You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
When I run the python integration tests on Spark 3.2.0 or spark_3.3.0
env -u SPARK_CONF_DIR SPARK_HOME=~/spark_3.2.0/ ./run_pyspark_from_build.sh -k 'parquet and encryption'
Most of them fail with an error like
Caused by: java.lang.ClassNotFoundException: Class org.apache.parquet.crypto.keytools.mocks.InMemoryKMS not found
at org.apache.hadoop.conf.Configuration.getClassByName(Configuration.java:2571)
at org.apache.parquet.hadoop.util.ConfigurationUtil.getClassFromConfig(ConfigurationUtil.java:33)
All I did was to run buildall.sh and then cd into the integration tests directory and run the tests.
I looks like
@pytest.mark.skipif(os.environ.get('INCLUDE_PARQUET_HADOOP_TEST_JAR', 'false') == 'false', reason='INCLUDE_PARQUET_HADOOP_TEST_JAR is disabled')
is the culprit. It is being set to true in the run_pyspark_from_build.sh without ever even checking if the hadoop dependency was placed where it is expected to be. But honestly I have no idea because I don't know what jar is expected to be on the classpath. It could be the fact that I build all of the versions and somehow the wrong dependency was downloaded. Either way this is a bad user experience.
The text was updated successfully, but these errors were encountered:
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs3-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs0-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs1-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs2-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs5-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs2-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs4-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs3-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs0-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs1-] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs4-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o310.parquet.
FAILED ../../src/main/python/parquet_test.py::test_parquet_read_encryption[reader_confs5-parquet] - py4j.protocol.Py4JJavaError: An error occurred while calling o400.parquet.
Describe the bug
When I run the python integration tests on Spark 3.2.0 or spark_3.3.0
Most of them fail with an error like
All I did was to run buildall.sh and then cd into the integration tests directory and run the tests.
I looks like
is the culprit. It is being set to true in the run_pyspark_from_build.sh without ever even checking if the hadoop dependency was placed where it is expected to be. But honestly I have no idea because I don't know what jar is expected to be on the classpath. It could be the fact that I build all of the versions and somehow the wrong dependency was downloaded. Either way this is a bad user experience.
The text was updated successfully, but these errors were encountered: