RSView to multiple SLC5/05s on different networks

[glenncooper]

I'm going to show my networking weakness. I have an installation in which we have five production lines and one common feed system. All of these use SLC5/05 processors, and each production line has its own fully isolated network (isolated from each other and the outside world). A production line has several RSView32 nodes and 2 SLC5/05 processors. All of these networks use different fixed IP addressing following the structure 10.10.1.xx for line 1, 10.10.2.xx for line 2, etc. and 10.10.6.xx for the common system.

I now have to get an RSView32 node on each production line to communicate with the SLC5/05 on the common system. So an RSView node at 10.10.1.32 will still be connecting with PLCs on its own network but also one at 10.10.6.31.

This seems like an application for a router. I've read through various posts about using routers and setting up the ports, etc., but am still unclear about how the best way to do this, and how to configure RSLinx Classic to do this.

Does anyone have any ideas or similar experiences? Thanks....

 

[Ken Roach]

Networking is complex, which is why so many bright people make their living untangling them.

This might be as straightforward as broadening the network by editing the Subnet Mask value in the TCP/IP configuration of each RSView32 station and SLC controller so that they are actually on the same IP network.

I'll bet that your system has been set up with the Subnet Mask configured for 255.255.255.0. This is typical of a "Class C" network with 256 possible hosts, while the 10.x.y.z addressing is typical of a Class A network with 16,777,216 possible hosts. That shouldn't hurt your effort.

What you need is a subnet mask that includes all the networks: 10.10.1.x, 10.10.2.x, 10.10.3.x, up to 10.10.6.x.

I'd say change the Subnet Mask to 255.255.248.0

255 = 11111111
248 = 11111000

This will create a network where any host within the address space can communicate with any other:
10.10.0.x
10.10.1.x
10.10.2.x
10.10.3.x
10.10.4.x
10.10.5.x
10.10.6.x
10.10.7.x

You have to edit both the target SLC and the RSView32 station. Think about the SLC-5/05 with its 10.10.2.x address and subnet mask of 255.255.255.0. It knows it can communicate with any node on the 10.10.2.x network but it needs to go through the Default Gateway address to route to another network.

If you don't want to or cannot by policy change the subnet mask of each device, you're going to need a router. It doesn't have to be a complicated or expensive or fast one. In fact, if all your networks physically connect to one another the router really only needs one network port; the so-called "router on a stick" approach.

It may be that some of your switches already offer routing features. Post about the basic physical connectivity betweenn these production lines and about what sort of Ethernet switch hardware is in use.

 

[caughtinthemiddle1]

That is exactly what I had to do for our management team to be able to see production numbers across different networks using the same default gateway. It too was a RSView app.

 

[The Plc Kid]

If hardware arrangements allow a second nic could be added to each pc on each line that is running view 32 and the second nic could be configured to the feed system.

Or as ken said you can change the subnet mask to make it work.

Or a router could be used also. I use the allen bradley stratix 8300 which is a layer 3 switch which is a basic router and is made by cisco so it people like it also.

Hirschman makes some nice offerings on layer 3 switchs and routers also.

 

[The Plc Kid]

Posting the connection topology or proposed topology will help.

Do you plan to use ring,bus,star ?

 

[JBJennings]

I agree with Ken, that will allow all of the processors and View stations to communicate with each other. Check your subnet mask, it is possible that (at least on the PCs) the subnet mask is 255.0.0.0 (Windows default for 10.10.xxx.xxx)

Good luck.

 

[glenncooper]

Thanks for your responses. I ended up doing some tests with just changing the subnet to 255.255.248.0 as Ken had suggested. This worked perfectly, and will work well in our setup. We're going to be connecting these networks together through a Cisco managed switch, and so this should work well. I had played with the subnet before, but had only changed it one end to see if I could ping the other end. In hindsight it seems obvious that I need to change the subnet mask on both ends, but that just hadn't occurred to me.

Thanks again for everyone's help!