MAPREDUCE-5932. Provide an option to use a dedicated reduce-side shuffle log. Contributed by Gera Shegalov

This commit is contained in:
Jason Lowe 2014-12-03 17:02:14 +00:00
parent 22afae890d
commit 03ab24aa01
13 changed files with 243 additions and 56 deletions

View File

@ -235,6 +235,9 @@ Release 2.7.0 - UNRELEASED
IMPROVEMENTS IMPROVEMENTS
MAPREDUCE-5932. Provide an option to use a dedicated reduce-side shuffle
log (Gera Shegalov via jlowe)
OPTIMIZATIONS OPTIMIZATIONS
MAPREDUCE-6169. MergeQueue should release reference to the current item MAPREDUCE-6169. MergeQueue should release reference to the current item

View File

@ -20,16 +20,14 @@ package org.apache.hadoop.mapred;
import java.net.InetSocketAddress; import java.net.InetSocketAddress;
import java.util.HashMap; import java.util.HashMap;
import java.util.Iterator;
import java.util.List; import java.util.List;
import java.util.Map; import java.util.Map;
import java.util.Vector; import java.util.Vector;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.Path;
import org.apache.hadoop.mapred.TaskLog.LogName; import org.apache.hadoop.mapred.TaskLog.LogName;
import org.apache.hadoop.mapreduce.ID;
import org.apache.hadoop.mapreduce.MRJobConfig; import org.apache.hadoop.mapreduce.MRJobConfig;
import org.apache.hadoop.mapreduce.TypeConverter;
import org.apache.hadoop.mapreduce.v2.util.MRApps; import org.apache.hadoop.mapreduce.v2.util.MRApps;
import org.apache.hadoop.yarn.api.ApplicationConstants; import org.apache.hadoop.yarn.api.ApplicationConstants;
import org.apache.hadoop.yarn.api.ApplicationConstants.Environment; import org.apache.hadoop.yarn.api.ApplicationConstants.Environment;
@ -52,20 +50,6 @@ public class MapReduceChildJVM {
jobConf.get(JobConf.MAPRED_TASK_ENV)); jobConf.get(JobConf.MAPRED_TASK_ENV));
} }
private static String getChildLogLevel(JobConf conf, boolean isMap) {
if (isMap) {
return conf.get(
MRJobConfig.MAP_LOG_LEVEL,
JobConf.DEFAULT_LOG_LEVEL.toString()
);
} else {
return conf.get(
MRJobConfig.REDUCE_LOG_LEVEL,
JobConf.DEFAULT_LOG_LEVEL.toString()
);
}
}
public static void setVMEnv(Map<String, String> environment, public static void setVMEnv(Map<String, String> environment,
Task task) { Task task) {
@ -79,7 +63,7 @@ public class MapReduceChildJVM {
// streaming) it will have the correct loglevel. // streaming) it will have the correct loglevel.
environment.put( environment.put(
"HADOOP_ROOT_LOGGER", "HADOOP_ROOT_LOGGER",
getChildLogLevel(conf, task.isMapTask()) + ",console"); MRApps.getChildLogLevel(conf, task.isMapTask()) + ",console");
// TODO: The following is useful for instance in streaming tasks. Should be // TODO: The following is useful for instance in streaming tasks. Should be
// set in ApplicationMaster's env by the RM. // set in ApplicationMaster's env by the RM.
@ -147,15 +131,6 @@ public class MapReduceChildJVM {
return adminClasspath + " " + userClasspath; return adminClasspath + " " + userClasspath;
} }
private static void setupLog4jProperties(Task task,
Vector<String> vargs,
long logSize, Configuration conf) {
String logLevel = getChildLogLevel(task.conf, task.isMapTask());
int numBackups = task.conf.getInt(MRJobConfig.TASK_LOG_BACKUPS,
MRJobConfig.DEFAULT_TASK_LOG_BACKUPS);
MRApps.addLog4jSystemProperties(logLevel, logSize, numBackups, vargs, conf);
}
public static List<String> getVMCommand( public static List<String> getVMCommand(
InetSocketAddress taskAttemptListenerAddr, Task task, InetSocketAddress taskAttemptListenerAddr, Task task,
JVMId jvmID) { JVMId jvmID) {
@ -206,10 +181,7 @@ public class MapReduceChildJVM {
Path childTmpDir = new Path(MRApps.crossPlatformifyMREnv(conf, Environment.PWD), Path childTmpDir = new Path(MRApps.crossPlatformifyMREnv(conf, Environment.PWD),
YarnConfiguration.DEFAULT_CONTAINER_TEMP_DIR); YarnConfiguration.DEFAULT_CONTAINER_TEMP_DIR);
vargs.add("-Djava.io.tmpdir=" + childTmpDir); vargs.add("-Djava.io.tmpdir=" + childTmpDir);
MRApps.addLog4jSystemProperties(task, vargs, conf);
// Setup the log4j prop
long logSize = TaskLog.getTaskLogLength(conf);
setupLog4jProperties(task, vargs, logSize, conf);
if (conf.getProfileEnabled()) { if (conf.getProfileEnabled()) {
if (conf.getProfileTaskRange(task.isMapTask() if (conf.getProfileTaskRange(task.isMapTask()

