查看Spark进程的JVM配置及内存使用
如何查看正在运行的Spark进程的JVM配置以及分代的内存使用情况,是线上运行作业常用的监控手段:
1、通过ps命令查询PID
- ps -ef | grep 5661
可以根据命令中的特殊字符来定位pid
2、使用jinfo命令查询该进程的JVM参数设置
- jinfo 105007
可以得到详细的JVM配置信息
- Attaching to process ID 105007, please wait...
- Debugger attached successfully.
- Server compiler detected.
- JVM version is 24.65-b04
- Java System Properties:
- spark.local.dir = /diskb/sparktmp,/diskc/sparktmp,/diskd/sparktmp,/diske/sparktmp,/diskf/sparktmp,/diskg/sparktmp
- java.runtime.name = Java(TM) SE Runtime Environment
- java.vm.version = 24.65-b04
- sun.boot.library.path = /usr/java/jdk1.7.0_67-cloudera/jre/lib/amd64
- java.vendor.url = http://java.oracle.com/
- java.vm.vendor = Oracle Corporation
- path.separator = :
- file.encoding.pkg = sun.io
- java.vm.name = Java HotSpot(TM) 64-Bit Server VM
- sun.os.patch.level = unknown
- sun.java.launcher = SUN_STANDARD
- user.country = CN
- user.dir = /opt/bin/spark_dev_job
- java.vm.specification.name = Java Virtual Machine Specification
- java.runtime.version = 1.7.0_67-b01
- java.awt.graphicsenv = sun.awt.X11GraphicsEnvironment
- SPARK_SUBMIT = true
- os.arch = amd64
- java.endorsed.dirs = /usr/java/jdk1.7.0_67-cloudera/jre/lib/endorsed
- spark.executor.memory = 24g
- line.separator =
- java.io.tmpdir = /tmp
- java.vm.specification.vendor = Oracle Corporation
- os.name = Linux
- spark.driver.memory = 15g
- spark.master = spark://10.130.2.220:7077
- sun.jnu.encoding = UTF-8
- java.library.path = :/opt/cloudera/parcels/CDH/lib/hadoop/lib/native:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib
- sun.nio.ch.bugLevel =
- java.class.version = 51.0
- java.specification.name = Java Platform API Specification
- sun.management.compiler = HotSpot 64-Bit Tiered Compilers
- spark.submit.deployMode = client
- spark.executor.extraJavaOptions = -XX:PermSize=8m -XX:+PrintGCDetails -XX:+PrintGCTimeStamps
- os.version = 2.6.32-573.8.1.el6.x86_64
- user.home = /root
- user.timezone = PRC
- java.awt.printerjob = sun.print.PSPrinterJob
- file.encoding = UTF-8
- java.specification.version = 1.7
- spark.app.name = com.hexun.streaming.NewsTopNRealRankOffsetRise
- spark.eventLog.enabled = true
- user.name = root
- java.class.path = /opt/cloudera/parcels/CDH/lib/hadoop/lib/snappy-java-1.0.4.1.jar:/opt/modules/spark-1.6.1-bin-hadoop2.6/conf/:/opt/modules/spark-1.6.1-bin-hadoop2.6/lib/spark-assembly-1.6.1-hadoop2.6.0.jar:/opt/modules/spark-1.6.1-bin-hadoop2.6/lib/datanucleus-core-3.2.10.jar:/opt/modules/spark-1.6.1-bin-hadoop2.6/lib/datanucleus-api-jdo-3.2.6.jar:/opt/modules/spark-1.6.1-bin-hadoop2.6/lib/datanucleus-rdbms-3.2.9.jar:/etc/hadoop/conf/
- java.vm.specification.version = 1.7
- sun.arch.data.model = 64
- sun.java.command = org.apache.spark.deploy.SparkSubmit --master spark://10.130.2.220:7077 --conf spark.driver.memory=15g --conf spark.executor.extraJavaOptions=-XX:PermSize=8m -XX:+PrintGCDetails -XX:+PrintGCTimeStamps --conf spark.ui.port=5661 --class com.hexun.streaming.NewsTopNRealRankOffsetRise --executor-memory 24g --total-executor-cores 24 --jars /opt/bin/sparkJars/kafka_2.10-0.8.2.1.jar,/opt/bin/sparkJars/spark-streaming-kafka_2.10-1.6.1.jar,/opt/bin/sparkJars/metrics-core-2.2.0.jar,/opt/bin/sparkJars/mysql-connector-java-5.1.26-bin.jar NewsTopNRealRankOffsetRise.jar
