Ignition Server Memory Spike

The CPU Trend and Memory Trend on the Ignition Gateway are showing a large and inconsistent spike, resulting in around 10 minutes of downtime after each spike. Logs aren’t showing any immediate causes. Looking for solution.
Operating System: Windows Server 2008 R2 amd 64
Java Version 1.8.0_171-b11
Ignition Version 7.9.6
are what is being used

I suggest, better provide OS, Java version, Ignition version and wrapper.log file to speed up response from Ignition support.

1 Like

Switch to G1GC. Turn on logging of garbage collector performance. Look for pause-the-world events (aka evacuation pauses) in the logs. See this post for details:

1 Like

Don’t mean to revive a dead topic but how do you turn on logging of garbage collection in Ignition? I’ve seen you mention it a few times but I can’t find anything in the docs about it.

1 Like

See wrapper.java.additional.3, 6, 7, & 8 in this post:

Also see the the section shown below here:

2 Likes

Note that those java -XX options have changed in current java versions (9+). This is what I’m using in Ignition 8.0.0:

wrapper.java.additional.1=-XX:+UseG1GC
wrapper.java.additional.2=-XX:MaxGCPauseMillis=100
wrapper.java.additional.3=-Xlog:gc*:file=/var/log/ignition/javagc-%WRAPPER_TIME_YYYYMMDDHHIISS%.log:t,tm:filecount=5,filesize=16m
wrapper.java.additional.4=-Ddata.dir=/usr/share/ignition/data
wrapper.java.additional.5=-Dorg.apache.catalina.loader.WebappClassLoader.ENABLE_CLEAR_REFERENCES=false
wrapper.java.additional.6=-Dignition.developer.moduleupload=true

{ Adjust folder names to suit your actual install, of course. }

3 Likes