|
| 1 | +/* |
| 2 | + * Licensed to the Apache Software Foundation (ASF) under one or more |
| 3 | + * contributor license agreements. See the NOTICE file distributed with |
| 4 | + * this work for additional information regarding copyright ownership. |
| 5 | + * The ASF licenses this file to You under the Apache License, Version 2.0 |
| 6 | + * (the "License"); you may not use this file except in compliance with |
| 7 | + * the License. You may obtain a copy of the License at |
| 8 | + * |
| 9 | + * http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | + * |
| 11 | + * Unless required by applicable law or agreed to in writing, software |
| 12 | + * distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | + * See the License for the specific language governing permissions and |
| 15 | + * limitations under the License. |
| 16 | + */ |
| 17 | + |
| 18 | +package org.apache.spark |
| 19 | + |
| 20 | +/** |
| 21 | + * Low-level status reporting APIs for monitoring job and stage progress. |
| 22 | + * |
| 23 | + * These APIs intentionally provide very weak consistency semantics; consumers of these APIs should |
| 24 | + * be prepared to handle empty / missing information. For example, a job's stage ids may be known |
| 25 | + * but the status API may not have any information about the details of those stages, so |
| 26 | + * `getStageInfo` could potentially return `None` for a valid stage id. |
| 27 | + * |
| 28 | + * To limit memory usage, these APIs only provide information on recent jobs / stages. These APIs |
| 29 | + * will provide information for the last `spark.ui.retainedStages` stages and |
| 30 | + * `spark.ui.retainedJobs` jobs. |
| 31 | + * |
| 32 | + * NOTE: this class's constructor should be considered private and may be subject to change. |
| 33 | + */ |
| 34 | +class SparkStatusTracker private[spark] (sc: SparkContext) { |
| 35 | + |
| 36 | + private val jobProgressListener = sc.jobProgressListener |
| 37 | + |
| 38 | + /** |
| 39 | + * Return a list of all known jobs in a particular job group. If `jobGroup` is `null`, then |
| 40 | + * returns all known jobs that are not associated with a job group. |
| 41 | + * |
| 42 | + * The returned list may contain running, failed, and completed jobs, and may vary across |
| 43 | + * invocations of this method. This method does not guarantee the order of the elements in |
| 44 | + * its result. |
| 45 | + */ |
| 46 | + def getJobIdsForGroup(jobGroup: String): Array[Int] = { |
| 47 | + jobProgressListener.synchronized { |
| 48 | + val jobData = jobProgressListener.jobIdToData.valuesIterator |
| 49 | + jobData.filter(_.jobGroup.orNull == jobGroup).map(_.jobId).toArray |
| 50 | + } |
| 51 | + } |
| 52 | + |
| 53 | + /** |
| 54 | + * Returns an array containing the ids of all active stages. |
| 55 | + * |
| 56 | + * This method does not guarantee the order of the elements in its result. |
| 57 | + */ |
| 58 | + def getActiveStageIds(): Array[Int] = { |
| 59 | + jobProgressListener.synchronized { |
| 60 | + jobProgressListener.activeStages.values.map(_.stageId).toArray |
| 61 | + } |
| 62 | + } |
| 63 | + |
| 64 | + /** |
| 65 | + * Returns an array containing the ids of all active jobs. |
| 66 | + * |
| 67 | + * This method does not guarantee the order of the elements in its result. |
| 68 | + */ |
| 69 | + def getActiveJobIds(): Array[Int] = { |
| 70 | + jobProgressListener.synchronized { |
| 71 | + jobProgressListener.activeJobs.values.map(_.jobId).toArray |
| 72 | + } |
| 73 | + } |
| 74 | + |
| 75 | + /** |
| 76 | + * Returns job information, or `None` if the job info could not be found or was garbage collected. |
| 77 | + */ |
| 78 | + def getJobInfo(jobId: Int): Option[SparkJobInfo] = { |
| 79 | + jobProgressListener.synchronized { |
| 80 | + jobProgressListener.jobIdToData.get(jobId).map { data => |
| 81 | + new SparkJobInfoImpl(jobId, data.stageIds.toArray, data.status) |
| 82 | + } |
| 83 | + } |
| 84 | + } |
| 85 | + |
| 86 | + /** |
| 87 | + * Returns stage information, or `None` if the stage info could not be found or was |
| 88 | + * garbage collected. |
| 89 | + */ |
| 90 | + def getStageInfo(stageId: Int): Option[SparkStageInfo] = { |
| 91 | + jobProgressListener.synchronized { |
| 92 | + for ( |
| 93 | + info <- jobProgressListener.stageIdToInfo.get(stageId); |
| 94 | + data <- jobProgressListener.stageIdToData.get((stageId, info.attemptId)) |
| 95 | + ) yield { |
| 96 | + new SparkStageInfoImpl( |
| 97 | + stageId, |
| 98 | + info.attemptId, |
| 99 | + info.name, |
| 100 | + info.numTasks, |
| 101 | + data.numActiveTasks, |
| 102 | + data.numCompleteTasks, |
| 103 | + data.numFailedTasks) |
| 104 | + } |
| 105 | + } |
| 106 | + } |
| 107 | +} |
0 commit comments