View File

@ -61,7 +61,7 @@ public class TestMapReduceChildJVM {
" -Dlog4j.configuration=container-log4j.properties" + " -Dlog4j.configuration=container-log4j.properties" +
" -Dyarn.app.container.log.dir=<LOG_DIR>" + " -Dyarn.app.container.log.dir=<LOG_DIR>" +
" -Dyarn.app.container.log.filesize=0" + " -Dyarn.app.container.log.filesize=0" +
" -Dhadoop.root.logger=INFO,CLA" + " -Dhadoop.root.logger=INFO,CLA -Dhadoop.root.logfile=syslog" +
" org.apache.hadoop.mapred.YarnChild 127.0.0.1" + " org.apache.hadoop.mapred.YarnChild 127.0.0.1" +
" 54321" + " 54321" +
" attempt_0_0000_m_000000_0" + " attempt_0_0000_m_000000_0" +
@ -78,6 +78,73 @@ public class TestMapReduceChildJVM {
Assert.assertEquals("", app.cmdEnvironment.get("HADOOP_CLIENT_OPTS")); Assert.assertEquals("", app.cmdEnvironment.get("HADOOP_CLIENT_OPTS"));
} }
@Test (timeout = 30000)
public void testReduceCommandLineWithSeparateShuffle() throws Exception {
final Configuration conf = new Configuration();
conf.setBoolean(MRJobConfig.REDUCE_SEPARATE_SHUFFLE_LOG, true);
testReduceCommandLine(conf);
}
@Test (timeout = 30000)
public void testReduceCommandLineWithSeparateCRLAShuffle() throws Exception {
final Configuration conf = new Configuration();
conf.setBoolean(MRJobConfig.REDUCE_SEPARATE_SHUFFLE_LOG, true);
conf.setLong(MRJobConfig.SHUFFLE_LOG_KB, 1L);
conf.setInt(MRJobConfig.SHUFFLE_LOG_BACKUPS, 3);
testReduceCommandLine(conf);
}
@Test (timeout = 30000)
public void testReduceCommandLine() throws Exception {
final Configuration conf = new Configuration();
testReduceCommandLine(conf);
}
private void testReduceCommandLine(Configuration conf)
throws Exception {
MyMRApp app = new MyMRApp(0, 1, true, this.getClass().getName(), true);
conf.setBoolean(MRConfig.MAPREDUCE_APP_SUBMISSION_CROSS_PLATFORM, true);
Job job = app.submit(conf);
app.waitForState(job, JobState.SUCCEEDED);
app.verifyCompleted();
final long shuffleLogSize =
conf.getLong(MRJobConfig.SHUFFLE_LOG_KB, 0L) * 1024L;
final int shuffleBackups = conf.getInt(MRJobConfig.SHUFFLE_LOG_BACKUPS, 0);
final String appenderName = shuffleLogSize > 0L && shuffleBackups > 0
? "shuffleCRLA"
: "shuffleCLA";
Assert.assertEquals(
"[" + MRApps.crossPlatformify("JAVA_HOME") + "/bin/java" +
" -Djava.net.preferIPv4Stack=true" +
" -Dhadoop.metrics.log.level=WARN" +
" -Xmx200m -Djava.io.tmpdir=" + MRApps.crossPlatformify("PWD") + "/tmp" +
" -Dlog4j.configuration=container-log4j.properties" +
" -Dyarn.app.container.log.dir=<LOG_DIR>" +
" -Dyarn.app.container.log.filesize=0" +
" -Dhadoop.root.logger=INFO,CLA -Dhadoop.root.logfile=syslog" +
" -Dyarn.app.mapreduce.shuffle.logger=INFO," + appenderName +
" -Dyarn.app.mapreduce.shuffle.logfile=syslog.shuffle" +
" -Dyarn.app.mapreduce.shuffle.log.filesize=" + shuffleLogSize +
" -Dyarn.app.mapreduce.shuffle.log.backups=" + shuffleBackups +
" org.apache.hadoop.mapred.YarnChild 127.0.0.1" +
" 54321" +
" attempt_0_0000_r_000000_0" +
" 0" +
" 1><LOG_DIR>/stdout" +
" 2><LOG_DIR>/stderr ]", app.myCommandLine);
Assert.assertTrue("HADOOP_ROOT_LOGGER not set for job",
app.cmdEnvironment.containsKey("HADOOP_ROOT_LOGGER"));
Assert.assertEquals("INFO,console",
app.cmdEnvironment.get("HADOOP_ROOT_LOGGER"));
Assert.assertTrue("HADOOP_CLIENT_OPTS not set for job",
app.cmdEnvironment.containsKey("HADOOP_CLIENT_OPTS"));
Assert.assertEquals("", app.cmdEnvironment.get("HADOOP_CLIENT_OPTS"));
}
@Test (timeout = 30000) @Test (timeout = 30000)
public void testCommandLineWithLog4JConifg() throws Exception { public void testCommandLineWithLog4JConifg() throws Exception {
@ -99,7 +166,7 @@ public class TestMapReduceChildJVM {
" -Dlog4j.configuration=" + testLogPropertieFile + " -Dlog4j.configuration=" + testLogPropertieFile +
" -Dyarn.app.container.log.dir=<LOG_DIR>" + " -Dyarn.app.container.log.dir=<LOG_DIR>" +
" -Dyarn.app.container.log.filesize=0" + " -Dyarn.app.container.log.filesize=0" +
" -Dhadoop.root.logger=INFO,CLA" + " -Dhadoop.root.logger=INFO,CLA -Dhadoop.root.logfile=syslog" +
" org.apache.hadoop.mapred.YarnChild 127.0.0.1" + " org.apache.hadoop.mapred.YarnChild 127.0.0.1" +
" 54321" + " 54321" +
" attempt_0_0000_m_000000_0" + " attempt_0_0000_m_000000_0" +

