This closes #11241

Test scope is not included in this fix.

This closes #10976

support vectorized read for hive 1.x.

closes #10730

If the module was built beforehand a pom.xml would be detected in the target directory.

Adding the e2e-pre-commit profile activation only to the misc profile broke the caching

Azure exhibits some problems when using printf.

- consolidate travis activations in .travis.yml
- introduce dedicated profile for each category to reduce verbosity
- use gmavenplus-plugin to merge inclusion/exclusions across profiles
- add java11 profile (i.e., don't rely on exclusion setup in root pom)
- disable tests with Hadoop category by default
- add Hadoop activations to e2e profiles

This closes #11023

This closes #10951.

If the StreamTask is being cancelled, let the Task handle the exception.

This test does not rely on job execution success, but verifies the correctness in the test scripts.
The occastional exceptions are unrelated and result of the cancellation process, hence can be ignored.
(Cleaning these log pollution exceptions up is a separate unrelated issue)

Some cleanups in RocksDBStateMemoryControlTestProgram
  - skip ".returns" when not necessary
  - peel off command line parameter abstraction earlier (don't pass ParameterTool to functions).
  - Add proper class-level JavaDoc.

Remove script from 'run-nightly-tests.sh' (unused) and add to a different CRON split file instead.

Some improvements to the test script
  - Added comment explainig where the upper limit value comes from
  - Added fixed managed memory size, so the test is not affected by changes in the managed memory fraction settings
  - Always log measured cache size for information

[FLINK-15638][release][python] Change version of pyflink to the release version when creating release branch

This closes #10876.

The partial compilation in nightly only compiles flind-end-to-end-tests
for performance reasons. However, that resulted in outdated walkthrough
archetypes among other issues.

This closes #10665.

[FLINK-15338][python] Cherry-pick BEAM-9006#10462 to fix the TM Metaspace memory leak problem when submitting PyFlink UDF jobs multiple times.

This closes #10772.

[FLINK-15338][python] Cherry-pick NETTY#8955 to fix the TM Metaspace memory leak problem in shaded netty when submitting PyFlink UDF jobs multiple times.

This closes #10666.