The following test fails in the Valhalla CI when --enabled-preview is used:

vmTestbase/nsk/jdb/exclude/exclude001/exclude001.java

Here's a snippet from the log file:

[13:59:56.922] reply[1]: Step completed: "thread=MyThread-0", nsk.jdb.exclude.exclude001.MyThread.run(), line=90 bci=3
[13:59:56.922] reply[2]:
[13:59:56.922] reply[3]: MyThread-0[1]
[13:59:56.922] Sending command: exclude java.*,javax.*,sun.*,com.sun.*,jdk.*
[13:59:57.125] reply[0]: MyThread-0[1]
[13:59:57.126] Sending command: trace methods 830
[13:59:57.342] reply[0]: MyThread-0[1]
[13:59:57.347] Sending command: cont
[14:10:12.284] WARNING: message not recieved: MyThread-0[1]
[14:10:12.284] Remaining debugger output follows:
[14:16:11.913] reply[0]: >
[14:16:11.913] reply[1]: Method exited: return value = <void value>, "thread=MyThread-0", nsk.jdb.exclude.exclude001.MyThread.run(), line=93 bci=14
[14:16:11.913] reply[2]:
[14:16:11.913] reply[3]: MyThread-0[1]
[14:16:11.932] # ERROR: Caught unexpected exception while executing the test: nsk.share.Failure: Expected message not received during 600200 milliseconds:
[14:16:11.933] # ERROR: MyThread-0[1]
The following stacktrace is for failure analysis.
nsk.share.TestFailure: Caught unexpected exception while executing the test: nsk.share.Failure: Expected message not received during 600200 milliseconds: MyThread-0[1]
at nsk.share.Log.logExceptionForFailureAnalysis(Log.java:309)
at nsk.share.Log.complain(Log.java:280)
at nsk.share.jdb.JdbTest.failure(JdbTest.java:72)
at nsk.share.jdb.JdbTest.runTest(JdbTest.java:157)
at nsk.jdb.exclude.exclude001.exclude001.main(exclude001.java:88)
at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
at java.base/java.lang.reflect.Method.invoke(Method.java:565)
at com.sun.javatest.regtest.agent.MainActionHelper$AgentVMRunnable.run(MainActionHelper.java:335)
at java.base/java.lang.Thread.run(Thread.java:1516)
nsk.share.Failure: Expected message not received during 600200 milliseconds:
MyThread-0[1]
at nsk.share.jdb.Jdb.waitForMessage(Jdb.java:551)
at nsk.share.jdb.Jdb.receiveReplyForWithMessageWait(Jdb.java:363)
at nsk.jdb.exclude.exclude001.exclude001.runCases(exclude001.java:158)
at nsk.share.jdb.JdbTest.runTest(JdbTest.java:148)
at nsk.jdb.exclude.exclude001.exclude001.main(exclude001.java:88)
at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
at java.base/java.lang.reflect.Method.invoke(Method.java:565)
at com.sun.javatest.regtest.agent.MainActionHelper$AgentVMRunnable.run(MainActionHelper.java:335)
at java.base/java.lang.Thread.run(Thread.java:1516)
[14:16:11.944] Waiting for jdb exits
[14:26:11.953] # ERROR: jdb did not exit after timeout.
[14:26:11.955] Sending quit command to jdb.
[14:26:11.956] Sending command: quit
[14:26:12.445] # ERROR: TEST FAILED

These are the critical lines that show the failure:

[13:59:57.347] Sending command: cont
[14:10:12.284] WARNING: message not recieved: MyThread-0[1]
[14:10:12.284] Remaining debugger output follows:
[14:16:11.913] reply[0]: >
[14:16:11.913] reply[1]: Method exited: return value = <void value>, "thread=MyThread-0", nsk.jdb.exclude.exclude001.MyThread.run(), line=93 bci=14
[14:16:11.913] reply[2]:
[14:16:11.913] reply[3]: MyThread-0[1]
[14:16:11.932] # ERROR: Caught unexpected exception while executing the test: nsk.share.Failure: Expected message not received during 600200 milliseconds:
[14:16:11.933] # ERROR: MyThread-0[1]

The debugger sends the "cont" command and 10 minutes later
the expected reply of "MyThread-0[1]" has not been received.
The reply that was received is dump in "reply[0]" thru "reply[3]"
and the last line ("reply[3]") does show the expected value.

Is this line in the reply:

[14:16:11.913] reply[1]: Method exited: return value = <void value>, "thread=MyThread-0", nsk.jdb.exclude.exclude001.MyThread.run(), line=93 bci=14

messing up the logic that's looking for "MyThread-0[1]" ?

There are currently 40 sightings of this test's failure mode in
the Valhalla CI and all of them are on windows-x64. In the
Tier7 that I'm currently analyzing (jdk-27-valhalla+1-20), this
test executes 20 times: 19 on linux-aarch64 and linux-x64 that
all pass and one execution on windows-x64 that failed.

The windows-x64 execution that failed is the only configuration
that was invoked with --enable-preview.

Update: However, of the 40 sightings of this test's failure mode
in the Valhalla CI, ONLY ONE of those failures specified the
--enable-preview option.