View File

@ -43,6 +43,9 @@ import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.Path;
import org.apache.hadoop.mapred.InvalidJobConfException; import org.apache.hadoop.mapred.InvalidJobConfException;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.Task;
import org.apache.hadoop.mapred.TaskLog;
import org.apache.hadoop.mapreduce.JobID; import org.apache.hadoop.mapreduce.JobID;
import org.apache.hadoop.mapreduce.MRConfig; import org.apache.hadoop.mapreduce.MRConfig;
import org.apache.hadoop.mapreduce.MRJobConfig; import org.apache.hadoop.mapreduce.MRJobConfig;
@ -59,6 +62,7 @@ import org.apache.hadoop.mapreduce.v2.api.records.TaskType;
import org.apache.hadoop.util.ApplicationClassLoader; import org.apache.hadoop.util.ApplicationClassLoader;
import org.apache.hadoop.util.StringUtils; import org.apache.hadoop.util.StringUtils;
import org.apache.hadoop.yarn.ContainerLogAppender; import org.apache.hadoop.yarn.ContainerLogAppender;
import org.apache.hadoop.yarn.ContainerRollingLogAppender;
import org.apache.hadoop.yarn.api.ApplicationConstants; import org.apache.hadoop.yarn.api.ApplicationConstants;
import org.apache.hadoop.yarn.api.ApplicationConstants.Environment; import org.apache.hadoop.yarn.api.ApplicationConstants.Environment;
import org.apache.hadoop.yarn.api.records.LocalResource; import org.apache.hadoop.yarn.api.records.LocalResource;
@ -68,7 +72,6 @@ import org.apache.hadoop.yarn.conf.YarnConfiguration;
import org.apache.hadoop.yarn.exceptions.YarnRuntimeException; import org.apache.hadoop.yarn.exceptions.YarnRuntimeException;
import org.apache.hadoop.yarn.util.Apps; import org.apache.hadoop.yarn.util.Apps;
import org.apache.hadoop.yarn.util.ConverterUtils; import org.apache.hadoop.yarn.util.ConverterUtils;
import org.apache.log4j.RollingFileAppender;
/** /**
* Helper class for MR applications * Helper class for MR applications
@ -593,17 +596,31 @@ public class MRApps extends Apps {
return result; return result;
} }
public static String getChildLogLevel(Configuration conf, boolean isMap) {
if (isMap) {
return conf.get(
MRJobConfig.MAP_LOG_LEVEL,
JobConf.DEFAULT_LOG_LEVEL.toString()
);
} else {
return conf.get(
MRJobConfig.REDUCE_LOG_LEVEL,
JobConf.DEFAULT_LOG_LEVEL.toString()
);
}
}
/** /**
* Add the JVM system properties necessary to configure {@link ContainerLogAppender}. * Add the JVM system properties necessary to configure
* @param logLevel the desired log level (eg INFO/WARN/DEBUG) * {@link ContainerLogAppender} or
* @param logSize See {@link ContainerLogAppender#setTotalLogFileSize(long)} * {@link ContainerRollingLogAppender}.
* @param numBackups See {@link RollingFileAppender#setMaxBackupIndex(int)} *
* @param task for map/reduce, or null for app master
* @param vargs the argument list to append to * @param vargs the argument list to append to
* @param conf configuration of MR job * @param conf configuration of MR job
*/ */
public static void addLog4jSystemProperties( public static void addLog4jSystemProperties(Task task,
