Github user squito commented on a diff in the pull request:
https://github.com/apache/spark/pull/22612#discussion_r222373667
--- Diff:
core/src/main/scala/org/apache/spark/executor/ProcfsBasedSystems.scala ---
@@ -0,0 +1,268 @@
+/*
+ * 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 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 extends Logging {
+ var procfsDir = "/proc/"
+ val procfsStatFile = "stat"
+ var pageSize = 0
+ var isAvailable: Boolean = isItProcfsBased
+ private val pid: Int = computePid()
+ private val ptree: scala.collection.mutable.Map[ Int, Set[Int]] =
+ scala.collection.mutable.Map[ Int, Set[Int]]()
+
+ var allMetrics: ProcfsBasedSystemsMetrics = ProcfsBasedSystemsMetrics(0,
0, 0, 0, 0, 0)
+ private var latestJVMVmemTotal: Long = 0
+ private var latestJVMRSSTotal: Long = 0
+ private var latestPythonVmemTotal: Long = 0
+ private var latestPythonRSSTotal: Long = 0
+ private var latestOtherVmemTotal: Long = 0
+ private var latestOtherRSSTotal: Long = 0
+
+ computeProcessTree()
+
+ private def isItProcfsBased: 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[Byte] = 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")
+ isAvailable = false
+ return -1
+ case _ => logDebug("Some exception occurred when trying to compute
process tree. " +
+ "As a result reporting of ProcessTree metrics is stopped")
+ isAvailable = false
--- End diff --
include the exception as the second argument to `logDebug`, so we get the
full stack trace. Also the two cases are basically the same, I'd combine them,
no real value in a separate msg just for IOException
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
