Our wrapper logs are being inundated with messages related to something with the tags that shouldBeEnabled. I plan to restart the Gateway to try to fix it when we have a break in production, but does anyone know what might be causing this?
There is probably a StackOverflowError at the head of each batch of lines - this looks like a classic recursive function call. What version of Ignition is this? (This is unambiguously an Ignition bug, but it's possible we've already fixed it).
We're running 8.1.24. Also, eventually the logs did load on the gateway, and all of them said, "Uncaught exception executing batch context."
I'd send your logs in to support. Like I said, definitely a bug in Ignition, and the nature of it should be clear enough from the logs. There may or may not be a known workaround.
I'll try to get the logs for you tomorrow. Just as an FYI, I rebooted the server, and the issue started happening right away again. I then reinstalled 8.1.24 on the server, and it is still happening. However, it is not happening on the backup server, so I'm wondering if there is some issue with the server. Also, we are running 8.1.24 on 7 other gateways without any issues. We run all of our Ignition servers on Ubuntu.
Have you found what was causing these messages?
We are experiencing the exact same issue.
INFO | jvm 1 | 2023/08/04 08:46:55 | E [t.e.batchoperations ] [12:46:55]: Uncaught exception executing batch context.
INFO | jvm 1 | 2023/08/04 08:46:55 | java.lang.StackOverflowError: null
INFO | jvm 1 | 2023/08/04 08:46:55 | at com.inductiveautomation.ignition.gateway.tags.runtime.nodes.BasicTagDefinition.shouldBeEnabled(BasicTagDefinition.java:362)
INFO | jvm 1 | 2023/08/04 08:46:55 | at com.inductiveautomation.ignition.gateway.tags.runtime.nodes.FolderNode.shouldBeEnabled(FolderNode.java:482)
No, I don't think we discovered the root cause, and I'm almost certain the issue stopped on its own at some point.
1 Like
Did anyone get a fix on this? The message I am getting is the following: Uncaught exception executing batch context. I am currently experiencing this and logs and wrapper logs are filling up so quickly I can't even troubleshoot other problems. I'm currently on version 8.1.17. Should I maybe upgrade?
I haven't seen this happen since the last posts, and we are now running 8.1.28. I don't know if the issue is actually resolved in newer versions.
I assume this issue isn't resolved I am seeing this same problem on 8.1.32. Is there any word on this matter.
Infinite recursion (in java, not jython) is always a bug worthy of reporting to support. (This forum is not support.)
2 Likes
Just had this occur with 8.1.33. Same log messages; we restarted the gateway to resolve those issues (this problem used a fair amount of i/o, writing 200-300mb/s), but then the logs were corrupted (system log idb) on gateway restart. Next, while the gateway was shut down, I moved the log file (and journal) to a different folder and then started the gateway. It recreated the log files for me, and the system returned to normal operation with logging.
I'll mention that I worked with support to resolve this.
Also, that this has been happening more frequently in the last few months, and they're looking for examples of system behavior like this so they can recreate the bug and fix it.