String logLevel, long logSize, int numBackups, List<String> vargs, List<String> vargs, Configuration conf) {
Configuration conf) {
String log4jPropertyFile = String log4jPropertyFile =
conf.get(MRJobConfig.MAPREDUCE_JOB_LOG4J_PROPERTIES_FILE, ""); conf.get(MRJobConfig.MAPREDUCE_JOB_LOG4J_PROPERTIES_FILE, "");
if (log4jPropertyFile.isEmpty()) { if (log4jPropertyFile.isEmpty()) {
@ -619,10 +636,29 @@ public class MRApps extends Apps {
vargs.add("-Dlog4j.configuration="+log4jPath.getName()); vargs.add("-Dlog4j.configuration="+log4jPath.getName());
} }
long logSize;
String logLevel;
int numBackups;
if (task == null) {
logSize = conf.getLong(MRJobConfig.MR_AM_LOG_KB,
MRJobConfig.DEFAULT_MR_AM_LOG_KB) << 10;
logLevel = conf.get(
MRJobConfig.MR_AM_LOG_LEVEL, MRJobConfig.DEFAULT_MR_AM_LOG_LEVEL);
numBackups = conf.getInt(MRJobConfig.MR_AM_LOG_BACKUPS,
MRJobConfig.DEFAULT_MR_AM_LOG_BACKUPS);
} else {
logSize = TaskLog.getTaskLogLimitBytes(conf);
logLevel = getChildLogLevel(conf, task.isMapTask());
numBackups = conf.getInt(MRJobConfig.TASK_LOG_BACKUPS,
MRJobConfig.DEFAULT_TASK_LOG_BACKUPS);
}
vargs.add("-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_DIR + "=" + vargs.add("-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_DIR + "=" +
ApplicationConstants.LOG_DIR_EXPANSION_VAR); ApplicationConstants.LOG_DIR_EXPANSION_VAR);
vargs.add( vargs.add(
"-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_SIZE + "=" + logSize); "-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_SIZE + "=" + logSize);
if (logSize > 0L && numBackups > 0) { if (logSize > 0L && numBackups > 0) {
// log should be rolled // log should be rolled
vargs.add("-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_BACKUPS + "=" vargs.add("-D" + YarnConfiguration.YARN_APP_CONTAINER_LOG_BACKUPS + "="
@ -631,6 +667,30 @@ public class MRApps extends Apps {
} else { } else {
vargs.add("-Dhadoop.root.logger=" + logLevel + ",CLA"); vargs.add("-Dhadoop.root.logger=" + logLevel + ",CLA");
} }
vargs.add("-Dhadoop.root.logfile=" + TaskLog.LogName.SYSLOG);
if ( task != null
&& !task.isMapTask()
&& conf.getBoolean(MRJobConfig.REDUCE_SEPARATE_SHUFFLE_LOG,
MRJobConfig.DEFAULT_REDUCE_SEPARATE_SHUFFLE_LOG)) {
final int numShuffleBackups = conf.getInt(MRJobConfig.SHUFFLE_LOG_BACKUPS,
MRJobConfig.DEFAULT_SHUFFLE_LOG_BACKUPS);
final long shuffleLogSize = conf.getLong(MRJobConfig.SHUFFLE_LOG_KB,
MRJobConfig.DEFAULT_SHUFFLE_LOG_KB) << 10;
final String shuffleLogger = logLevel
+ (shuffleLogSize > 0L && numShuffleBackups > 0
? ",shuffleCRLA"
: ",shuffleCLA");
vargs.add("-D" + MRJobConfig.MR_PREFIX
+ "shuffle.logger=" + shuffleLogger);
vargs.add("-D" + MRJobConfig.MR_PREFIX
+ "shuffle.logfile=" + TaskLog.LogName.SYSLOG + ".shuffle");
vargs.add("-D" + MRJobConfig.MR_PREFIX
+ "shuffle.log.filesize=" + shuffleLogSize);
vargs.add("-D" + MRJobConfig.MR_PREFIX
+ "shuffle.log.backups=" + numShuffleBackups);
}
} }
/** /**

