001/**
002* Licensed to the Apache Software Foundation (ASF) under one
003* or more contributor license agreements.  See the NOTICE file
004* distributed with this work for additional information
005* regarding copyright ownership.  The ASF licenses this file
006* to you under the Apache License, Version 2.0 (the
007* "License"); you may not use this file except in compliance
008* with the License.  You may obtain a copy of the License at
009*
010*     http://www.apache.org/licenses/LICENSE-2.0
011*
012* Unless required by applicable law or agreed to in writing, software
013* distributed under the License is distributed on an "AS IS" BASIS,
014* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015* See the License for the specific language governing permissions and
016* limitations under the License.
017*/
018
019package org.apache.hadoop.yarn.api;
020
021import java.io.IOException;
022
023import org.apache.hadoop.classification.InterfaceAudience.Public;
024import org.apache.hadoop.classification.InterfaceStability.Stable;
025import org.apache.hadoop.yarn.api.protocolrecords.GetContainerStatusesRequest;
026import org.apache.hadoop.yarn.api.protocolrecords.GetContainerStatusesResponse;
027import org.apache.hadoop.yarn.api.protocolrecords.StartContainerRequest;
028import org.apache.hadoop.yarn.api.protocolrecords.StartContainersRequest;
029import org.apache.hadoop.yarn.api.protocolrecords.StartContainersResponse;
030import org.apache.hadoop.yarn.api.protocolrecords.StopContainersRequest;
031import org.apache.hadoop.yarn.api.protocolrecords.StopContainersResponse;
032import org.apache.hadoop.yarn.api.records.Container;
033import org.apache.hadoop.yarn.api.records.ContainerId;
034import org.apache.hadoop.yarn.api.records.ContainerLaunchContext;
035import org.apache.hadoop.yarn.api.records.ContainerStatus;
036import org.apache.hadoop.yarn.exceptions.NMNotYetReadyException;
037import org.apache.hadoop.yarn.exceptions.YarnException;
038
039/**
040 * <p>The protocol between an <code>ApplicationMaster</code> and a 
041 * <code>NodeManager</code> to start/stop containers and to get status
042 * of running containers.</p>
043 * 
044 * <p>If security is enabled the <code>NodeManager</code> verifies that the
045 * <code>ApplicationMaster</code> has truly been allocated the container
046 * by the <code>ResourceManager</code> and also verifies all interactions such 
047 * as stopping the container or obtaining status information for the container.
048 * </p>
049 */
050@Public
051@Stable
052public interface ContainerManagementProtocol {
053
054  /**
055   * <p>
056   * The <code>ApplicationMaster</code> provides a list of
057   * {@link StartContainerRequest}s to a <code>NodeManager</code> to
058   * <em>start</em> {@link Container}s allocated to it using this interface.
059   * </p>
060   * 
061   * <p>
062   * The <code>ApplicationMaster</code> has to provide details such as allocated
063   * resource capability, security tokens (if enabled), command to be executed
064   * to start the container, environment for the process, necessary
065   * binaries/jar/shared-objects etc. via the {@link ContainerLaunchContext} in
066   * the {@link StartContainerRequest}.
067   * </p>
068   * 
069   * <p>
070   * The <code>NodeManager</code> sends a response via
071   * {@link StartContainersResponse} which includes a list of
072   * {@link Container}s of successfully launched {@link Container}s, a
073   * containerId-to-exception map for each failed {@link StartContainerRequest} in
074   * which the exception indicates errors from per container and a
075   * allServicesMetaData map between the names of auxiliary services and their
076   * corresponding meta-data. Note: None-container-specific exceptions will
077   * still be thrown by the API method itself.
078   * </p>
079   * <p>
080   * The <code>ApplicationMaster</code> can use
081   * {@link #getContainerStatuses(GetContainerStatusesRequest)} to get updated
082   * statuses of the to-be-launched or launched containers.
083   * </p>
084   * 
085   * @param request
086   *          request to start a list of containers
087   * @return response including conatinerIds of all successfully launched
088   *         containers, a containerId-to-exception map for failed requests and
089   *         a allServicesMetaData map.
090   * @throws YarnException
091   * @throws IOException
092   * @throws NMNotYetReadyException
093   *           This exception is thrown when NM starts from scratch but has not
094   *           yet connected with RM.
095   */
096  @Public
097  @Stable
098  StartContainersResponse startContainers(StartContainersRequest request)
099      throws YarnException, IOException;
100
101  /**
102   * <p>
103   * The <code>ApplicationMaster</code> requests a <code>NodeManager</code> to
104   * <em>stop</em> a list of {@link Container}s allocated to it using this
105   * interface.
106   * </p>
107   * 
108   * <p>
109   * The <code>ApplicationMaster</code> sends a {@link StopContainersRequest}
110   * which includes the {@link ContainerId}s of the containers to be stopped.
111   * </p>
112   * 
113   * <p>
114   * The <code>NodeManager</code> sends a response via
115   * {@link StopContainersResponse} which includes a list of {@link ContainerId}
116   * s of successfully stopped containers, a containerId-to-exception map for
117   * each failed request in which the exception indicates errors from per
118   * container. Note: None-container-specific exceptions will still be thrown by
119   * the API method itself. <code>ApplicationMaster</code> can use
120   * {@link #getContainerStatuses(GetContainerStatusesRequest)} to get updated
121   * statuses of the containers.
122   * </p>
123   * 
124   * @param request
125   *          request to stop a list of containers
126   * @return response which includes a list of containerIds of successfully
127   *         stopped containers, a containerId-to-exception map for failed
128   *         requests.
129   * @throws YarnException
130   * @throws IOException
131   */
132  @Public
133  @Stable
134  StopContainersResponse stopContainers(StopContainersRequest request)
135      throws YarnException, IOException;
136
137  /**
138   * <p>
139   * The API used by the <code>ApplicationMaster</code> to request for current
140   * statuses of <code>Container</code>s from the <code>NodeManager</code>.
141   * </p>
142   * 
143   * <p>
144   * The <code>ApplicationMaster</code> sends a
145   * {@link GetContainerStatusesRequest} which includes the {@link ContainerId}s
146   * of all containers whose statuses are needed.
147   * </p>
148   * 
149   * <p>
150   * The <code>NodeManager</code> responds with
151   * {@link GetContainerStatusesResponse} which includes a list of
152   * {@link ContainerStatus} of the successfully queried containers and a
153   * containerId-to-exception map for each failed request in which the exception
154   * indicates errors from per container. Note: None-container-specific
155   * exceptions will still be thrown by the API method itself.
156   * </p>
157   * 
158   * @param request
159   *          request to get <code>ContainerStatus</code>es of containers with
160   *          the specified <code>ContainerId</code>s
161   * @return response containing the list of <code>ContainerStatus</code> of the
162   *         successfully queried containers and a containerId-to-exception map
163   *         for failed requests.
164   * 
165   * @throws YarnException
166   * @throws IOException
167   */
168  @Public
169  @Stable
170  GetContainerStatusesResponse getContainerStatuses(
171      GetContainerStatusesRequest request) throws YarnException,
172      IOException;
173}