I downloaded 7.2.1 and installed it a couple of nights ago on a widows machine. I now am having issues with tags going stale and not repopulating on a plc5 and 10 compactlogix that just totally disconnect after they run a while and also the gateway just shutting down and restarting in which I will lose some of my contrologix processors. I can go in and tell it to browse and the will come back but sometimes I have to do this a couple of times. Along with this I have some transactions that will not “read” from the processor but will trigger. I saved a wrapper log for you to look into. I work nights in the midwest so I probably won’t get back to you today.
Thanks Paul 
wrapper.log (723 KB)
There’s a bug fix in the 7.2.2 development build posted that will address your driver problem.
What normally happens is if a driver request times out 3 consecutive times it then will attempt to reconnect to the PLC. The bug was that it was simply disconnecting and leaving it at that.
As to the gateway shutting down and restarting by itself - can you send ALL of the wrapper.log files to support@inductiveautomation.com? You are probably running out of memory in the gateway and the service is restarting it for you.
Try upping the max memory to 1536MB in ignition.conf. You cannot go higher than that unless you are using 64-bit Ignition/64-bit Java.
I’ll give it a try when I go back in tonight. Thanks!!
Here are all of the wrapper logs.
wrapper3.log (1 MB)
wrapper2.log (1 MB)
wrapper1.log (1 MB)
wrapper.log (959 KB)
I upgraded to the development version Friday night. It did seem to help my tags connect back up but I am still having issues with stale data. I think we need to check into why the connections are dropping out in the first place. I did not see this before the 7.2 upgrade except for the occasional flicker through when trying pull up some charts or browse the processors that have a lot of i/o points. Also I do know where to increase my usable memory all I found is where I can increase the designer usable memory. Thanks Paul
Assuming you mean you don't know where to edit the usable memory.
You can change this by editing the 'ignition.conf' file in the ignition install directory. You'll want to edit this line (line 56 I believe):
Maximum Java Heap Size (in MB)
wrapper.java.maxmemory=1536
Thanks Dave I found it.
I went back to 7.1.8 last night and all my tags are back to updating as they have in the past. I’m not exactly sure why I am having the problems with stale data with 7.2. The fix for 7.2.2 did get them to reconnect but sometimes they would stay stale for a half hour or more. any ideas to look at?
It would be ideal if you could call into support at some point during our business hours and do a GoToMeeting with them. Is that a possibility at all?
Yea I’ll try tomorrow I work nights in the midwest so I’m at home now just going to bed. Thanks.
OK, I got a chance to look at this issue and the reason he is seeing the ‘Stale’ quality more often is because Ignition is a little more strict than it was in 7.1.x.
We looked at the diagnostics for the devices that were showing stale data (in the gateway under Configure->OPC-UA->Devices) and the request throughput was much lower than the number of requests. This means that the device or the network are running slow. The way to fix this problem is to look into why your network is taking so long.
If you can’t do that, one way to stop seeing the stale overlays is to go into the OPC-UA settings and increase the Stale Threshold. This will only cover up the problem though. Your other option in Ignition is to reduce the scan class speed to something your network can handle. IE: if you have 60 requests on a 1 second scan class and your device is seeing a throughput of 10/sec then you must change the scan class to 6 seconds.
Robert, I presently have a similar problem with stale data using 7.2.3 in windows 7 connected to a CompactLogix. I did not have this problem with the 7.2.0 version. I look at my requests vs. throughput and they are the same. I have tried deleting unused tags and increasing my scan time with no success. The whole reason I waited to upgrade was due to some of the communication problems mentioned in this forum. I see you mention a difference in 7.1 and 7.2 in the way stale data is processed. Is there a difference between 7.2.0 and 7.2.3 or is there possibly some other issue?
Go to the [diagnostics] link for the device that has the stale tags on the device list page. It’s under “Ignition OPC-UA > Devices” on the gateway. Can you post a screenshot of that page or call in and let us take a look?
Kevin, I’ll be away from the plant for a while but I’ll see if I have the same issues on our test setup. I’ll send that if I do or I’ll get you something when I return to the facility. The PC doesn’t currently have internet connectivity but maybe they can hook it up for a while for a quick gotomeeting.
Here’s a snapshot of the diagnostic page captured using our test setup. The only items on this network are the Ignition PC and the PLC.
I don’t know if this should be tied to this thread or a new one started but we had a incident over the weekend that involves networking with Ignition. One of our contrologix would to connect to ignition or rslinx. Maintainance said everything was working fine on the floor. This plc talks to4 rsview hmi’s and to 2 other contrologix. I was able to ping the ethernet card and to go onto it in a web browser and it said evething was good. I replace the ethernet and lost comms to the 2 contrologix processor’s and 1 hmi. I downloaded a copy of wireshark onto the ignition pc and ran a 30 second scan of the network and found that ignition was just pounding on the processor that wouldn’t connect. I disabled that connection in ignition and everything came back to normal. Later on in the day they thought that they were having the same issues with a compactlogix and my boss had all contrologix and compactlogix disabled in ignition. I am to look at it tonght. I’ll post more if I know more. Paul
This is known about and will be resolved in the 7.2.4 release.
thanks when is it due out?
Shooting for this week. Thursday perhaps?
Kevin, will that take care of the issue I am having also?