Stephan Schroevers

> I'm sure @ben-manes can explain better since he wrote it I'm sure he can explain it (:smile:), but he's not the author; @sheepdreamofandroids is. See #336.

> How do I do that? Setting MAVEN_OPTS maybe? If I try to set up: > > `set MAVEN_OPTS="-Xmx1024M, -XX:ReservedCodeCacheSize=1024M,-XX:-UseGCOverheadLimit" ` > > I get an error complaining about the...

Then I wonder whether there are any non-visible/non-space characters in there. For example, with both JDK 11 and 17 I get: ``` $ MAVEN_OPTS="-XmxFOO -XX:ReservedCodeCacheSize=1024M -XX:-UseGCOverheadLimit" mvn clean Invalid maximum...

Hmm, afaik Surefire doesn't honor this flag either. It's documentation does [mention](http://maven.apache.org/surefire/maven-surefire-plugin/test-mojo.html#forkCount) it in the context of the `forkCount` property, but only in the context of parallel Surefire executions, not...

@spinnakerbot remove-label to-be-closed

A release with this change would help us, as our build enforces [`requireUpperBoundDeps`](https://maven.apache.org/enforcer/enforcer-rules/requireUpperBoundDeps.html). If we explicitly downgrade OkHttp to avoid an alpha release (which we don't want to use in...

I filed #839, which should resolve this issue.

It would be nice if one of the maintainers had a look at #839.

@elucash thanks for jumping on this! I just ran a test against a0010e93e41402c33f45d01befc433662c385c2e and heap space ended up at ~560MB while meta space ended up at 105MB:  So your...

Performed a new run based on ca0fa7ca6d1b4533815333c6ad0e4512997f1c3e, and the results look very promising: - Meta space ended up at ~80MB (~30% reduction compared to version 2.8.8!). - Heap space ended...