Github user dhruve commented on a diff in the pull request:
https://github.com/apache/spark/pull/22612#discussion_r222781707
--- Diff:
core/src/main/scala/org/apache/spark/executor/ProcfsBasedSystems.scala ---
@@ -0,0 +1,266 @@
+/*
+ * 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.
+ */
+
+package org.apache.spark.executor
+
+import java.io._
+import java.nio.charset.Charset
+import java.nio.file.{Files, Paths}
+import java.util.Locale
+
+import scala.collection.mutable
+import scala.collection.mutable.ArrayBuffer
+import scala.collection.mutable.Queue
+
+import org.apache.spark.SparkEnv
+import org.apache.spark.internal.{config, Logging}
+
+private[spark] case class ProcfsBasedSystemsMetrics(
+ jvmVmemTotal: Long,
+ jvmRSSTotal: Long,
+ pythonVmemTotal: Long,
+ pythonRSSTotal: Long,
+ otherVmemTotal: Long,
+ otherRSSTotal: Long)
+
+// Some of the ideas here are taken from the ProcfsBasedProcessTree class
in hadoop
+// project.
+private[spark] class ProcfsBasedSystems(val procfsDir: String = "/proc/")
extends Logging {
+ val procfsStatFile = "stat"
+ var pageSize = computePageSize()
+ var isAvailable: Boolean = isProcfsAvailable
+ private val pid = computePid()
+ private val ptree = mutable.Map[ Int, Set[Int]]()
+
+ var allMetrics: ProcfsBasedSystemsMetrics = ProcfsBasedSystemsMetrics(0,
0, 0, 0, 0, 0)
+ private var latestJVMVmemTotal = 0L
+ private var latestJVMRSSTotal = 0L
+ private var latestPythonVmemTotal = 0L
+ private var latestPythonRSSTotal = 0L
+ private var latestOtherVmemTotal = 0L
+ private var latestOtherRSSTotal = 0L
+
+ computeProcessTree()
+
+ private def isProcfsAvailable: Boolean = {
+ val testing = sys.env.contains("SPARK_TESTING") ||
sys.props.contains("spark.testing")
+ if (testing) {
+ return true
+ }
+ try {
+ if (!Files.exists(Paths.get(procfsDir))) {
+ return false
+ }
+ }
+ catch {
+ case f: FileNotFoundException => return false
+ }
+ val shouldLogStageExecutorMetrics =
+ SparkEnv.get.conf.get(config.EVENT_LOG_STAGE_EXECUTOR_METRICS)
+ val shouldLogStageExecutorProcessTreeMetrics =
+ SparkEnv.get.conf.get(config.EVENT_LOG_PROCESS_TREE_METRICS)
+ shouldLogStageExecutorProcessTreeMetrics &&
shouldLogStageExecutorMetrics
+ }
+
+ private def computePid(): Int = {
+ if (!isAvailable) {
+ return -1;
+ }
+ try {
+ // This can be simplified in java9:
+ //
https://docs.oracle.com/javase/9/docs/api/java/lang/ProcessHandle.html
+ val cmd = Array("bash", "-c", "echo $PPID")
+ val length = 10
+ val out = Array.fill[Byte](length)(0)
+ Runtime.getRuntime.exec(cmd).getInputStream.read(out)
+ val pid = Integer.parseInt(new String(out, "UTF-8").trim)
+ return pid;
+ }
+ catch {
+ case e: IOException => logDebug("IO Exception when trying to compute
process tree." +
+ " As a result reporting of ProcessTree metrics is stopped", e)
+ isAvailable = false
+ return -1
+ }
+ }
+
+ private def computePageSize(): Long = {
+ val testing = sys.env.contains("SPARK_TESTING") ||
sys.props.contains("spark.testing")
+ if (testing) {
+ return 0;
+ }
+ val cmd = Array("getconf", "PAGESIZE")
+ val out = Array.fill[Byte](10)(0)
+ Runtime.getRuntime.exec(cmd).getInputStream.read(out)
+ return Integer.parseInt(new String(out, "UTF-8").trim)
+ }
+
+ private def computeProcessTree(): Unit = {
+ if (!isAvailable) {
+ return
+ }
+ val queue = mutable.Queue.empty[Int]
+ queue += pid
+ while( !queue.isEmpty ) {
+ val p = queue.dequeue()
+ val c = getChildPids(p)
+ if(!c.isEmpty) {
+ queue ++= c
+ ptree += (p -> c.toSet)
+ }
+ else {
+ ptree += (p -> Set[Int]())
+ }
+ }
+ }
+
+ private def getChildPids(pid: Int): ArrayBuffer[Int] = {
+ try {
+ val cmd = Array("pgrep", "-P", pid.toString)
+ val input = Runtime.getRuntime.exec(cmd).getInputStream
+ val childPidsInByte = mutable.ArrayBuffer.empty[Byte]
+ var d = input.read()
+ while (d != -1) {
+ childPidsInByte.append(d.asInstanceOf[Byte])
+ d = input.read()
+ }
+ input.close()
+ val childPids = new String(childPidsInByte.toArray,
"UTF-8").split("\n")
+ val childPidsInInt = mutable.ArrayBuffer.empty[Int]
+ for (p <- childPids) {
+ if (p != "") {
+ childPidsInInt += Integer.parseInt(p)
+ }
+ }
+ childPidsInInt
+ } catch {
+ case e: IOException => logDebug("IO Exception when trying to compute
process tree." +
+ " As a result reporting of ProcessTree metrics is stopped", e)
+ isAvailable = false
+ return mutable.ArrayBuffer.empty[Int]
+ }
+ }
+
+ def getProcessInfo(pid: Int): Unit = {
--- End diff --
getProcessInfo is not returning anything here. What you could do is to set
it to perhaps setProcessMetrics and pass in a temporary
ProcfsBasedSystemsMetrics. And after setting the metrics for all the childPIDs
replace/update the metrics instance.
More like:
``` scala
def setProcessMetrics(pid: Int, metrics: ProcfsBasedSystemsMetrics) {
....
if (java) {
metrics.memUsage(rssMemory) += ....
} else if (python) {
metrics.memUsage(pythonRssMemory) += ....
}
}
```
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
