[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526307#comment-13526307 ] Hudson commented on MAPREDUCE-4049: --- Integrated in Hadoop-Yarn-trunk #58 (See [https://builds.apache.org/job/Hadoop-Yarn-trunk/58/]) MAPREDUCE-4049. Experimental api to allow for alternate shuffle plugins. Contributed by Anver BenHanoch. (Revision 1418173) Result = SUCCESS acmurthy : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVNview=revrev=1418173 Files : * /hadoop/common/trunk/hadoop-mapreduce-project/CHANGES.txt * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ReduceTask.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ShuffleConsumerPlugin.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/MRConfig.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/Shuffle.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/resources/mapred-default.xml * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/TestShufflePlugin.java plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526365#comment-13526365 ] Hudson commented on MAPREDUCE-4049: --- Integrated in Hadoop-Hdfs-trunk #1247 (See [https://builds.apache.org/job/Hadoop-Hdfs-trunk/1247/]) MAPREDUCE-4049. Experimental api to allow for alternate shuffle plugins. Contributed by Anver BenHanoch. (Revision 1418173) Result = FAILURE acmurthy : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVNview=revrev=1418173 Files : * /hadoop/common/trunk/hadoop-mapreduce-project/CHANGES.txt * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ReduceTask.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ShuffleConsumerPlugin.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/MRConfig.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/Shuffle.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/resources/mapred-default.xml * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/TestShufflePlugin.java plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526401#comment-13526401 ] Hudson commented on MAPREDUCE-4049: --- Integrated in Hadoop-Mapreduce-trunk #1278 (See [https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1278/]) MAPREDUCE-4049. Experimental api to allow for alternate shuffle plugins. Contributed by Anver BenHanoch. (Revision 1418173) Result = SUCCESS acmurthy : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVNview=revrev=1418173 Files : * /hadoop/common/trunk/hadoop-mapreduce-project/CHANGES.txt * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ReduceTask.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/ShuffleConsumerPlugin.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/MRConfig.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/Shuffle.java * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/resources/mapred-default.xml * /hadoop/common/trunk/hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/TestShufflePlugin.java plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Tsuyoshi OZAWA updated MAPREDUCE-4502: -- Attachment: MAPREDUCE-4525-pof.diff I've prototyped proof of concept code of node-level aggregation. I made sure that it works well, so I share it! Current situation is as follows: * Node level aggregation in distributed mode is supported. I made sure of the correctness of the result by using diff command. * Fault tolerance is partially supported. At design level, it can be supported(WIP). * No test codes. * No switch on/off feature. Quick benchmark is as follows: |Default Hadoop(sec)|Node level aggregation(sec)| |105.0|82.0| Experiment Environment is as follows: * Input file is 1GB, generated by RandomTextWriter. * 8 CPU/machines. * 2 machines. I'm appreciate if you review the design and code. And, I have several questions to deal with: 1. How can I write tests such a complex case? Any ideas or references? 2. I've extended mr_protos.proto. Is it acceptable? My next step is creating patch sets with tests and fully fault tolerance support to merge it into trunk. I'm also planning to benchmark of node-level aggregation. I'll show you here when I do it. Note that design note is outdated, so I'm updating it now. If you'd like to track changes my code changes, you can see it on [my github|https://github.com/oza/hadoop-common/tree/MAPREDUCE-4525-Node-level-aggregation]. Multi-level aggregation with combining the result of maps per node/rack --- Key: MAPREDUCE-4502 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 Project: Hadoop Map/Reduce Issue Type: Improvement Components: applicationmaster, mrv2 Reporter: Tsuyoshi OZAWA Assignee: Tsuyoshi OZAWA Attachments: design_v2.pdf, MAPREDUCE-4525-pof.diff, speculative_draft.pdf The shuffle costs is expensive in Hadoop in spite of the existence of combiner, because the scope of combining is limited within only one MapTask. To solve this problem, it's a good way to aggregate the result of maps per node/rack by launch combiner. This JIRA is to implement the multi-level aggregation infrastructure, including combining per container(MAPREDUCE-3902 is related), coordinating containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4808) Allow reduce-side merge to be pluggable
[ https://issues.apache.org/jira/browse/MAPREDUCE-4808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526506#comment-13526506 ] Arun C Murthy commented on MAPREDUCE-4808: -- On second thoughts, a simpler solution: why don't we use a simplified version of Merger as the interface? This way Syncsort can just implement that, why bother with trying to deal with memory reservation etc.? Allow reduce-side merge to be pluggable --- Key: MAPREDUCE-4808 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4808 Project: Hadoop Map/Reduce Issue Type: Sub-task Affects Versions: 2.0.2-alpha Reporter: Arun C Murthy Assignee: Mariappan Asokan Fix For: 2.0.3-alpha Attachments: COMBO-mapreduce-4809-4812-4808.patch, mapreduce-4808.patch Allow reduce-side merge to be pluggable for MAPREDUCE-2454 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526516#comment-13526516 ] Arun C Murthy commented on MAPREDUCE-4049: -- [~tucu00] I'm confused. MAPREDUCE-2454 was a *huge* patch by a different contributor which got broken up to aid through reviews. MAPREDUCE-4049 is, effectively, a trivial patch from another contributor after he (Avner) has very patiently taken in all feedback. We should be thankful. I don't see why we need to block Avner's work on MAPREDUCE-2454. Furthermore, Avner has made it crystal clear that he has issues working with people working on MAPREDUCE-2454 (see http://s.apache.org/MRT, http://s.apache.org/6bh, http://s.apache.org/fR4). Why coerce him? Do you have any technical reason to revert the commit? Else, we can close this discussion. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526565#comment-13526565 ] Alejandro Abdelnur commented on MAPREDUCE-4049: --- Arun, focusing on the technical side of your comments. My reasons to revert the patch from trunk are: All these components are highly interrelated as you know. During the review of MAPREDUCE-4049 we found inconsistencies in the naming and we aligned them with the other sub-tasks. We may need to do some more of that. This was your motivation to create MAPREDUCE-2454 branch after a similar comment I've made in MAPREDUCE-4809. You want to have gridmix runs in a reasonable size cluster to ensure there are not performance degradation due to the subtasks of MAPREDUCE-2454. I don' t see why MAPREDUCE-4049 should be excluded from those tests. Personally I think this is not needed for any of the patches as a change from 'new' to 'ReflectionUtils.newInstance()' outside of the processing loop cannot affect things, but you strongly asked me for this over the phone. Thus, I think your 'requirements' for the other tasks to MAPREDUCE-2454 do also apply to MAPREDUCE-4049 and until they are satisfied, MAPREDUCE-2454 is not ready for going to trunk. Said this, again, please revert. I'm confident we can do a last push and get the branch MAPREDUCE-2454 merge into trunk at fast pace. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526579#comment-13526579 ] Arun C Murthy commented on MAPREDUCE-4049: -- I'm getting tired of the lawyering. MAPREDUCE-2454 is a couple of orders of magnitude larger than MAPREDUCE-4049. Anyway to stop wasting my time arguing, I just started a 300-node gridmix run with MAPREDUCE-4049, I'll report back by eod. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun C Murthy updated MAPREDUCE-4049: - Issue Type: Improvement (was: Sub-task) Parent: (was: MAPREDUCE-2454) plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526598#comment-13526598 ] Alejandro Abdelnur commented on MAPREDUCE-4049: --- Sounds good on the 300 node gridmix test, mind running a full MR-2454 patch if Asokan posts a updated patch soon? Still this is not addressing: bq. During the review of MAPREDUCE-4049 we found inconsistencies in the naming and we aligned them with the other sub-tasks. We may need to do some more of that. This was your motivation to create MAPREDUCE-2454 branch after a similar comment I've made in MAPREDUCE-4809. And this was significant enough for you to create MR-2454 branch. So again, revert MAPREDUCE-4049 from trunk until we iron out the whole branch. PS: the common ratio for order of magnitude is 10. I think you are a bit off with your comment. Lets keep things objective. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (MAPREDUCE-4859) TestRecoveryManager fails on branch-1
Arun C Murthy created MAPREDUCE-4859: Summary: TestRecoveryManager fails on branch-1 Key: MAPREDUCE-4859 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4859 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Looks like the tests are extremely flaky and just hang. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4858) TestWebUIAuthorization fails on branch-1
[ https://issues.apache.org/jira/browse/MAPREDUCE-4858?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526648#comment-13526648 ] Arun C Murthy commented on MAPREDUCE-4858: -- FYI - I had to add a minor addendum to fix the merge since HADOOP-8988 isn't in branch-1.1. My bad. {noformat} Index: src/test/org/apache/hadoop/mapred/TestWebUIAuthorization.java === --- src/test/org/apache/hadoop/mapred/TestWebUIAuthorization.java (revision 1418203) +++ src/test/org/apache/hadoop/mapred/TestWebUIAuthorization.java (working copy) @@ -782,9 +782,6 @@ true); props.setProperty(PolicyProvider.POLICY_PROVIDER_CONFIG, HadoopPolicyProvider.class.getName()); -props.setProperty( - CommonConfigurationKeys.HADOOP_SECURITY_INSTRUMENTATION_REQUIRES_ADMIN, -true); props.setProperty(JobConf.MR_ADMINS, mrAdminUser + + mrAdminGroup); startCluster(true, props); {noformat} TestWebUIAuthorization fails on branch-1 Key: MAPREDUCE-4858 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4858 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Attachments: MAPREDUCE-4858.patch, MAPREDUCE-4858.patch TestWebUIAuthorization fails on branch-1 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526653#comment-13526653 ] Alejandro Abdelnur commented on MAPREDUCE-4049: --- bq. During the review of MAPREDUCE-4049 we found inconsistencies in the naming and we aligned them with the other sub-tasks. We may need to do some more of that. This was your motivation to create MAPREDUCE-2454 branch after a similar comment I've made in MAPREDUCE-4809. Arun, until we address this, are you reverting the patch? Or I'll have to do it? plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526677#comment-13526677 ] Arun C Murthy commented on MAPREDUCE-4049: -- Alejandro - you committed the patch. I'm confused. We can change them via MAPREDUCE-2454 if you are pedantic, or file a follow on issue. I've already started the runs, I'm not going to be able to get cluster time to re-run them. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526684#comment-13526684 ] Arun C Murthy commented on MAPREDUCE-4049: -- I'm done discussing. Can we please move on? File a follow on issue for naming nits. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
Karthik Kambatla created MAPREDUCE-4860: --- Summary: Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 2.0.2-alpha, 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526697#comment-13526697 ] Karthik Kambatla commented on MAPREDUCE-4860: - In fact, I wonder why we can't use common.DelegationTokenRenewer for token renewal. That avoids duplicating code too. Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1, 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4860: Description: mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} was: mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1, 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4812) Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle
[ https://issues.apache.org/jira/browse/MAPREDUCE-4812?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526707#comment-13526707 ] Arun C Murthy commented on MAPREDUCE-4812: -- Alejandro - since you +1ed this jira, can you pls explain the rationale for the apis I've asked in MAPREDUCE-4808? Maybe you and Asokan had an offline conversation? Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle --- Key: MAPREDUCE-4812 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4812 Project: Hadoop Map/Reduce Issue Type: Sub-task Affects Versions: 2.0.2-alpha Reporter: Mariappan Asokan Assignee: Mariappan Asokan Fix For: 2.0.3-alpha Attachments: COMBO-mapreduce-4809-4812.patch, COMBO-mapreduce-4809-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch This is part of MAPREDUCE-2454. This further breaks down MAPREDUCE-4808 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4859) TestRecoveryManager fails on branch-1
[ https://issues.apache.org/jira/browse/MAPREDUCE-4859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526721#comment-13526721 ] Arun C Murthy commented on MAPREDUCE-4859: -- Sigh, I give up. TestRecoveryManager is hopeless. Mainly in the sense that it uses the confounded UtilsForTests which are broken. testJobTrackerRestartsWithMissingJobFile testJobTrackerRestartWithBadJobs *hang* on both Linux and MacOSX. testJobResubmission works on MacOSX and hangs on Linux similar to the other two. I managed to track and fix one bug in testJobTrackerInfoCreation. I'll ignore them for 1.1.2 (sad to have a stable release with unit-test failures due to flaky test code) so we can revisit them. TestRecoveryManager fails on branch-1 - Key: MAPREDUCE-4859 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4859 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Attachments: MAPREDUCE-4859.patch Looks like the tests are extremely flaky and just hang. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4859) TestRecoveryManager fails on branch-1
[ https://issues.apache.org/jira/browse/MAPREDUCE-4859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun C Murthy updated MAPREDUCE-4859: - Attachment: MAPREDUCE-4859.patch Fixed testJobTrackerInfoCreation, ignored others after converting them to junit4. TestRecoveryManager fails on branch-1 - Key: MAPREDUCE-4859 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4859 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Attachments: MAPREDUCE-4859.patch Looks like the tests are extremely flaky and just hang. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4859) TestRecoveryManager fails on branch-1
[ https://issues.apache.org/jira/browse/MAPREDUCE-4859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526725#comment-13526725 ] Matt Foley commented on MAPREDUCE-4859: --- +1. Please commit to branch-1 and branch-1.1. Thanks, Arun! TestRecoveryManager fails on branch-1 - Key: MAPREDUCE-4859 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4859 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Attachments: MAPREDUCE-4859.patch Looks like the tests are extremely flaky and just hang. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (MAPREDUCE-4859) TestRecoveryManager fails on branch-1
[ https://issues.apache.org/jira/browse/MAPREDUCE-4859?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun C Murthy resolved MAPREDUCE-4859. -- Resolution: Fixed I just committed this. Thanks for the quick check Matt! TestRecoveryManager fails on branch-1 - Key: MAPREDUCE-4859 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4859 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 1.1.1 Reporter: Arun C Murthy Assignee: Arun C Murthy Fix For: 1.1.2 Attachments: MAPREDUCE-4859.patch Looks like the tests are extremely flaky and just hang. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526773#comment-13526773 ] Alejandro Abdelnur commented on MAPREDUCE-4049: --- My +1 was for the work in the branch. My rationale it only echoing yours: https://issues.apache.org/jira/browse/MAPREDUCE-4809?focusedCommentId=13501245page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13501245 Arun C Murthy added a comment - 20/Nov/12 07:50 I've also created a MR-2454 branch in svn, let's commit to that branch first. This way we can change our mind before we do the final merge if necessary. MAPREDUCE-4812 will require some changes in MAPREDUCE-4049 So, revert from trunk, thx. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Improvement Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-1700) User supplied dependencies may conflict with MapReduce system JARs
[ https://issues.apache.org/jira/browse/MAPREDUCE-1700?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526775#comment-13526775 ] Kihwal Lee commented on MAPREDUCE-1700: --- {quote} bq. Tom, one thing I've forgot to mention in my previous comment, we should see how to enable the classloader on the client side as well as it may be required (to use different JARs) for the submission code. I think this is a slightly different problem, since users generally have more control over the JVM they submit from than the JVM the task runs in. So, yes, another JIRA would be appropriate. {quote} I think AM also runs user code, if a custom output format is defined. User supplied dependencies may conflict with MapReduce system JARs -- Key: MAPREDUCE-1700 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1700 Project: Hadoop Map/Reduce Issue Type: Bug Components: task Reporter: Tom White Assignee: Tom White Attachments: MAPREDUCE-1700-ccl.patch, MAPREDUCE-1700-ccl.patch, MAPREDUCE-1700.patch, MAPREDUCE-1700.patch, MAPREDUCE-1700.patch, MAPREDUCE-1700.patch, MAPREDUCE-1700.patch, MAPREDUCE-1700.patch If user code has a dependency on a version of a JAR that is different to the one that happens to be used by Hadoop, then it may not work correctly. This happened with user code using a different version of Avro, as reported [here|https://issues.apache.org/jira/browse/AVRO-493?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12852081#action_12852081]. The problem is analogous to the one that application servers have with WAR loading. Using a specialized classloader in the Child JVM is probably the way to solve this. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4812) Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle
[ https://issues.apache.org/jira/browse/MAPREDUCE-4812?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526783#comment-13526783 ] Alejandro Abdelnur commented on MAPREDUCE-4812: --- MAPREDUCE-4812 (this JIRA) makes the MergerManager is made pluggable, nothing else. In the current code, in the local case where Shuffle is not involved, the MergerManager is not being used; instead the Merger class is used directly. With MAPREDUCE-4808, the MergerManager is augmented to also handle the local case by adding a new method and moving the Merger.merge() invocation to it. MAPREDUCE-4808 introduces the merge pluggability for the local case when shuffle is not in the picture. My understanding is that Asokan split the pluggability of MergeManager from augmenting its functionality to handle the local case to keep the changes focused. I'm OK with folding both in a single JIRA if you think it makes more sense.. Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle --- Key: MAPREDUCE-4812 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4812 Project: Hadoop Map/Reduce Issue Type: Sub-task Affects Versions: 2.0.2-alpha Reporter: Mariappan Asokan Assignee: Mariappan Asokan Fix For: 2.0.3-alpha Attachments: COMBO-mapreduce-4809-4812.patch, COMBO-mapreduce-4809-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch This is part of MAPREDUCE-2454. This further breaks down MAPREDUCE-4808 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4795) TestDelegationTokenRenewal should not use static variables in Renewer
[ https://issues.apache.org/jira/browse/MAPREDUCE-4795?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4795: Resolution: Not A Problem Status: Resolved (was: Patch Available) Spoke to Sandy offline about this: the static variables in renewer don't seem to be causing the test failures we noticed. TestDelegationTokenRenewal should not use static variables in Renewer - Key: MAPREDUCE-4795 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4795 Project: Hadoop Map/Reduce Issue Type: Improvement Affects Versions: 1.0.4 Reporter: Sandy Ryza Assignee: Sandy Ryza Attachments: MAPREDUCE-4795-branch-1.patch TestDelegationTokenRenewal uses static variables to access what's going on inside its Renewer class, making it so problems can occur if the tests are run in parallel. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4808) Allow reduce-side merge to be pluggable
[ https://issues.apache.org/jira/browse/MAPREDUCE-4808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526795#comment-13526795 ] Alejandro Abdelnur commented on MAPREDUCE-4808: --- Arun, what do you exactly mean? Both Merger and MergeManager are classes. The reducer side of logic in the Merger class is used only in 2 places, within the MergeManager (used by the Shuffle) and in the ReduceTask (for the local case). This patch is moving the later use into the MergeManager as well. By doing this all reduce merge logic, for the local and the distributed case, is encapsulated in the MergerManager. Then, when an alternate implementation is provided, it can handle both cases, local and distributed case. The resulting interface being introduced as ReduceInputMerger is quite simple: {code} public void init(ContextK, V reduceMergerContext); public void waitForResource() throws InterruptedException; public MapOutputK, V reserve(TaskAttemptID mapId, long requestedSize, int fetcher) throws IOException; public RawKeyValueIterator close() throws Throwable; // To merge files created for a local job. public RawKeyValueIterator mergeLocalFiles(Path localFiles[]) throws IOException; {code} I think this is much simpler than trying modify things in the Merger, given that the merger is not directly used by the Shuffle, but through the MergeManager. Allow reduce-side merge to be pluggable --- Key: MAPREDUCE-4808 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4808 Project: Hadoop Map/Reduce Issue Type: Sub-task Affects Versions: 2.0.2-alpha Reporter: Arun C Murthy Assignee: Mariappan Asokan Fix For: 2.0.3-alpha Attachments: COMBO-mapreduce-4809-4812-4808.patch, mapreduce-4808.patch Allow reduce-side merge to be pluggable for MAPREDUCE-2454 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4812) Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle
[ https://issues.apache.org/jira/browse/MAPREDUCE-4812?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526802#comment-13526802 ] Mariappan Asokan commented on MAPREDUCE-4812: - Hi Arun, Sorry I did not get back sooner. The intention of {{ReduceInputMerger}} interface is to have a pluggable {{MergeManager}} implementation. For a non-local job, {{Shuffle}} and {{MergeManager}} interact and synchronize with each other using the three methods {{waitForInMemoryMerge(),}} {{reserve(),}} and {{close()}}. So in order to use the {{Shuffle}} these methods are captured in {{ReduceInputMerger}} interface. I renamed {{waitForInMemoryMerge()}} to a generic name {{waitForResource()}} since the plugin implementation may not have the concept of in-memory merge. Since the return value from {{reserve()}} is {{MapOutput}}, I did some refactoring of {{MapOutput}} so that plugin can return its own implementation of it. I kept the refactoring done on {{MapOutput}} in MAPREDUCE-4808. With just MAPREDUCE-4812, an external plugin is not possible, but it has the core part of the concepts so that it is easy to review just {{ReduceInputMerger}} design. Similarly, for a local job the input is coming from local files. I enhanced {{ReduceInputMerger}} with one more method for this. It is also kept in MAPREDUCE-4808. Hope I explained well. Please let me know if you have any more questions. Thanks. -- Asokan Create reduce input merger plugin in ReduceTask.java and pass it to Shuffle --- Key: MAPREDUCE-4812 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4812 Project: Hadoop Map/Reduce Issue Type: Sub-task Affects Versions: 2.0.2-alpha Reporter: Mariappan Asokan Assignee: Mariappan Asokan Fix For: 2.0.3-alpha Attachments: COMBO-mapreduce-4809-4812.patch, COMBO-mapreduce-4809-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch, mapreduce-4812.patch This is part of MAPREDUCE-2454. This further breaks down MAPREDUCE-4808 -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4860: Affects Version/s: (was: 2.0.2-alpha) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (MAPREDUCE-4861) Cleanup: mapreduce.security.token.DelegationTokenRenewal is not used in branch-2. Can remove it.
Karthik Kambatla created MAPREDUCE-4861: --- Summary: Cleanup: mapreduce.security.token.DelegationTokenRenewal is not used in branch-2. Can remove it. Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4861: Summary: Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal (was: Cleanup: mapreduce.security.token.DelegationTokenRenewal is not used in branch-2. Can remove it.) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4861: Status: Patch Available (was: Open) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4861: Attachment: mr-4861.patch Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526818#comment-13526818 ] Sandy Ryza commented on MAPREDUCE-4861: --- +1 Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-461) Enable ServicePlugins for the JobTracker
[ https://issues.apache.org/jira/browse/MAPREDUCE-461?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Arun C Murthy updated MAPREDUCE-461: Fix Version/s: 1.2.0 Enable ServicePlugins for the JobTracker Key: MAPREDUCE-461 URL: https://issues.apache.org/jira/browse/MAPREDUCE-461 Project: Hadoop Map/Reduce Issue Type: New Feature Reporter: Fredrik Hedberg Assignee: Fredrik Hedberg Priority: Minor Fix For: 0.23.0, 1.2.0 Attachments: MAPREDUCE-461.branch-1.patch, MAPREDUCE-461.patch, sp-jt-1.diff Allow ServicePlugins (see HADOOP-5257) for the JobTracker. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526841#comment-13526841 ] Daryn Sharp commented on MAPREDUCE-4861: Presumably the code was copied elsewhere? If so, were the tests also copied? Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526856#comment-13526856 ] Hadoop QA commented on MAPREDUCE-4861: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12559961/mr-4861.patch against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:red}-1 tests included{color}. The patch doesn't appear to include any new or modified tests. Please justify why no new tests are needed for this patch. Also please list what manual steps were performed to verify this patch. {color:green}+1 javac{color}. The applied patch does not increase the total number of javac compiler warnings. {color:green}+1 javadoc{color}. The javadoc tool did not generate any warning messages. {color:green}+1 eclipse:eclipse{color}. The patch built with eclipse:eclipse. {color:green}+1 findbugs{color}. The patch does not introduce any new Findbugs (version 1.3.9) warnings. {color:green}+1 release audit{color}. The applied patch does not increase the total number of release audit warnings. {color:green}+1 core tests{color}. The patch passed unit tests in hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-jobclient. {color:green}+1 contrib tests{color}. The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3108//testReport/ Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3108//console This message is automatically generated. Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4860: Attachment: mr-4860.patch Taking a closer look, the problem seems to be with the use of TimerTask/Timer. If a task is already scheduled, calling task.cancel() from outside the task doesn't stop the scheduled task, but prevents the task from getting scheduled another time. However, the task can cancel itself. The patch: # adds a check within the task if it has been stopped from outside # synchronizes changes to delegationTokens Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4860.patch mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526868#comment-13526868 ] Karthik Kambatla commented on MAPREDUCE-4861: - Daryn, the code has been copied to yarn.server.rm.security.DelegationTokenRenewer and the corresponding tests have been copied to rm.security.TestDelegationTokenRenewer (except one test - testing renewal after close()). Do you suggest I add the missing test before deleting these? The copied code suffers from issues similar to MAPREDUCE-4860. However, I still think o.a.h.fs.DelegationTokenRenewer can be re-used here instead. Comments? Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4861: Attachment: mr-4861.patch Updated the patch to copy testDTRenewalAfterClose to TestDelegationTokenRenewer Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch, mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4860: Status: Patch Available (was: Open) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4860.patch mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526886#comment-13526886 ] Hadoop QA commented on MAPREDUCE-4860: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12559974/mr-4860.patch against trunk revision . {color:red}-1 patch{color}. The patch command could not apply the patch. Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3110//console This message is automatically generated. Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4860.patch mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Created] (MAPREDUCE-4862) Inconsistent synchronization in yarn.server.rm.security.DelegationTokenRenewal
Karthik Kambatla created MAPREDUCE-4862: --- Summary: Inconsistent synchronization in yarn.server.rm.security.DelegationTokenRenewal Key: MAPREDUCE-4862 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4862 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4862) Inconsistent synchronization in yarn.server.rm.security.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4862?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4862: Description: yarn.s.rm.security.DelegationTokenRenewer uses TimerTask/Timer. When such a timer task is canceled, already scheduled tasks run to completion. The task should check for such cancellation before running. Also, delegationTokens needs to be synchronized on all accesses. was: mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} Inconsistent synchronization in yarn.server.rm.security.DelegationTokenRenewal -- Key: MAPREDUCE-4862 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4862 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla yarn.s.rm.security.DelegationTokenRenewer uses TimerTask/Timer. When such a timer task is canceled, already scheduled tasks run to completion. The task should check for such cancellation before running. Also, delegationTokens needs to be synchronized on all accesses. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4860: Attachment: mr-4860.patch Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4860.patch, mr-4860.patch mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526923#comment-13526923 ] Hadoop QA commented on MAPREDUCE-4861: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12559978/mr-4861.patch against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:green}+1 tests included{color}. The patch appears to include 1 new or modified test files. {color:green}+1 javac{color}. The applied patch does not increase the total number of javac compiler warnings. {color:green}+1 javadoc{color}. The javadoc tool did not generate any warning messages. {color:green}+1 eclipse:eclipse{color}. The patch built with eclipse:eclipse. {color:green}+1 findbugs{color}. The patch does not introduce any new Findbugs (version 1.3.9) warnings. {color:green}+1 release audit{color}. The applied patch does not increase the total number of release audit warnings. {color:red}-1 core tests{color}. The patch failed these unit tests in hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-jobclient hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager: org.apache.hadoop.yarn.server.resourcemanager.security.TestDelegationTokenRenewer {color:green}+1 contrib tests{color}. The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3109//testReport/ Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3109//console This message is automatically generated. Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch, mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4860) Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4860?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526924#comment-13526924 ] Hadoop QA commented on MAPREDUCE-4860: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12559989/mr-4860.patch against trunk revision . {color:red}-1 patch{color}. The patch command could not apply the patch. Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3111//console This message is automatically generated. Inconsistent synchronization in mapreduce.security.token.DelegationTokenRenewal --- Key: MAPREDUCE-4860 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4860 Project: Hadoop Map/Reduce Issue Type: Bug Components: security Affects Versions: 1.1.1 Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4860.patch, mr-4860.patch mapreduce.security.token.DelegationTokenRenewal synchronizes on removeDelegationToken, but fails to synchronize on addToken, and renewing tokens in run(). This inconsistency is exposed by frequent failures of TestDelegationTokenRenewal: {noformat} Error Message renew wasn't called as many times as expected expected:4 but was:5 Stacktrace junit.framework.AssertionFailedError: renew wasn't called as many times as expected expected:4 but was:5 at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewal(TestDelegationTokenRenewal.java:317) at org.apache.hadoop.mapreduce.security.token.TestDelegationTokenRenewal.testDTRenewalAfterClose(TestDelegationTokenRenewal.java:338) {noformat} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526945#comment-13526945 ] Karthik Kambatla commented on MAPREDUCE-4861: - Looks like I ran the wrong TestDelegationTokenRenewer locally before uploading. Taking a closer look, I don't think the new test applies to YARN as the Renewer can't be stopped and restarted. Shall upload the first patch again. Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch, mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated MAPREDUCE-4861: Attachment: mr-4861.patch Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch, mr-4861.patch, mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Alejandro Abdelnur updated MAPREDUCE-4049: -- Issue Type: Sub-task (was: Improvement) Parent: MAPREDUCE-2454 plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526953#comment-13526953 ] Alejandro Abdelnur commented on MAPREDUCE-4049: --- Arun, I'd appreciate not to remove again this JIRA from being a subtask of MAPREDUCE-2454. You are leaving with no choice but to *-1* this patch to go in trunk until the work in the branch is completed. My rationale for this -1 it is yours: https://issues.apache.org/jira/browse/MAPREDUCE-4809?focusedCommentId=13501245page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13501245 Arun C Murthy added a comment - 20/Nov/12 07:50 I've also created a MR-2454 branch in svn, let's commit to that branch first. This way we can change our mind before we do the final merge if necessary. In addition, as we may need to make further changes to MAPREDUCE-4049, I'd like to see the whole scope of changes before merging back to trunk. plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Reopened] (MAPREDUCE-4049) plugin for generic shuffle service
[ https://issues.apache.org/jira/browse/MAPREDUCE-4049?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Alejandro Abdelnur reopened MAPREDUCE-4049: --- plugin for generic shuffle service -- Key: MAPREDUCE-4049 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4049 Project: Hadoop Map/Reduce Issue Type: Sub-task Components: performance, task, tasktracker Affects Versions: 1.0.3, 1.1.0, 2.0.0-alpha, 3.0.0 Reporter: Avner BenHanoch Assignee: Avner BenHanoch Labels: merge, plugin, rdma, shuffle Fix For: 3.0.0 Attachments: HADOOP-1.x.y.patch, Hadoop Shuffle Plugin Design.rtf, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch, mapreduce-4049.patch Support generic shuffle service as set of two plugins: ShuffleProvider ShuffleConsumer. This will satisfy the following needs: # Better shuffle and merge performance. For example: we are working on shuffle plugin that performs shuffle over RDMA in fast networks (10gE, 40gE, or Infiniband) instead of using the current HTTP shuffle. Based on the fast RDMA shuffle, the plugin can also utilize a suitable merge approach during the intermediate merges. Hence, getting much better performance. # Satisfy MAPREDUCE-3060 - generic shuffle service for avoiding hidden dependency of NodeManager with a specific version of mapreduce shuffle (currently targeted to 0.24.0). References: # Hadoop Acceleration through Network Levitated Merging, by Prof. Weikuan Yu from Auburn University with others, [http://pasl.eng.auburn.edu/pubs/sc11-netlev.pdf] # I am attaching 2 documents with suggested Top Level Design for both plugins (currently, based on 1.0 branch) # I am providing link for downloading UDA - Mellanox's open source plugin that implements generic shuffle service using RDMA and levitated merge. Note: At this phase, the code is in C++ through JNI and you should consider it as beta only. Still, it can serve anyone that wants to implement or contribute to levitated merge. (Please be advised that levitated merge is mostly suit in very fast networks) - [http://www.mellanox.com/content/pages.php?pg=products_dynproduct_family=144menu_section=69] -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4861) Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal
[ https://issues.apache.org/jira/browse/MAPREDUCE-4861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526978#comment-13526978 ] Hadoop QA commented on MAPREDUCE-4861: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12559994/mr-4861.patch against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:red}-1 tests included{color}. The patch doesn't appear to include any new or modified tests. Please justify why no new tests are needed for this patch. Also please list what manual steps were performed to verify this patch. {color:green}+1 javac{color}. The applied patch does not increase the total number of javac compiler warnings. {color:green}+1 javadoc{color}. The javadoc tool did not generate any warning messages. {color:green}+1 eclipse:eclipse{color}. The patch built with eclipse:eclipse. {color:green}+1 findbugs{color}. The patch does not introduce any new Findbugs (version 1.3.9) warnings. {color:green}+1 release audit{color}. The applied patch does not increase the total number of release audit warnings. {color:green}+1 core tests{color}. The patch passed unit tests in hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-jobclient. {color:green}+1 contrib tests{color}. The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3112//testReport/ Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3112//console This message is automatically generated. Cleanup: Remove unused mapreduce.security.token.DelegationTokenRenewal -- Key: MAPREDUCE-4861 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4861 Project: Hadoop Map/Reduce Issue Type: Bug Affects Versions: 2.0.2-alpha Reporter: Karthik Kambatla Assignee: Karthik Kambatla Attachments: mr-4861.patch, mr-4861.patch, mr-4861.patch mapreduce.security.token.DelegationTokenRenewal doesn't seem to be used in branch-2 at all. grep on trunk yields no results, not even ReflectionUtils related suff. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4594) Add init/shutdown methods to mapreduce Partitioner
[ https://issues.apache.org/jira/browse/MAPREDUCE-4594?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Radim Kolar updated MAPREDUCE-4594: --- Attachment: partitioner2.txt now with unit tests Add init/shutdown methods to mapreduce Partitioner -- Key: MAPREDUCE-4594 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4594 Project: Hadoop Map/Reduce Issue Type: Improvement Components: client Affects Versions: trunk Reporter: Radim Kolar Attachments: partitioner1.txt, partitioner2.txt, partitioner2.txt The Partitioner supports only the Configurable API, which can be used for basic init in setConf(). Problem is that there is no shutdown function. I propose to use standard setup() cleanup() functions like in mapper / reducer. Use case is that I need to start and stop spring context and datagrid client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4594) Add init/shutdown methods to mapreduce Partitioner
