build-alibaba-ubuntu1804-armv8-1 periodically hangs test testJCMMXBeanRemote_1

[andrew-m-leonard]

Please set the title to indicate the test name and machine name where known.

To make it easy for the infrastructure team to repeat and diagnose, please answer the following questions:

• test suite/name? Test_openjdk17_j9_extended.functional_aarch64_linux testJCMMXBeanRemote_1
• Is there an existing issue elsewhere covering this? https://github.com/eclipse/openj9/issues/12189
• Which machine(s) does it work on? build-docker-ubuntu1804-armv8-3
• Which machine(s) does it fail on? build-alibaba-ubuntu1804-armv8-1
• Do you have a link to a Grinder re-run if the test with the failure? https://ci.adoptopenjdk.net/job/Grinder/8023/console

Any other details:

[sxa]

There's something very odd on this host, over 2000 bash processes all dated March 24th like this:

jenkins  32591  7358  0 Mar24 ?        00:00:00 [bash] <defunct>

I'll kill your Grinder, try and clear it up, then run it again

[sxa]

@joeyleeeeeee97 Are your arm64 machines that you've supplied to us really docker containers? These defunct bash processes appear to have parent processes which suggest so. I was going to try and reboot the machine in order to try and clear this, but if it's a container attempting it could have unexpected side effects.

[sxa]

The machine has now been replaced, but I'm going to kick off a run at https://ci.adoptopenjdk.net/view/Test_functional/job/Test_openjdk17_j9_extended.functional_aarch64_linux/85/ on the -2 machine to see if it shows the same problem (@andrew-m-leonard do you have the correct TARGET parameters for this that I can stick into a Grinder to avoid having to run the whole suite?)

[sxa]

Machines are currently offine therefore not causing any failures :-) Unassigning myself for now - if the machines return we can revisit this.