I have been tasked with trying to clean up the Network. On this network i have 14 AB PLC’s and 6 Vision Clients and 8 PVP 700. Right now my Problem is there is to much traffic going to the server. We are Using a Karbon 750 as a server that is Running windows 10. I have all the devices running through unmanaged ethernet switches. When i try and move the 10 AB PLC’s and 6 vision Clients to the Stratix 5200 so that i can filter out the rest of the unnecessary Traffic I start to have issues with My Transaction Groups in ignition. When I have my System running through the managed switch i will get my trigger request then the transaction group will go into an errored state. Sorry if my grammar Is not correct.
Please do not double post, per the rules.
Please share actual details. Nothing you've declared gives any clues to what might be happening. Also, proper grammar and capitalization makes everything easier to read, especially for the large contingent of non-native-english speakers here.
Need to share a little about your actual architecture.
What model/version are the 14 AB PLCs? How are they connected to the network (EnxT modules or native L8 ports if CLX). Standard star network (no DLR)?
What driver are you using to communicate to the PLCs?
Is everything on a single star network and all on the same subnet? Any NAT going on? Are you using a single VLAN?
You mentioned stratix 5200s and managed switches. Are they one in the same? I’ve honestly had terrible experiences with 5200s…much prefer the 5700s.
How many tags and what poll rates are you trying to manage from the 14 PLCs?
Have you checked the port diagnostics of each device to make sure you’re not accumulating MAC or FCS errors?
What does the driver load look like for each PLC connection?
Need to check the fundamentals. 14 PLCs and 6 clients is not that heavy of a load unless you drug the entire global tag browse into Ignition for each PLC.
1 Like
What are the symptoms you are dealing with that suggest that there is too much traffic running to the server?
A server is likely to have a gigabit network connection, and will be easily able to handle those few PLCs provided you are sensible in your use of the PLCs.
You might want to double-check your VLAN setup or driver load; sometimes it’s just one chatty PLC or tag group flooding everything else. It’s a pain to troubleshoot, but once it’s clean, the difference is huge.
The models very from PLC to PLC. I have 6 1769-L33ER CompactLogix 5370 Controllers, 2 1669-L30ER CompactLogix 5370 Controllers 2 5069 - L320ER CompactLogix 5380 Controllers, and 2 1769-L19ER-BB1B CompactLogix 5370 Controller. The Subnet for this network is on a class B subnet mask. “Understand i did not set this up i came in to this” mean everything that has an ethernet connection can does talk to the Server. The Stratix 5200 is the managed switch. I was told by my supplier that the 5700’s are no longer being produced and forced us to buy the 5200’s. we are pulling in upwards of 85,000 tags in to ignition. The Polling Rates Vary from Transaction Group to Transaction Group. Anywheres Between 250 ms to 1 Second.
For an Example one of my UDT’s i am pulling in to ignition is 29596 bytes. For this specific Machine it is actually doing it 3 times to 3 different transaction groups.
The contractors that set this system up was not sensible about there set up. They differently did not go spearing on the PLCs or the server.
The VLAN setup is not the issue. I thought the same thing. I disabled everything on the managed switch and tried to run it as a an unmanaged switch. With that being said, The Stratix is still trying to filter out the packets. In lesser words it is doing it's job.
Onces I have more time to collect more information on some of the questions i did not answer or did not answer fully i will share them to the chat.
There is no real reason why this small amount of tags would saturate a 100mbit ethernet connection, are any of the PLCs running over 100% overload in the gateway?
At this time i do not have the answers to some of the questions being asked. With being said some of the things being asked was not even on my radar. I will share more information when i have collected the data/ Information. Thank you.
No bandwidth issue. All device Load % are low (≈0–12%).
Something i did notice on my servers CPU, The average usage is a bit higher than expected.
Ethernet is sending 996 kbps and receiving 1.8 Mbps
With the Transaction Groups that seems to be having issues when i switch over to the managed switch the execution scheduling is set to timer at 1 second. the others are set to 250 ms. I could be having a timing issue with my groups. Again Sorry about my Grammar.
You might be having issues on the database side, what are your longest query logs showing in the gateway?
Sir, you might just be on to something.