Github user vanzin commented on a diff in the pull request: https://github.com/apache/spark/pull/22504#discussion_r223853075 --- Diff: core/src/test/scala/org/apache/spark/util/logging/DriverLoggerSuite.scala --- @@ -0,0 +1,109 @@ +/* + * 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.util.logging + +import java.io.{BufferedInputStream, FileInputStream} + +import org.apache.commons.io.FileUtils + +import org.apache.spark._ +import org.apache.spark.{SparkContext, SparkFunSuite} +import org.apache.spark.internal.config._ +import org.apache.spark.launcher.SparkLauncher +import org.apache.spark.network.util.JavaUtils +import org.apache.spark.util.Utils + +class DriverLoggerSuite extends SparkFunSuite { + + test("driver logs are persisted") { + val sc = getSparkContext() + // Wait for application to start + Thread.sleep(1000) + + val app_id = sc.applicationId + // Run a simple spark application + sc.parallelize(1 to 1000).count() + + // Assert driver log file exists + val rootDir = Utils.getLocalDir(sc.getConf) + val driverLogsDir = FileUtils.getFile(rootDir, "driver_logs") + assert(driverLogsDir.exists()) + val files = driverLogsDir.listFiles() + assert(files.length === 1) + assert(files(0).getName.equals("driver.log")) + + sc.stop() + // On application end, file is moved to Hdfs (which is a local dir for this test) + assert(!driverLogsDir.exists()) + val hdfsDir = FileUtils.getFile("/tmp/hdfs_logs/", app_id) + assert(hdfsDir.exists()) + val hdfsFiles = hdfsDir.listFiles() + assert(hdfsFiles.length > 0) + JavaUtils.deleteRecursively(hdfsDir) + assert(!hdfsDir.exists()) + } + + test("driver logs are synced to hdfs continuously") { + val sc = getSparkContext() + // Wait for application to start + Thread.sleep(1000) + + val app_id = sc.applicationId + // Run a simple spark application + sc.parallelize(1 to 1000).count() + + // Assert driver log file exists + val rootDir = Utils.getLocalDir(sc.getConf) + val driverLogsDir = FileUtils.getFile(rootDir, "driver_logs") + assert(driverLogsDir.exists()) + val files = driverLogsDir.listFiles() + assert(files.length === 1) + assert(files(0).getName.equals("driver.log")) + for (i <- 1 to 1000) { + logInfo("Log enough data to log file so that it can be flushed") + } + + // After 5 secs, file contents are synced to Hdfs (which is a local dir for this test) + Thread.sleep(6000) --- End diff -- I'm not a fan of code that relies on exact timing between threads for things to work. This would be better if you drove the `HdfsAsyncWriter` class manually from the test, instead of indirectly through a `SparkContext`. The you can also control the log file flushing explicitly instead of the hack you have above.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org