[ https://issues.apache.org/jira/browse/MAPREDUCE-4594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13526992#comment-13526992 ] Hadoop QA commented on MAPREDUCE-4594: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12560007/partitioner2.txt against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:green}+1 tests included{color}. The patch appears to include 2 new or modified test files. {color:red}-1 javac{color:red}. The patch appears to cause the build to fail. Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3113//console This message is automatically generated. Add init/shutdown methods to mapreduce Partitioner -- Key: MAPREDUCE-4594 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4594 Project: Hadoop Map/Reduce Issue Type: Improvement Components: client Affects Versions: trunk Reporter: Radim Kolar Attachments: partitioner1.txt, partitioner2.txt, partitioner2.txt The Partitioner supports only the Configurable API, which can be used for basic init in setConf(). Problem is that there is no shutdown function. I propose to use standard setup() cleanup() functions like in mapper / reducer. Use case is that I need to start and stop spring context and datagrid client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4594) Add init/shutdown methods to mapreduce Partitioner
[ https://issues.apache.org/jira/browse/MAPREDUCE-4594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13527010#comment-13527010 ] Radim Kolar commented on MAPREDUCE-4594: hmm, it fails to compile with javac. In eclipse it works. Its problem with generic. i can not cast class Partitioner.class to Class? extends Partitioner?, ? because of type erasure. Add init/shutdown methods to mapreduce Partitioner -- Key: MAPREDUCE-4594 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4594 Project: Hadoop Map/Reduce Issue Type: Improvement Components: client Affects Versions: trunk Reporter: Radim Kolar Attachments: partitioner1.txt, partitioner2.txt, partitioner2.txt The Partitioner supports only the Configurable API, which can be used for basic init in setConf(). Problem is that there is no shutdown function. I propose to use standard setup() cleanup() functions like in mapper / reducer. Use case is that I need to start and stop spring context and datagrid client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MAPREDUCE-4594) Add init/shutdown methods to mapreduce Partitioner
[ https://issues.apache.org/jira/browse/MAPREDUCE-4594?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Radim Kolar updated MAPREDUCE-4594: --- Attachment: partitioner3.txt Add init/shutdown methods to mapreduce Partitioner -- Key: MAPREDUCE-4594 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4594 Project: Hadoop Map/Reduce Issue Type: Improvement Components: client Affects Versions: trunk Reporter: Radim Kolar Attachments: partitioner1.txt, partitioner2.txt, partitioner2.txt, partitioner3.txt The Partitioner supports only the Configurable API, which can be used for basic init in setConf(). Problem is that there is no shutdown function. I propose to use standard setup() cleanup() functions like in mapper / reducer. Use case is that I need to start and stop spring context and datagrid client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4594) Add init/shutdown methods to mapreduce Partitioner
[ https://issues.apache.org/jira/browse/MAPREDUCE-4594?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13527034#comment-13527034 ] Hadoop QA commented on MAPREDUCE-4594: -- {color:red}-1 overall{color}. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12560018/partitioner3.txt against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:green}+1 tests included{color}. The patch appears to include 2 new or modified test files. {color:green}+1 javac{color}. The applied patch does not increase the total number of javac compiler warnings. {color:green}+1 javadoc{color}. The javadoc tool did not generate any warning messages. {color:green}+1 eclipse:eclipse{color}. The patch built with eclipse:eclipse. {color:green}+1 findbugs{color}. The patch does not introduce any new Findbugs (version 1.3.9) warnings. {color:green}+1 release audit{color}. The applied patch does not increase the total number of release audit warnings. {color:red}-1 core tests{color}. The patch failed these unit tests in hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core: org.apache.hadoop.mapred.TestMapTask {color:green}+1 contrib tests{color}. The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3114//testReport/ Console output: https://builds.apache.org/job/PreCommit-MAPREDUCE-Build/3114//console This message is automatically generated. Add init/shutdown methods to mapreduce Partitioner -- Key: MAPREDUCE-4594 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4594 Project: Hadoop Map/Reduce Issue Type: Improvement Components: client Affects Versions: trunk Reporter: Radim Kolar Attachments: partitioner1.txt, partitioner2.txt, partitioner2.txt, partitioner3.txt The Partitioner supports only the Configurable API, which can be used for basic init in setConf(). Problem is that there is no shutdown function. I propose to use standard setup() cleanup() functions like in mapper / reducer. Use case is that I need to start and stop spring context and datagrid client. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MAPREDUCE-4502) Multi-level aggregation with combining the result of maps per node/rack
[ https://issues.apache.org/jira/browse/MAPREDUCE-4502?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13527037#comment-13527037 ] Tsuyoshi OZAWA commented on MAPREDUCE-4502: --- I'll create subtasks under MAPREDUCE-4525, and attach patches separately to make the review easy. Multi-level aggregation with combining the result of maps per node/rack --- Key: MAPREDUCE-4502 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4502 Project: Hadoop Map/Reduce Issue Type: Improvement Components: applicationmaster, mrv2 Reporter: Tsuyoshi OZAWA Assignee: Tsuyoshi OZAWA Attachments: design_v2.pdf, MAPREDUCE-4525-pof.diff, speculative_draft.pdf The shuffle costs is expensive in Hadoop in spite of the existence of combiner, because the scope of combining is limited within only one MapTask. To solve this problem, it's a good way to aggregate the result of maps per node/rack by launch combiner. This JIRA is to implement the multi-level aggregation infrastructure, including combining per container(MAPREDUCE-3902 is related), coordinating containers by application master without breaking fault tolerance of jobs. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira