could somehow be related to the dual setup, and that is what Rockwell Support thought initially too, however this particular Compactlogix only supports one network connection.

Rockwell Support was supposed to get back to me today after our long call but I have heard nothing new from them thus far.

 

[cb423]

I've received an answer from Rockwell today, not sure I fully understand however I guess the conclusion is that this is not possible and we will have to either allocate 50+ Public IP Addresses for our drives, downgrade our Contrologix to Compact, or install an additional Ethernet Card. Thanks to everyone that helped



See below for their response:


Those drives are being browsed via the L33ERM's ethernet channel A, so the perspective is from the internal ENBT - the NATR is not between the L33ERM and the drives.
Same situation with the EN2T. If the customer wants to see them at their NATted address they'll have to connect RSLinx to the public side of the NATR and put the IPs they want to see in an ETH Devices driver, and their PC will of course have to have an IP appropriate for the public's subnet range (10.67.whatever)