RSLinx 204 Comm Error

Hello All,
I'm trying to connect to a 1768-L43/B Via the 1768-ENBT/A through RSLinx and I am having a bit of trouble...the dreaded yellow ? with a red X! I'm able to ping the device and I can also pull up the web browser interface. I'm also able to connect with other device on the network.

Software: RSLinx 3.74
Other hardware: Stratix 6000 switch, unknow IT switch, which I am told is functioning properly.....

Is there anything that I need to look for in the Statix that could be blocking the 1768 from reaching linx?

Thanks!

Still having this issue. Has any seen this in the past?

Quick update,

The issue was resolved. It was a firewall issue with the server. IT was able to correct the problem.

Thanks,
Chris

Are you using the "Ethernet Devices" driver, or the "EtherNet/IP" driver in RSLinx Classic ?

The "Ethernet devices" driver requires you to enter the IP addresses manually and uses straightforward TCP connections to connect to them.

The "EtherNet/IP" driver can browse automatically, but it needs to be able to transmit broadcast commands and receive replies to those commands. Some firewalls and most routers will block those functions.

The red X means that the driver genuinely is getting no reply from the device.

Hi Ken,

I am using "Ethernet Devices". We were able to get the device talking after IT adjusted the Firewall settings.

Thanks for the reply!

Thanks for posting the followup !

I'm making a wild guess here, but I'm going to guess that it was a firewall that was blocking TCP probing. Cisco PIX firewalls had this enabled by default so I've seen it a few times.

The RSLinx Classic "Ethernet Devices" driver has to support both the old CSPv4 protocol (PLC-5E and early SLC-5/05) and the modern EtherNet/IP protocol.

CSPv4 runs on TCP Port 2222. EtherNet/IP runs on TCP Port 44818.

When you first browse a network after putting in the IP address of the target devices, RSLinx has to figure out which type of device and therefore which TCP port to use.

The way they decided to do that is to start with the older one, so they send out a TCP Connection request on Port 2222.

If the device rejects a TCP connection on Port 2222, then it's an EtherNet/IP-only device (like a ControlLogix) and RSLinx switches to Port 44818.

But RSLinx doesn't just send out one TCP Connection request and wait for it to be rejected. It sends out three, and waits for all three to be rejected.

A firewall looks at repeated rejected TCP port probes with no other traffic in between and says "hey, that's a TCP port probing attack. I'm going to block it."

The workaround when you have a firewall like this and need to use the Ethernet Devices driver with ControlLogix is to append the IP address with ":EIP" in the driver configuration.

That causes RSLinx Classic to skip the Port 2222 probe and go directly to using Port 44818. It was implemented quite a while ago; I think it's present in RSLinx Classic 2.51 and later.

As always Ken, thanks for all of the information. I will be adding this to my notes that way I can find it later.

Is the same true for 22 COMM-E devices? Do they communicate with Port 2222 or 44818?

Thanks!

Only the older PLC-5E, 1785-ENET, SLC-5/05, and Pyramid Integrator devices use the old CSPv4 protocol on TCP Port 2222.

The PLC-5E, 1785-ENET, and SLC-5/05 all got 'dual stack' firmware around 2000 that allows them to use both protocols. Most are upgradeable, though it's either a remanufacture or a chargeable item.

Everything else uses EtherNet/IP, on TCP port 44818, for unscheduled messaging and upload/download and browsing in RSLinx Classic.

To make thing slightly more confusing, EtherNet/IP I/O transactions use UDP (not TCP) port number 2222. But that's purely between the firmware on the modules; RSLinx doesn't participate in that at all and you'll only ever see it while doing protocol analysis.

Interesting. Thanks for all the great info Ken.