aarch64_linux build test ant task failed with ArrayIndexOutOfBoundsException in some machines

[JasonFengJ9]

Failure link

From an internal build(cent7-aarch64-4):

08:27:40  openjdk version "21-internal" 2023-09-19
08:27:40  OpenJDK Runtime Environment (build 21-internal-adhoc.jenkins.BuildJDK21aarch64linuxPersonal)
08:27:40  Eclipse OpenJ9 VM (build master-e57170e15e5, JRE 21 Linux aarch64-64-Bit Compressed References 20230903_60 (JIT enabled, AOT enabled)
08:27:40  OpenJ9   - e57170e15e5
08:27:40  OMR      - 75d233597f4
08:27:40  JCL      - fce06129139 based on jdk-21+35)

Rerun in Grinder - Change TARGET to run only the failed test targets.

Optional info

Failure output (captured from console output)

08:30:00  Caught an exception while logging the end of the build.  Exception was:
08:30:00  java.lang.ExceptionInInitializerError
08:30:00  	at java.base/java.lang.J9VMInternals.recordInitializationFailure(J9VMInternals.java:175)
08:30:00  	at org.apache.tools.ant.DefaultLogger.formatTime(DefaultLogger.java:304)
08:30:00  	at org.apache.tools.ant.DefaultLogger.buildFinished(DefaultLogger.java:175)
08:30:00  	at org.apache.tools.ant.Project.fireBuildFinished(Project.java:2089)
08:30:00  	at org.apache.tools.ant.Main.runBuild(Main.java:852)
08:30:00  	at org.apache.tools.ant.Main.startAnt(Main.java:223)
08:30:00  	at org.apache.tools.ant.launch.Launcher.run(Launcher.java:284)
08:30:00  	at org.apache.tools.ant.launch.Launcher.main(Launcher.java:101)
08:30:00  Caused by: java.lang.ArrayIndexOutOfBoundsException
08:30:00  	at java.base/sun.util.calendar.BaseCalendar.getCalendarDateFromFixedDate(BaseCalendar.java:453)
08:30:00  	at java.base/java.util.GregorianCalendar.computeFields(GregorianCalendar.java:2369)
08:30:00  	at java.base/java.util.GregorianCalendar.computeTime(GregorianCalendar.java:2787)
08:30:00  	at java.base/java.util.Calendar.updateTime(Calendar.java:3427)
08:30:00  	at java.base/java.util.Calendar.getTimeInMillis(Calendar.java:1814)
08:30:00  	at java.base/java.util.Calendar.getTime(Calendar.java:1785)
08:30:00  	at java.base/java.text.SimpleDateFormat.initializeDefaultCentury(SimpleDateFormat.java:905)
08:30:00  	at java.base/java.text.SimpleDateFormat.initialize(SimpleDateFormat.java:670)
08:30:00  	at java.base/java.text.SimpleDateFormat.<init>(SimpleDateFormat.java:628)
08:30:00  	at org.apache.tools.ant.util.DateUtils.<clinit>(DateUtils.java:72)
08:30:00  	... 7 more
08:30:00  1
08:30:00  compile.mk:45: recipe for target 'compile' failed

Similar failure jobs:
JDK21 aarch64_linux(rhel8-aarch64-2)
JDK21 aarch64_linux(ub20-aarch64-4)
JDK21 aarch64_linux(cent7-aarch64-6)
JDK17 aarch64_linux(rhel8-aarch64-1)

Pass jobs:
JDK21 aarch64_linux(sles15-aarch64-2)
JDK21 aarch64_linux(ub22-aarch64-5)
JDK21 aarch64_linux(rhel9-aarch64-1)

The Ant version is Apache Ant(TM) version 1.10.5 compiled on July 10 2018.
Re-running the job in the same machine seems failed again, so this appears to a machine-related instead of an intermittent issue.

[pshipton]

We don't see this on the OpenJ9 machines. The latest build is https://openj9-jenkins.osuosl.org/job/Pipeline_Build_Test_JDK21_aarch64_linux/139/ and it's fine.
Since this seems to be an internal machine problem, not sure why it's being tracked here.

[Akira1Saitoh]

I think this failure is the same problem as #18057. This build does not include eclipse-omr/omr#7104.

08:27:40 OMR - 75d233597f4

[JasonFengJ9]

Latest run JDK21 aarch64_linux(sles15-aarch64-2) did pass

20:18:53  openjdk version "21-internal" 2023-09-19
20:18:53  OpenJDK Runtime Environment (build 21-internal-adhoc.jenkins.BuildJDK21aarch64linuxPersonal)
20:18:53  Eclipse OpenJ9 VM (build master-d3388e1662d, JRE 21 Linux aarch64-64-Bit Compressed References 20230905_61 (JIT enabled, AOT enabled)
20:18:53  OpenJ9   - d3388e1662d
20:18:53  OMR      - 873ac5d377a
20:18:53  JCL      - 154f45ddce4 based on jdk-21+35)

Will see if it passes in previously failed machines.

[JasonFengJ9]

The ArrayIndexOutOfBoundsException wasn't reproduced in previously failed machines w/ latest JDK21 aarch64_linux as per #18082 (comment)
ub20-aarch64-4
rhel8-aarch64-2
cent7-aarch64-6
rhel8-aarch64-1

Thanks @Akira1Saitoh