View File

@ -27,7 +27,6 @@ import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path; import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.compress.CompressionCodec; import org.apache.hadoop.io.compress.CompressionCodec;
import org.apache.hadoop.mapred.FileAlreadyExistsException;
import org.apache.hadoop.mapreduce.security.TokenCache; import org.apache.hadoop.mapreduce.security.TokenCache;
import org.apache.hadoop.util.Progressable; import org.apache.hadoop.util.Progressable;
@ -287,7 +286,8 @@ public abstract class FileOutputFormat<K, V> implements OutputFormat<K, V> {
"This method can only be called from within a Job"); "This method can only be called from within a Job");
} }
String taskType = (conf.getBoolean(JobContext.TASK_ISMAP, true)) ? "m" : "r"; String taskType = conf.getBoolean(JobContext.TASK_ISMAP,
JobContext.DEFAULT_TASK_ISMAP) ? "m" : "r";
NumberFormat numberFormat = NumberFormat.getInstance(); NumberFormat numberFormat = NumberFormat.getInstance();
numberFormat.setMinimumIntegerDigits(5); numberFormat.setMinimumIntegerDigits(5);

View File

@ -473,6 +473,10 @@ public class TaskLog {
* @return the number of bytes to cap the log files at * @return the number of bytes to cap the log files at
*/ */
public static long getTaskLogLength(JobConf conf) { public static long getTaskLogLength(JobConf conf) {
return getTaskLogLimitBytes(conf);
}
public static long getTaskLogLimitBytes(Configuration conf) {
return conf.getLong(JobContext.TASK_USERLOG_LIMIT, 0) * 1024; return conf.getLong(JobContext.TASK_USERLOG_LIMIT, 0) * 1024;
} }

