Disable container support in opensearch test containers

[jaydeluca]

There have been a lot of flaky opensearch tests today (example)

I figured we could try disabling container support (-XX:-UseContainerSupport) and Log4j JMX to fix some NullPointerExceptions in CgroupV2Subsystem.getInstance() during container startup.

Not sure if this will fix it, but didnt seem like it could hurt. Seems like it helped for someone else

Hopefully resolves #15024

Some stack traces:

2025-10-16 20:04:05,784 main ERROR Could not reconfigure JMX java.lang.NullPointerException |  
-- | --
at java.base/jdk.internal.platform.cgroupv2.CgroupV2Subsystem.getInstance(CgroupV2Subsystem.java:81) |  
at java.base/jdk.internal.platform.CgroupSubsystemFactory.create(CgroupSubsystemFactory.java:113) |  
at java.base/jdk.internal.platform.CgroupMetrics.getInstance(CgroupMetrics.java:167) |  
at java.base/jdk.internal.platform.SystemMetrics.instance(SystemMetrics.java:29) |  
at java.base/jdk.internal.platform.Metrics.systemMetrics(Metrics.java:58) |  
at java.base/jdk.internal.platform.Container.metrics(Container.java:43) |  
at jdk.management/com.sun.management.internal.OperatingSystemImpl.(OperatingSystemImpl.java:182) |  
at jdk.management/com.sun.management.internal.PlatformMBeanProviderImpl.getOperatingSystemMXBean(PlatformMBeanProviderImpl.java:281) |  
at jdk.management/com.sun.management.internal.PlatformMBeanProviderImpl$3.nameToMBeanMap(PlatformMBeanProviderImpl.java:198) |  
at java.management/java.lang.management.ManagementFactory.lambda$getPlatformMBeanServer$0(ManagementFactory.java:487) |  
at java.base/java.util.stream.ReferencePipeline$7$1.accept(ReferencePipeline.java:271) |  
at java.base/java.util.stream.ReferencePipeline$2$1.accept(ReferencePipeline.java:177) |  
at java.base/java.util.HashMap$ValueSpliterator.forEachRemaining(HashMap.java:1693) |  
at java.base/java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:484) |  
at java.base/java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:474) |  
at java.base/java.util.stream.ForEachOps$ForEachOp.evaluateSequential(ForEachOps.java:150) |  
at java.base/java.util.stream.ForEachOps$ForEachOp$OfRef.evaluateSequential(ForEachOps.java:173) |  
at java.base/java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234) |  
at java.base/java.util.stream.ReferencePipeline.forEach(ReferencePipeline.java:497) |  
at java.management/java.lang.management.ManagementFactory.getPlatformMBeanServer(ManagementFactory.java:488) |  
at org.apache.logging.log4j.core.jmx.Server.reregisterMBeansAfterReconfigure(Server.java:140)

fatal error in thread [main], exiting |  
-- | --
java.lang.ExceptionInInitializerError |  
at org.opensearch.bootstrap.Bootstrap.initializeProbes(Bootstrap.java:177) |  
at org.opensearch.bootstrap.Bootstrap.setup(Bootstrap.java:199) |  
at org.opensearch.bootstrap.Bootstrap.init(Bootstrap.java:412) |  
at org.opensearch.bootstrap.OpenSearch.init(OpenSearch.java:178) |  
at org.opensearch.bootstrap.OpenSearch.execute(OpenSearch.java:169) |  
at org.opensearch.cli.EnvironmentAwareCommand.execute(EnvironmentAwareCommand.java:100) |  
at org.opensearch.cli.Command.mainWithoutErrorHandling(Command.java:138) |  
at org.opensearch.cli.Command.main(Command.java:101) |  
at org.opensearch.bootstrap.OpenSearch.main(OpenSearch.java:135) |  
at org.opensearch.bootstrap.OpenSearch.main(OpenSearch.java:101) |  
Caused by: java.lang.NullPointerException |  
at java.base/jdk.internal.platform.cgroupv2.CgroupV2Subsystem.getInstance(CgroupV2Subsystem.java:81) |  
at java.base/jdk.internal.platform.CgroupSubsystemFactory.create(CgroupSubsystemFactory.java:113) |  
at java.base/jdk.internal.platform.CgroupMetrics.getInstance(CgroupMetrics.java:167) |  
at java.base/jdk.internal.platform.SystemMetrics.instance(SystemMetrics.java:29) |  
at java.base/jdk.internal.platform.Metrics.systemMetrics(Metrics.java:58) |  
at java.base/jdk.internal.platform.Container.metrics(Container.java:43) |  
at jdk.management/com.sun.management.internal.OperatingSystemImpl.(OperatingSystemImpl.java:182) |  
at jdk.management/com.sun.management.internal.PlatformMBeanProviderImpl.getOperatingSystemMXBean(PlatformMBeanProviderImpl.java:281) |  
at jdk.management/com.sun.management.internal.PlatformMBeanProviderImpl$3.nameToMBeanMap(PlatformMBeanProviderImpl.java:198) |  
at java.management/sun.management.spi.PlatformMBeanProvider$PlatformComponent.getMBeans(PlatformMBeanProvider.java:195) |  
at java.management/java.lang.management.ManagementFactory.getPlatformMXBean(ManagementFactory.java:686) |  
at java.management/java.lang.management.ManagementFactory.getOperatingSystemMXBean(ManagementFactory.java:388) |  
at org.opensearch.monitor.process.ProcessProbe.(ProcessProbe.java:46) |  
... 10 more

[jaydeluca]

i dont know what version of log4j is in use, so added both configs
https://logging.apache.org/log4j/2.x/manual/jmx.html
https://logging.apache.org/log4j/2.3.x/manual/jmx.html

[trask]

@jaydeluca heads up looks like also impacting elasticsearch

[jaydeluca]

looking at some of the build failures, and researching this more, it might have to do with the linux kernel version in use. It looks like tests are passing on 6.11.x (example success - Linux 6.11.0-1018-azure) but not higher than that (example failure - Linux 6.14.0-1012-azure)

More resources:

• Java application in container killed after upgrading from 4.37.2 to 4.38.0 docker/for-mac#7573
• https://stackoverflow.com/questions/71532170/java-lang-nullpointerexception-cannot-invoke-jdk-internal-platform-cgroupinfo/71585718#71585718