This is an automated email from the ASF dual-hosted git repository.
yangjie01 pushed a commit to branch branch-4.0
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/branch-4.0 by this push:
new a547d4d08491 [SPARK-51876][PYTHON][TESTS] Add configuration for
`log4j.configurationFile` in the PySpark submission args of
`run_individual_python_test`
a547d4d08491 is described below
commit a547d4d08491f9bc371db575ed3c4d7c02d43845
Author: yangjie01 <yangji...@baidu.com>
AuthorDate: Wed Apr 23 19:18:26 2025 +0800
[SPARK-51876][PYTHON][TESTS] Add configuration for
`log4j.configurationFile` in the PySpark submission args of
`run_individual_python_test`
### What changes were proposed in this pull request?
This PR adds the configuration for `log4j.configurationFile` in the PySpark
submission args of the `run_individual_python_test` function. This allows the
Java processes initiated by the submitted PySpark jobs to use a reasonable
logging level.
### Why are the changes needed?
Prevent the Java processes corresponding to the PySpark jobs submitted by
`run_individual_python_test` from using unexpected logging levels and avoid
disk wastage in GA tasks.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
- Pass GitHub Actions
- Locally test:
run
```
python/run-tests --testnames
pyspark.sql.tests.pandas.test_pandas_transform_with_state
--python-executables=python3.11
```
**Before**
The Java process corresponding to the PySpark job unexpectedly uses the
log4j test configuration in the Scala test directory.
We can see the submission command for the Java process as follows:
```
/Users/yangjie01/Tools/zulu17/bin/java -cp
hive-jackson/*:/Users/yangjie01/spark/conf/:...:/Users/yangjie01/spark/common/kvstore/target/scala-2.13/test-classes/:/Users/yangjie01/spark/common/network-common/target/scala-2.13/test-classes/:/Users/yangjie01/spark/common/network-shuffle/target/scala-2.13/test-classes/:...:/Users/yangjie01/spark/assembly/target/scala-2.13/jars/*
-Xmx1g
-Djava.io.tmpdir=/Users/yangjie01/spark/python/target/7836a5a1-a39e-4b31-8ab1-f14b901668a4
-Xss4M -XX:+Ig [...]
```
The Java process launched by this PySpark job uses the log4j configuration
located in the `common/kvstore/target/scala-2.13/test-classes` directory. The
`rootLogger.level` in this configuration file is set to `debug`, which is why
we see a large amount of DEBUG logs in the `unit-tests.log`.
https://github.com/apache/spark/blob/master/common/kvstore/src/test/resources/log4j2.properties
```
25/04/23 17:02:04.688 main DEBUG MutableMetricsFactory: field
org.apache.hadoop.metrics2.lib.MutableRate
org.apache.hadoop.security.UserGroupInformation$UgiMetrics.getGroups with
annotation org.apache.hadoop.metrics2.annotation.Metric(sampleName="Ops",
always=false, valueName="Time", about="", interval=10, type=DEFAULT,
value={"GetGroups"})
25/04/23 17:02:04.694 main DEBUG MutableMetricsFactory: field
org.apache.hadoop.metrics2.lib.MutableRate
org.apache.hadoop.security.UserGroupInformation$UgiMetrics.loginFailure with
annotation org.apache.hadoop.metrics2.annotation.Metric(sampleName="Ops",
always=false, valueName="Time", about="", interval=10, type=DEFAULT,
value={"Rate of failed kerberos logins and latency (milliseconds)"})
25/04/23 17:02:04.694 main DEBUG MutableMetricsFactory: field
org.apache.hadoop.metrics2.lib.MutableRate
org.apache.hadoop.security.UserGroupInformation$UgiMetrics.loginSuccess with
annotation org.apache.hadoop.metrics2.annotation.Metric(sampleName="Ops",
always=false, valueName="Time", about="", interval=10, type=DEFAULT,
value={"Rate of successful kerberos logins and latency (milliseconds)"})
25/04/23 17:02:04.694 main DEBUG MutableMetricsFactory: field private
org.apache.hadoop.metrics2.lib.MutableGaugeInt
org.apache.hadoop.security.UserGroupInformation$UgiMetrics.renewalFailures with
annotation org.apache.hadoop.metrics2.annotation.Metric(sampleName="Ops",
always=false, valueName="Time", about="", interval=10, type=DEFAULT,
value={"Renewal failures since last successful login"})
25/04/23 17:02:04.695 main DEBUG MutableMetricsFactory: field private
org.apache.hadoop.metrics2.lib.MutableGaugeLong
org.apache.hadoop.security.UserGroupInformation$UgiMetrics.renewalFailuresTotal
with annotation org.apache.hadoop.metrics2.annotation.Metric(sampleName="Ops",
always=false, valueName="Time", about="", interval=10, type=DEFAULT,
value={"Renewal failures since startup"})
25/04/23 17:02:04.696 main DEBUG MetricsSystemImpl: UgiMetrics, User and
group related metrics
25/04/23 17:02:04.744 main DEBUG ShutdownHookManager: Adding shutdown hook
25/04/23 17:02:04.758 main DEBUG Shell: Failed to detect a valid hadoop
home directory
...
