Skip to content

Conversation

@Mulily0513
Copy link
Collaborator

…pectations

  • add bench_test and pxf_extension_test in run_tests.sh, plus matrix entries for bench and pxf_extension in CI
  • bump surefire heap to 4G to avoid OOM
  • update gpupgrade expected outputs to new PXF_HOME paths and JSON formatter error text
  • make ProtocolUtils/HiveBaseTest/JdbcHiveTest/OrcWriteTest/ParquetWriteTest more robust to env defaults (protocol, creds, hive JDBC URL)
  • keep MultiServerTest running under HDFS with a safe working directory fallback
  • set distribution key and INSERT pattern for performance test data load

fix #ISSUE_Number


Change logs

Describe your change clearly, including what problem is being solved or what document is being added or updated.

Contributor's checklist

Here are some reminders before you submit your pull request:

  • Make sure that your Pull Request has a clear title and commit message. You can take the Git commit template as a reference.
  • Sign the Contributor License Agreement as prompted for your first-time contribution (One-time setup).
  • Learn the code contribution and doc contribution guides for better collaboration.
  • List your communications in the GitHub Issues or Discussions (if has or needed).
  • Feel free to ask for the cloudberrydb team to help review and approve.

else
export JAVA_HOME=$(readlink -f /usr/bin/java | sed 's:/bin/java::')
fi
export JAVA_HOME=/usr/lib/jvm/java-8-openjdk
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

delete Auto-detect Java 8 path for different architectures?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If the /usr/lib/jvm/java-8-openjdk/ directory indeed contains JDK versions for multiple architectures, then the JDK should automatically select the appropriate version based on the architecture.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I mean this commit gphd-conf.sh is deleted automatically, set export JAVA_HOME=/usr/lib/jvm/java-8-openjdk. I mean it shouldn't be removed.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok

return System.getProperty(AWS_ACCESS_KEY_ID);
String access = System.getProperty(AWS_ACCESS_KEY_ID);
String result = access != null ? access : System.getenv(AWS_ACCESS_KEY_ID);
System.out.println("DEBUG: doInit() method called getAccess " + result);
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is a debug print?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, this is a debug print. It should be removed.

hdfs.getWorkingDirectory() + "/" + fileName);
// load to hive table
hive.loadData(tableName, hdfsPath, false);
hive.loadData(tableName, hdfs.getWorkingDirectory() + "/" + fileName, false);
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sometimes HDFS metadata loading isn't that fast. Would it be better to add hdfs.waitForFile? ensure data load successfully?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok

- s3
- features
- gpdb
- bench
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bench runs for 5 hours. Not sure that we should run it on every PR.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@tuhaihe bench runs for 5 hours, we need to distinguish between functional testing and performance testing?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

5 hours feels too long for regular PR validation. I don't think it’s necessary to run this on every PR.

It might make more sense to separate performance/benchmark testing from functional testing, and run the benchmark as a scheduled job (for example, weekly) instead of blocking day-to-day PRs.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So we can remove it from this PR? How about it?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I can try to further split the load tests and performance tests first and see how long each part takes. Then we can decide the best way to run them.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since the load test takes about 20 minutes and the performance tests take over 2 hours, we could keep just the load test for regular PR validation and move the performance tests to a separate schedule.

@Mulily0513 Mulily0513 force-pushed the merge-with-upstream_hlhf branch 4 times, most recently from bc7f101 to d474dce Compare December 29, 2025 07:16
…pectations

- add bench_test and pxf_extension_test in run_tests.sh, plus matrix entries for bench and pxf_extension in CI
- bump surefire heap to 4G to avoid OOM
- update gpupgrade expected outputs to new PXF_HOME paths and JSON formatter error text
- make ProtocolUtils/HiveBaseTest/JdbcHiveTest/OrcWriteTest/ParquetWriteTest more robust to env defaults (protocol, creds, hive JDBC URL)
- keep MultiServerTest running under HDFS with a safe working directory fallback
- set distribution key and INSERT pattern for performance test data load
@Mulily0513 Mulily0513 force-pushed the merge-with-upstream_hlhf branch from d474dce to 17c8ad6 Compare December 29, 2025 08:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants