How to resolve "Thread limit exceeded replacing blocked worker"?

[efenderbosch-atg]

java.util.concurrent.RejectedExecutionException: Thread limit exceeded replacing blocked worker
	at java.base/java.util.concurrent.ForkJoinPool.tryCompensate(ForkJoinPool.java:2000)
	at java.base/java.util.concurrent.ForkJoinPool.compensatedBlock(ForkJoinPool.java:3737)
	at java.base/java.util.concurrent.ForkJoinPool.managedBlock(ForkJoinPool.java:3723)
	at org.spockframework.runtime.SpockNode.sneakyInvoke(SpockNode.java:40)
	at org.spockframework.runtime.SpecNode.lambda$around$0(SpecNode.java:72)
	at org.spockframework.runtime.PlatformSpecRunner.lambda$createMethodInfoForDoRunSpec$0(PlatformSpecRunner.java:66)
	at org.spockframework.runtime.model.MethodInfo.invoke(MethodInfo.java:157)
	at org.spockframework.runtime.PlatformSpecRunner.invokeRaw(PlatformSpecRunner.java:442)
	at org.spockframework.runtime.PlatformSpecRunner.invoke(PlatformSpecRunner.java:425)
	at org.spockframework.runtime.PlatformSpecRunner.runSpec(PlatformSpecRunner.java:59)
	at org.spockframework.runtime.SpecNode.around(SpecNode.java:72)
	at org.spockframework.runtime.SpecNode.around(SpecNode.java:12)
	at java.base/java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:387)
	at java.base/java.util.concurrent.ForkJoinPool$WorkQueue.topLevelExec(ForkJoinPool.java:1312)
	at java.base/java.util.concurrent.ForkJoinPool.scan(ForkJoinPool.java:1843)
	at java.base/java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1808)
	at java.base/java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:188)

I've tried tweaking the thread configuration, but it keeps popping up. Current SpockConfig.groovy

runner {
    parallel {
        // see https://spockframework.org/spock/docs/2.3/parallel_execution.html
        enabled(true)
        defaultSpecificationExecutionMode('CONCURRENT')
        defaultExecutionMode('SAME_THREAD')

        // see spock.config.ParallelConfiguration
        int availableProcessors = Runtime.getRuntime().availableProcessors()
        // Spock defaults parallelism to availableProcessors - 2
        // In CI environments, be more conservative to avoid thread exhaustion
        int parallelism = System.getenv('CI')
            ? Math.max(1, availableProcessors - 3)
            : Math.max(1, availableProcessors - 2)

        // Spock defaults minimumRunnable to the same as parallelism
        int minimumRunnable = parallelism
        // Math.round(availableProcessors / 2).intValue()

        // Spock defaults maxPoolSize to 256 + parallelism
        int maxPoolSize = 256 + parallelism

        // Spock defaults corePoolSize to the same as parallelism
        int corePoolSize = parallelism

        // Spock defaults keepAliveSeconds to 30
        int keepAliveSeconds = System.getenv('SPOCK_THREAD_KEEP_ALIVE_SECONDS')?.toInteger() ?: 60

        custom(parallelism, minimumRunnable, maxPoolSize, corePoolSize, keepAliveSeconds)
    }
}

[Vampire]

Please disable the StackTrace filtering and get a more complete StackTrace.

[efenderbosch-atg]

For anyone else w/ a similar issue, this seems to have fixed it for us for now. Reduced parallelism in CI and increased corePoolSize.

runner {
    parallel {
        // see https://spockframework.org/spock/docs/2.3/parallel_execution.html
        enabled(true)
        defaultSpecificationExecutionMode('CONCURRENT')
        defaultExecutionMode('SAME_THREAD')

        // see spock.config.ParallelConfiguration
        int availableProcessors = Runtime.getRuntime().availableProcessors()
        // Spock defaults parallelism to availableProcessors - 2
        // In CI environments, be more conservative to avoid thread exhaustion
        int parallelism = System.getenv('CI')
            ? Math.max(1, (availableProcessors / 2).intValue())
            : Math.max(1, availableProcessors - 1)

        // Spock defaults minimumRunnable to the same as parallelism
        int minimumRunnable = System.getenv('CI') ? 1 : parallelism

        // Spock defaults maxPoolSize to 256 + parallelism
        int maximumSpares =
            System.getProperty('java.util.concurrent.ForkJoinPool.common.maximumSpares', '256').toInteger()
        int maxPoolSize = maximumSpares + parallelism

        // Spock defaults corePoolSize to the same as parallelism
        int corePoolSize = parallelism * 2

        // Spock defaults keepAliveSeconds to 30
        int keepAliveSeconds = System.getenv('SPOCK_THREAD_KEEP_ALIVE_SECONDS')?.toInteger() ?: 60

        custom(parallelism, minimumRunnable, maxPoolSize, corePoolSize, keepAliveSeconds)
    }
}