25/04/23 17:02:04.763 main DEBUG Shell: setsid is not available on this
machine. So not using it.
25/04/23 17:02:04.763 main DEBUG Shell: setsid exited with exit code 0
25/04/23 17:02:04.808 main DEBUG PythonGatewayServer: Started
PythonGatewayServer on localhost/127.0.0.1 with port 50689
25/04/23 17:02:04.921 Thread-2 INFO SparkContext: Running Spark version
4.1.0-SNAPSHOT
25/04/23 17:02:04.922 Thread-2 INFO SparkContext: OS info Mac OS X, 15.4,
aarch64
25/04/23 17:02:04.922 Thread-2 INFO SparkContext: Java version 17.0.14
25/04/23 17:02:04.929 Thread-2 DEBUG SecurityUtil: Setting
hadoop.security.token.service.use_ip to true
25/04/23 17:02:04.955 Thread-2 DEBUG Groups: Creating new Groups object
25/04/23 17:02:04.955 Thread-2 DEBUG NativeCodeLoader: Trying to load the
custom-built native-hadoop library...
...
```
**After**
The Java process corresponding to the PySpark job will use the log4j
configuration specified by `-Dlog4j.configurationFile`:
```
/Users/yangjie01/Tools/zulu17/bin/java -cp
hive-jackson/*:/Users/yangjie01/spark/conf/:...:/Users/yangjie01/spark/common/kvstore/target/scala-2.13/test-classes/:/Users/yangjie01/spark/common/network-common/target/scala-2.13/test-classes/:...:/Users/yangjie01/spark/core/target/jars/*:/Users/yangjie01/spark/mllib/target/jars/*:/Users/yangjie01/spark/assembly/target/scala-2.13/jars/slf4j-api-2.0.17.jar:/Users/yangjie01/spark/assembly/target/scala-2.13/jars/*
-Xmx1g -Djava.io.tmpdir=/User [...]
```
From `spark.*.extraJavaOptions`, we can see the configuration
`-Dlog4j.configurationFile=/Users/yangjie01/spark/python/test_support/log4j2.properties`.
At the same time, there are no longer DEBUG-level logs in unit-tests.log:
```
25/04/23 17:08:23.584 Thread-2 INFO SparkContext: Running Spark version
4.1.0-SNAPSHOT
25/04/23 17:08:23.588 Thread-2 INFO SparkContext: OS info Mac OS X, 15.4,
aarch64
25/04/23 17:08:23.588 Thread-2 INFO SparkContext: Java version 17.0.14
25/04/23 17:08:23.625 Thread-2 WARN NativeCodeLoader: Unable to load
native-hadoop library for your platform... using builtin-java classes where
applicable
25/04/23 17:08:23.655 Thread-2 INFO ResourceUtils:
==============================================================
25/04/23 17:08:23.655 Thread-2 INFO ResourceUtils: No custom resources
configured for spark.driver.
25/04/23 17:08:23.656 Thread-2 INFO ResourceUtils:
==============================================================
25/04/23 17:08:23.656 Thread-2 INFO SparkContext: Submitted application:
ReusedSQLTestCase
25/04/23 17:08:23.669 Thread-2 INFO ResourceProfile: Default
ResourceProfile created, executor resources: Map(cores -> name: cores, amount:
1, script: , vendor: , memory -> name: memory, amount: 1024, script: , vendor:
, offHeap -> name: offHeap, amount: 0, script: , vendor: ), task resources:
Map(cpus -> name: cpus, amount: 1.0)
25/04/23 17:08:23.670 Thread-2 INFO ResourceProfile: Limiting resource is
cpu
25/04/23 17:08:23.671 Thread-2 INFO ResourceProfileManager: Added
ResourceProfile id: 0
25/04/23 17:08:23.698 Thread-2 INFO SecurityManager: Changing view acls to:
yangjie01
25/04/23 17:08:23.699 Thread-2 INFO SecurityManager: Changing modify acls
to: yangjie01
25/04/23 17:08:23.699 Thread-2 INFO SecurityManager: Changing view acls
groups to: yangjie01
25/04/23 17:08:23.699 Thread-2 INFO SecurityManager: Changing modify acls
groups to: yangjie01
25/04/23 17:08:23.700 Thread-2 INFO SecurityManager: SecurityManager:
authentication disabled; ui acls disabled; users with view permissions:
yangjie01 groups with view permissions: EMPTY; users with modify permissions:
yangjie01; groups with modify permissions: EMPTY; RPC SSL disabled
25/04/23 17:08:23.823 Thread-2 INFO Utils: Successfully started service
'sparkDriver' on port 51764.
25/04/23 17:08:23.840 Thread-2 INFO SparkEnv: Registering MapOutputTracker
25/04/23 17:08:23.846 Thread-2 INFO SparkEnv: Registering BlockManagerMaster
25/04/23 17:08:23.855 Thread-2 INFO BlockManagerMasterEndpoint: Using
org.apache.spark.storage.DefaultTopologyMapper for getting topology information
25/04/23 17:08:23.856 Thread-2 INFO BlockManagerMasterEndpoint:
BlockManagerMasterEndpoint up
25/04/23 17:08:23.857 Thread-2 INFO SparkEnv: Registering
BlockManagerMasterHeartbeat
25/04/23 17:08:23.871 Thread-2 INFO DiskBlockManager: Created local
directory at
/Users/yangjie01/SourceCode/git/spark-sbt/python/target/37c5fe4f-2abd-4591-90bf-01b745c1af4c/blockmgr-790a3b12-ed2c-4c57-bfd1-eb88ac3d4ebf
...
```
### Was this patch authored or co-authored using generative AI tooling?
No
Closes #50675 from LuciferYang/pyspark-log4j.
Authored-by: yangjie01 <yangji...@baidu.com>
Signed-off-by: yangjie01 <yangji...@baidu.com>
(cherry picked from commit a9128a08c82d1fa847c231d655c1ff58baf022fd)
Signed-off-by: yangjie01 <yangji...@baidu.com>
---
python/run-tests.py | 7 +++++--
python/test_support/log4j2.properties | 31 +++++++++++++++++++++++++++++++
2 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/python/run-tests.py b/python/run-tests.py
index 64ac48e210db..d020e6d38108 100755
--- a/python/run-tests.py
+++ b/python/run-tests.py
@@ -117,8 +117,11 @@ def run_individual_python_test(target_dir, test_name,
pyspark_python, keep_test_
metastore_dir = os.path.join(metastore_dir, str(uuid.uuid4()))
os.mkdir(metastore_dir)
- # Also override the JVM's temp directory by setting driver and executor
options.
- java_options = "-Djava.io.tmpdir={0}".format(tmp_dir)
+ # Also override the JVM's temp directory and log4j conf by setting driver
and executor options.
+ log4j2_path = os.path.join(SPARK_HOME,
"python/test_support/log4j2.properties")
+ java_options = "-Djava.io.tmpdir={0} -Dlog4j.configurationFile={1}".format(
+ tmp_dir, log4j2_path
+ )
java_options = java_options + " -Xss4M"
spark_args = [
"--conf", "spark.driver.extraJavaOptions='{0}'".format(java_options),
diff --git a/python/test_support/log4j2.properties
b/python/test_support/log4j2.properties
new file mode 100644
index 000000000000..6629658c1d28
--- /dev/null
+++ b/python/test_support/log4j2.properties
@@ -0,0 +1,31 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements. See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+# Set everything to be logged to the file target/unit-tests.log
+rootLogger.level = info
+rootLogger.appenderRef.file.ref = File
+
+appender.file.type = File
+appender.file.name = File
+appender.file.fileName = target/unit-tests.log
+appender.file.append = true
+appender.file.layout.type = PatternLayout
+appender.file.layout.pattern = %d{yy/MM/dd HH:mm:ss.SSS} %t %p %c{1}: %m%n%ex
+
+# Silence verbose logs from 3rd-party libraries.
+logger.netty.name = io.netty
+logger.netty.level = info
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