View File

@ -177,6 +177,7 @@ public interface MRJobConfig {
public static final String TASK_ATTEMPT_ID = "mapreduce.task.attempt.id"; public static final String TASK_ATTEMPT_ID = "mapreduce.task.attempt.id";
public static final String TASK_ISMAP = "mapreduce.task.ismap"; public static final String TASK_ISMAP = "mapreduce.task.ismap";
public static final boolean DEFAULT_TASK_ISMAP = true;
public static final String TASK_PARTITION = "mapreduce.task.partition"; public static final String TASK_PARTITION = "mapreduce.task.partition";
@ -773,6 +774,18 @@ public interface MRJobConfig {
MR_PREFIX + "task.container.log.backups"; MR_PREFIX + "task.container.log.backups";
public static final int DEFAULT_TASK_LOG_BACKUPS = 0; // don't roll public static final int DEFAULT_TASK_LOG_BACKUPS = 0; // don't roll
public static final String REDUCE_SEPARATE_SHUFFLE_LOG =
MR_PREFIX + "shuffle.log.separate";
public static final boolean DEFAULT_REDUCE_SEPARATE_SHUFFLE_LOG = true;
public static final String SHUFFLE_LOG_BACKUPS =
MR_PREFIX + "shuffle.log.backups";
public static final int DEFAULT_SHUFFLE_LOG_BACKUPS = 0; // don't roll
public static final String SHUFFLE_LOG_KB =
MR_PREFIX + "shuffle.log.limit.kb";
public static final long DEFAULT_SHUFFLE_LOG_KB = 0L;
public static final String WORKFLOW_NAME = "mapreduce.workflow.name"; public static final String WORKFLOW_NAME = "mapreduce.workflow.name";
public static final String WORKFLOW_NODE_NAME = public static final String WORKFLOW_NODE_NAME =
@ -812,4 +825,5 @@ public interface MRJobConfig {
"mapreduce.job.encrypted-intermediate-data.buffer.kb"; "mapreduce.job.encrypted-intermediate-data.buffer.kb";
public static final int DEFAULT_MR_ENCRYPTED_INTERMEDIATE_DATA_BUFFER_KB = public static final int DEFAULT_MR_ENCRYPTED_INTERMEDIATE_DATA_BUFFER_KB =
128; 128;
} }

View File

@ -713,6 +713,34 @@
</description> </description>
</property> </property>
<property>
<name>yarn.app.mapreduce.shuffle.log.separate</name>
<value>true</value>
<description>If enabled ('true') logging generated by the client-side shuffle
classes in a reducer will be written in a dedicated log file
'syslog.shuffle' instead of 'syslog'.
</description>
</property>
<property>
<name>yarn.app.mapreduce.shuffle.log.limit.kb</name>
<value>0</value>
<description>Maximum size of the syslog.shuffle file in kilobytes
(0 for no limit).
</description>
</property>
<property>
<name>yarn.app.mapreduce.shuffle.log.backups</name>
<value>0</value>
<description>If yarn.app.mapreduce.shuffle.log.limit.kb and
yarn.app.mapreduce.shuffle.log.backups are greater than zero
then a ContainerRollngLogAppender is used instead of ContainerLogAppender
for syslog.shuffle. See
org.apache.log4j.RollingFileAppender.maxBackupIndex
</description>
</property>
<property> <property>
<name>mapreduce.job.maxtaskfailures.per.tracker</name> <name>mapreduce.job.maxtaskfailures.per.tracker</name>
<value>3</value> <value>3</value>

View File

@ -392,14 +392,7 @@ public class YARNRunner implements ClientProtocol {
vargs.add(MRApps.crossPlatformifyMREnv(jobConf, Environment.JAVA_HOME) vargs.add(MRApps.crossPlatformifyMREnv(jobConf, Environment.JAVA_HOME)
+ "/bin/java"); + "/bin/java");
// TODO: why do we use 'conf' some places and 'jobConf' others? MRApps.addLog4jSystemProperties(null, vargs, conf);
long logSize = jobConf.getLong(MRJobConfig.MR_AM_LOG_KB,
MRJobConfig.DEFAULT_MR_AM_LOG_KB) << 10;
String logLevel = jobConf.get(
MRJobConfig.MR_AM_LOG_LEVEL, MRJobConfig.DEFAULT_MR_AM_LOG_LEVEL);
int numBackups = jobConf.getInt(MRJobConfig.MR_AM_LOG_BACKUPS,
MRJobConfig.DEFAULT_MR_AM_LOG_BACKUPS);
MRApps.addLog4jSystemProperties(logLevel, logSize, numBackups, vargs, conf);
// Check for Java Lib Path usage in MAP and REDUCE configs // Check for Java Lib Path usage in MAP and REDUCE configs
warnForJavaLibPath(conf.get(MRJobConfig.MAP_JAVA_OPTS,""), "map", warnForJavaLibPath(conf.get(MRJobConfig.MAP_JAVA_OPTS,""), "map",

