We are running Ignition version 7.6.3 (b2013090513) on Windows Server 2008 R2. We are using the Mobile module (1.6.3) to continuously display a screen on a thin client PC. At least once a day, we cannot access Ignition (both the web page and clients) and have to restart the service on the server to get it working again. The error message from the wrapper.log that we have seen each time this happens is:
INFO | jvm 1 | 2013/10/09 15:39:30 | Oct 09, 2013 3:39:30 PM org.apache.tomcat.util.net.JIoEndpoint createWorkerThread
INFO | jvm 1 | 2013/10/09 15:39:30 | INFO: Maximum number of threads (300) created for connector with address null and port 80
Are there any suggestions on something we should look at or check?
Can you zip up and attach your wrapper.log files as well? Sometimes these can contain a little bit of extra information…
Any idea what the memory usage is like when you have to restart it? How many mobile clients are open at that point?
I’m not sure what the specific memory usage is like at that point, but Windows showed that it still had plenty of memory overhead overall. There is only 1 mobile client connected at this point, but it is connected continuously (24/7). At the point where it crashes, that mobile client generally has been connected for 12-24 hours (depending on how long since the last crash).
We’ve also tried restarting via the Launch Gateway Control Utility and it will say it stops both the Web Server and the Gateway, but afterwards, the Gateway status will go back and forth between running and stopped and will not start completely and the Web Server will stay stopped. If we restart the Windows Service, it will restart fully functional.
So this is somehow causing the gateway web interface to be unresponsive also?
Do you think you can open the GCU and get a thread dump from there the next time this happens?
Yeah, the web interface is inaccessible too. Well, I say too, but I don’t know what is actually causing this, it’s just what I’m aware of that changed was we started using the mobile module.
One of the other developers just informed me this issue is known and has been fixed already for the next 7.6.4 release candidate.
Is that the 220.127.116.113100210-rc1 version, or is another going to follow with this fix? Would you recommend updating to the release candidate or waiting for the stable version?
It’s going to be the next one, 7.6.4-rc2.
The release candidates are usually pretty stable, but I think in this case you might be able to get away with just using the rc2 mobile module when it’s released.
Code has been frozen and merged… it’s in QA’s hands right now. Should probably be available middle-to-end of next week. If it’s urgent I could probably get you a build of just the mobile module that includes the fix today some time - email support at inductiveautomation.com if you want it.