- java.home = /usr/java/jdk1.7.0_67-cloudera/jre
- user.language = zh
- java.specification.vendor = Oracle Corporation
- awt.toolkit = sun.awt.X11.XToolkit
- spark.ui.port = 5661
- java.vm.info = mixed mode
- java.version = 1.7.0_67
- java.ext.dirs = /usr/java/jdk1.7.0_67-cloudera/jre/lib/ext:/usr/java/packages/lib/ext
- sun.boot.class.path = /usr/java/jdk1.7.0_67-cloudera/jre/lib/resources.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/rt.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/sunrsasign.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/jsse.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/jce.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/charsets.jar:/usr/java/jdk1.7.0_67-cloudera/jre/lib/jfr.jar:/usr/java/jdk1.7.0_67-cloudera/jre/classes
- java.vendor = Oracle Corporation
- file.separator = /
- spark.cores.max = 24
- spark.eventLog.dir = hdfs://nameservice1/spark-log
- java.vendor.url.bug = http://bugreport.sun.com/bugreport/
- sun.io.unicode.encoding = UnicodeLittle
- sun.cpu.endian = little
- spark.jars = file:/opt/bin/sparkJars/kafka_2.10-0.8.2.1.jar,file:/opt/bin/sparkJars/spark-streaming-kafka_2.10-1.6.1.jar,file:/opt/bin/sparkJars/metrics-core-2.2.0.jar,file:/opt/bin/sparkJars/mysql-connector-java-5.1.26-bin.jar,file:/opt/bin/spark_dev_job/NewsTopNRealRankOffsetRise.jar
- sun.cpu.isalist =
- VM Flags:
- -Xms15g -Xmx15g -XX:MaxPermSize=256m
3、使用jmap查看进程中内存分代使用的情况
- jmap -heap 105007
可以得到该Java进程使用内存的详细情况,包括新生代老年代内存的使用
- Attaching to process ID 105007, please wait...
- Debugger attached successfully.
- Server compiler detected.
- JVM version is 24.65-b04
- using thread-local object allocation.
- Parallel GC with 18 thread(s)
- Heap Configuration:
- MinHeapFreeRatio = 0
- MaxHeapFreeRatio = 100
- MaxHeapSize = 16106127360 (15360.0MB)
- NewSize = 1310720 (1.25MB)
- MaxNewSize = 17592186044415 MB
- OldSize = 5439488 (5.1875MB)
- NewRatio = 2
- SurvivorRatio = 8
- PermSize = 21757952 (20.75MB)
- MaxPermSize = 268435456 (256.0MB)
- G1HeapRegionSize = 0 (0.0MB)
- Heap Usage:
- PS Young Generation
- Eden Space:
- capacity = 4945084416 (4716.0MB)
- used = 2674205152 (2550.320770263672MB)
- free = 2270879264 (2165.679229736328MB)
- 54.07804856369109% used
- From Space:
- capacity = 217579520 (207.5MB)
- used = 37486624 (35.750030517578125MB)
- free = 180092896 (171.74996948242188MB)
- 17.22893036991717% used
- To Space:
- capacity = 206045184 (196.5MB)
- used = 0 (0.0MB)
- free = 206045184 (196.5MB)
- 0.0% used
- PS Old Generation
- capacity = 10737418240 (10240.0MB)
- used = 7431666880 (7087.389831542969MB)
- free = 3305751360 (3152.6101684570312MB)
- 69.2127913236618% used
- PS Perm Generation
- capacity = 268435456 (256.0MB)
- used = 128212824 (122.27327728271484MB)
- free = 140222632 (133.72672271728516MB)
- 47.762998938560486% used