View File

@ -38,6 +38,7 @@ public class ContainerLogAppender extends FileAppender
implements Flushable implements Flushable
{ {
private String containerLogDir; private String containerLogDir;
private String containerLogFile;
//so that log4j can configure it from the configuration(log4j.properties). //so that log4j can configure it from the configuration(log4j.properties).
private int maxEvents; private int maxEvents;
private Queue<LoggingEvent> tail = null; private Queue<LoggingEvent> tail = null;
@ -49,7 +50,7 @@ public class ContainerLogAppender extends FileAppender
if (maxEvents > 0) { if (maxEvents > 0) {
tail = new LinkedList<LoggingEvent>(); tail = new LinkedList<LoggingEvent>();
} }
setFile(new File(this.containerLogDir, "syslog").toString()); setFile(new File(this.containerLogDir, containerLogFile).toString());
setAppend(true); setAppend(true);
super.activateOptions(); super.activateOptions();
} }
@ -102,6 +103,14 @@ public class ContainerLogAppender extends FileAppender
this.containerLogDir = containerLogDir; this.containerLogDir = containerLogDir;
} }
public String getContainerLogFile() {
return containerLogFile;
}
public void setContainerLogFile(String containerLogFile) {
this.containerLogFile = containerLogFile;
}
private static final int EVENT_SIZE = 100; private static final int EVENT_SIZE = 100;
public long getTotalLogFileSize() { public long getTotalLogFileSize() {

View File

@ -34,11 +34,12 @@ import java.io.Flushable;
public class ContainerRollingLogAppender extends RollingFileAppender public class ContainerRollingLogAppender extends RollingFileAppender
implements Flushable { implements Flushable {
private String containerLogDir; private String containerLogDir;
private String containerLogFile;
@Override @Override
public void activateOptions() { public void activateOptions() {
synchronized (this) { synchronized (this) {
setFile(new File(this.containerLogDir, "syslog").toString()); setFile(new File(this.containerLogDir, containerLogFile).toString());
setAppend(true); setAppend(true);
super.activateOptions(); super.activateOptions();
} }
@ -62,4 +63,12 @@ public class ContainerRollingLogAppender extends RollingFileAppender
public void setContainerLogDir(String containerLogDir) { public void setContainerLogDir(String containerLogDir) {
this.containerLogDir = containerLogDir; this.containerLogDir = containerLogDir;
} }
public String getContainerLogFile() {
return containerLogFile;
}
public void setContainerLogFile(String containerLogFile) {
this.containerLogFile = containerLogFile;
}
} }

View File

@ -30,6 +30,7 @@ public class TestContainerLogAppender {
claAppender.setName("testCLA"); claAppender.setName("testCLA");
claAppender.setLayout(new PatternLayout("%-5p [%t]: %m%n")); claAppender.setLayout(new PatternLayout("%-5p [%t]: %m%n"));
claAppender.setContainerLogDir("target/testAppendInClose/logDir"); claAppender.setContainerLogDir("target/testAppendInClose/logDir");
claAppender.setContainerLogFile("syslog");
claAppender.setTotalLogFileSize(1000); claAppender.setTotalLogFileSize(1000);
claAppender.activateOptions(); claAppender.activateOptions();
final Logger claLog = Logger.getLogger("testAppendInClose-catergory"); final Logger claLog = Logger.getLogger("testAppendInClose-catergory");

View File

@ -13,6 +13,7 @@
# #
# Define some default values that can be overridden by system properties # Define some default values that can be overridden by system properties
hadoop.root.logger=DEBUG,CLA hadoop.root.logger=DEBUG,CLA
yarn.app.mapreduce.shuffle.logger=${hadoop.root.logger}
# Define the root logger to the system property "hadoop.root.logger". # Define the root logger to the system property "hadoop.root.logger".
log4j.rootLogger=${hadoop.root.logger}, EventCounter log4j.rootLogger=${hadoop.root.logger}, EventCounter
@ -30,18 +31,44 @@ yarn.app.container.log.filesize=100
log4j.appender.CLA=org.apache.hadoop.yarn.ContainerLogAppender log4j.appender.CLA=org.apache.hadoop.yarn.ContainerLogAppender
log4j.appender.CLA.containerLogDir=${yarn.app.container.log.dir} log4j.appender.CLA.containerLogDir=${yarn.app.container.log.dir}
log4j.appender.CLA.containerLogFile=${hadoop.root.logfile}
log4j.appender.CLA.totalLogFileSize=${yarn.app.container.log.filesize} log4j.appender.CLA.totalLogFileSize=${yarn.app.container.log.filesize}
log4j.appender.CLA.layout=org.apache.log4j.PatternLayout log4j.appender.CLA.layout=org.apache.log4j.PatternLayout
log4j.appender.CLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n log4j.appender.CLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n
log4j.appender.CRLA=org.apache.hadoop.yarn.ContainerRollingLogAppender log4j.appender.CRLA=org.apache.hadoop.yarn.ContainerRollingLogAppender
log4j.appender.CRLA.containerLogDir=${yarn.app.container.log.dir} log4j.appender.CRLA.containerLogDir=${yarn.app.container.log.dir}
log4j.appender.CRLA.containerLogFile=${hadoop.root.logfile}
log4j.appender.CRLA.maximumFileSize=${yarn.app.container.log.filesize} log4j.appender.CRLA.maximumFileSize=${yarn.app.container.log.filesize}
log4j.appender.CRLA.maxBackupIndex=${yarn.app.container.log.backups} log4j.appender.CRLA.maxBackupIndex=${yarn.app.container.log.backups}
log4j.appender.CRLA.layout=org.apache.log4j.PatternLayout log4j.appender.CRLA.layout=org.apache.log4j.PatternLayout
log4j.appender.CRLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n log4j.appender.CRLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n
log4j.appender.shuffleCLA=org.apache.hadoop.yarn.ContainerLogAppender
log4j.appender.shuffleCLA.containerLogDir=${yarn.app.container.log.dir}
log4j.appender.shuffleCLA.containerLogFile=${yarn.app.mapreduce.shuffle.logfile}
log4j.appender.shuffleCLA.totalLogFileSize=${yarn.app.mapreduce.shuffle.log.filesize}
log4j.appender.shuffleCLA.layout=org.apache.log4j.PatternLayout
log4j.appender.shuffleCLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n
log4j.appender.shuffleCRLA=org.apache.hadoop.yarn.ContainerRollingLogAppender
log4j.appender.shuffleCRLA.containerLogDir=${yarn.app.container.log.dir}
log4j.appender.shuffleCRLA.containerLogFile=${yarn.app.mapreduce.shuffle.logfile}
log4j.appender.shuffleCRLA.maximumFileSize=${yarn.app.mapreduce.shuffle.log.filesize}
log4j.appender.shuffleCRLA.maxBackupIndex=${yarn.app.mapreduce.shuffle.log.backups}
log4j.appender.shuffleCRLA.layout=org.apache.log4j.PatternLayout
log4j.appender.shuffleCRLA.layout.ConversionPattern=%d{ISO8601} %p [%t] %c: %m%n
################################################################################
# Shuffle Logger
#
log4j.logger.org.apache.hadoop.mapreduce.task.reduce=${yarn.app.mapreduce.shuffle.logger}
log4j.additivity.org.apache.hadoop.mapreduce.task.reduce=false
# Merger is used for both map-side and reduce-side spill merging. On the map
# side yarn.app.mapreduce.shuffle.logger == hadoop.root.logger
#
log4j.logger.org.apache.hadoop.mapred.Merger=${yarn.app.mapreduce.shuffle.logger}
log4j.additivity.org.apache.hadoop.mapred.Merger=false
# #
# Event Counter Appender # Event Counter Appender
# Sends counts of logging messages at different severity levels to Hadoop Metrics. # Sends counts of logging messages at different severity levels to Hadoop Metrics.