Hadoop YARN
  1. Hadoop YARN
  2. YARN-1913

With Fair Scheduler, cluster can logjam when all resources are consumed by AMs

    Details

    • Type: Bug Bug
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: 2.3.0
    • Fix Version/s: 2.5.0
    • Component/s: scheduler
    • Labels:
      None

      Description

      It's possible to deadlock a cluster by submitting many applications at once, and have all cluster resources taken up by AMs.

      One solution is for the scheduler to limit resources taken up by AMs, as a percentage of total cluster resources, via a "maxApplicationMasterShare" config.

      1. YARN-1913.patch
        40 kB
        Wei Yan
      2. YARN-1913.patch
        40 kB
        Wei Yan
      3. YARN-1913.patch
        40 kB
        Wei Yan
      4. YARN-1913.patch
        36 kB
        Wei Yan
      5. YARN-1913.patch
        36 kB
        Wei Yan
      6. YARN-1913.patch
        87 kB
        Wei Yan
      7. YARN-1913.patch
        86 kB
        Wei Yan
      8. YARN-1913.patch
        34 kB
        Wei Yan
      9. YARN-1913.patch
        35 kB
        Wei Yan
      10. YARN-1913.patch
        35 kB
        Wei Yan
      11. YARN-1913.patch
        38 kB
        Wei Yan

        Activity

        bc Wong created issue -
        Hide
        Jason Lowe added a comment -

        Which scheduler are you using? The CapacityScheduler already has a yarn.scheduler.capacity.maximum-am-resource-percent property, and there's a per-queue form of it as well.

        Show
        Jason Lowe added a comment - Which scheduler are you using? The CapacityScheduler already has a yarn.scheduler.capacity.maximum-am-resource-percent property, and there's a per-queue form of it as well.
        Sandy Ryza made changes -
        Field Original Value New Value
        Summary Cluster logjam when all resources are consumed by AM With Fair Scheduler, cluster can logjam when all resources are consumed by AMs
        Hide
        Sandy Ryza added a comment -

        This is a Fair Scheduler issue. We need to add an equivalent property to the Fair Scheduler.

        Show
        Sandy Ryza added a comment - This is a Fair Scheduler issue. We need to add an equivalent property to the Fair Scheduler.
        Karthik Kambatla (Inactive) made changes -
        Assignee Karthik Kambatla [ kkambatl ]
        Hide
        Wei Yan added a comment -

        init patch for review.
        Add queueMaxAMShare configuration for each queue. And update the code in MaxRunningAppsEnforcer.java by considering AM share. Instead of using accurate AM resource usage, here use an easier way. The max_app_limited_by_AM = (queue.queueMaxAMShare * queue.maxShare) / scheduler.minAllocation.

        Show
        Wei Yan added a comment - init patch for review. Add queueMaxAMShare configuration for each queue. And update the code in MaxRunningAppsEnforcer.java by considering AM share. Instead of using accurate AM resource usage, here use an easier way. The max_app_limited_by_AM = (queue.queueMaxAMShare * queue.maxShare) / scheduler.minAllocation.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12646146 ]
        Wei Yan made changes -
        Assignee Karthik Kambatla [ kkambatl ] Wei Yan [ ywskycn ]
        Wei Yan made changes -
        Status Open [ 1 ] Patch Available [ 10002 ]
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12646404 ]
        Wei Yan made changes -
        Status Patch Available [ 10002 ] Open [ 1 ]
        Wei Yan made changes -
        Status Open [ 1 ] Patch Available [ 10002 ]
        Hide
        Hadoop QA added a comment -

        -1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12646404/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        -1 core tests. The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager:

        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFairScheduler

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3816//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3816//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - -1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12646404/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. -1 core tests . The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager: org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFairScheduler +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3816//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3816//console This message is automatically generated.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12646641 ]
        Hide
        Hadoop QA added a comment -

        -1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12646641/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        -1 core tests. The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager:

        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFairScheduler

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3822//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3822//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - -1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12646641/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. -1 core tests . The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager: org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFairScheduler +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3822//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3822//console This message is automatically generated.
        Xu Yang made changes -
        Labels easyfix
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647029 ]
        Hide
        Hadoop QA added a comment -

        +1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647029/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3842//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3842//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647029/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3842//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3842//console This message is automatically generated.
        Hide
        Sandy Ryza added a comment -

        I think we should avoid doing approximate calculation through the minimum allocation. We need to handle situations where AM resources are much larger than the min, and situations where the minimum allocation will be 0 (common on Llama-enabled clusters).

        This would have the added benefit of avoiding touching the "runnability" machinery, which is already bordering on over-complicated.

        Show
        Sandy Ryza added a comment - I think we should avoid doing approximate calculation through the minimum allocation. We need to handle situations where AM resources are much larger than the min, and situations where the minimum allocation will be 0 (common on Llama-enabled clusters). This would have the added benefit of avoiding touching the "runnability" machinery, which is already bordering on over-complicated.
        Hide
        Wei Yan added a comment -

        Thanks, Sandy.
        Upload a new patch using the accurate way.

        Show
        Wei Yan added a comment - Thanks, Sandy. Upload a new patch using the accurate way.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647487 ]
        Wei Yan made changes -
        Labels easyfix
        Hide
        Hadoop QA added a comment -

        +1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647487/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 4 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3860//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3860//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647487/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 4 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3860//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3860//console This message is automatically generated.
        Hide
        Sandy Ryza added a comment -

        Thanks for the updated patch Wei. For queues, maxAMShare should be defined as a fraction of the queue's fair share, not maxShare. The majority of queues are configured with infinite maxResources. We need to be careful with this, as fair shares can change when queues are created dynamically.

        I think it might make sense to only allow the queue-level maxAMShare on leaf queues for the moment. I can't think of a strong reason somebody would want to set it on a parent queue, and doing this would allow us to avoid the complex logic in MaxRunningAppsEnforcer, and merely enforce the AM max share by checking in AppSchedulable.assignContainer. This is also what the Capacity Scheduler has at the moment.

        Show
        Sandy Ryza added a comment - Thanks for the updated patch Wei. For queues, maxAMShare should be defined as a fraction of the queue's fair share, not maxShare. The majority of queues are configured with infinite maxResources. We need to be careful with this, as fair shares can change when queues are created dynamically. I think it might make sense to only allow the queue-level maxAMShare on leaf queues for the moment. I can't think of a strong reason somebody would want to set it on a parent queue, and doing this would allow us to avoid the complex logic in MaxRunningAppsEnforcer, and merely enforce the AM max share by checking in AppSchedulable.assignContainer. This is also what the Capacity Scheduler has at the moment.
        Hide
        Wei Yan added a comment -

        Thanks, Sandy. Will update a patch.

        Show
        Wei Yan added a comment - Thanks, Sandy. Will update a patch.
        Hide
        Ashwin Shankar added a comment -

        Hey Sandy Ryza, quick comment

        I think it might make sense to only allow the queue-level maxAMShare on leaf queues for the moment. I can't think of a strong reason somebody would want to set it on a parent queue

        For NestedUserQueue rule, user queues would be created dynamically under a parent. For this use case,
        maxAMShare at the parent would be useful, since leaf user queues are not configured in the alloc xml.
        I see your point that it would complicate the logic at MaxRunningAppsEnforcer,but just wanted to bring this up in case you
        didn't consider this use case.

        Show
        Ashwin Shankar added a comment - Hey Sandy Ryza , quick comment I think it might make sense to only allow the queue-level maxAMShare on leaf queues for the moment. I can't think of a strong reason somebody would want to set it on a parent queue For NestedUserQueue rule, user queues would be created dynamically under a parent. For this use case, maxAMShare at the parent would be useful, since leaf user queues are not configured in the alloc xml. I see your point that it would complicate the logic at MaxRunningAppsEnforcer,but just wanted to bring this up in case you didn't consider this use case.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647733 ]
        Hide
        Wei Yan added a comment -

        Update a new patch to fix Sandy's comments.
        Ashwin Shankar, if the leaf queue is not configured, the default AM resource limit is (leaf_queue_fair_share * 1.0f), still limited by its fair share.

        Show
        Wei Yan added a comment - Update a new patch to fix Sandy's comments. Ashwin Shankar , if the leaf queue is not configured, the default AM resource limit is (leaf_queue_fair_share * 1.0f), still limited by its fair share.
        Hide
        Hadoop QA added a comment -

        +1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647733/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 4 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3872//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3872//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647733/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 4 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3872//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3872//console This message is automatically generated.
        Hide
        Sandy Ryza added a comment -

        The primary benefit of the logic in MaxRunningAppsEnforcer is that it allows us to enforce maxRunningApps constraints from queues higher up in the hierarchy, and integrate these with user maxRunningApps constraints

        As we won't have these issues for queue maxAMShares, I think we can avoid touching MaxRunningAppsEnforcer entirely and just do the checking inside AppSchedulable.assignContainer or FSLeafQueue.assignContainer.

        Show
        Sandy Ryza added a comment - The primary benefit of the logic in MaxRunningAppsEnforcer is that it allows us to enforce maxRunningApps constraints from queues higher up in the hierarchy, and integrate these with user maxRunningApps constraints As we won't have these issues for queue maxAMShares, I think we can avoid touching MaxRunningAppsEnforcer entirely and just do the checking inside AppSchedulable.assignContainer or FSLeafQueue.assignContainer.
        Hide
        Wei Yan added a comment -

        Thanks, Sandy.
        Upload a new patch, move the AM resource usage check to AppSchedulable.

        Show
        Wei Yan added a comment - Thanks, Sandy. Upload a new patch, move the AM resource usage check to AppSchedulable.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647873 ]
        Hide
        Hadoop QA added a comment -

        -1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647873/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 2 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        -1 findbugs. The patch appears to introduce 1 new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3884//testReport/
        Findbugs warnings: https://builds.apache.org/job/PreCommit-YARN-Build/3884//artifact/trunk/patchprocess/newPatchFindbugsWarningshadoop-yarn-server-resourcemanager.html
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3884//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - -1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647873/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 2 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. -1 findbugs . The patch appears to introduce 1 new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3884//testReport/ Findbugs warnings: https://builds.apache.org/job/PreCommit-YARN-Build/3884//artifact/trunk/patchprocess/newPatchFindbugsWarningshadoop-yarn-server-resourcemanager.html Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3884//console This message is automatically generated.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647905 ]
        Hide
        Hadoop QA added a comment -

        +1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647905/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 2 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3887//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3887//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647905/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 2 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3887//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3887//console This message is automatically generated.
        Hide
        Sandy Ryza added a comment -

        This is looking good. A small things.

        AppSchedulingInfo is only used to track pending resources. We should hold amResource in SchedulerApplicationAttempt.

        +          if (! queue.canRunAppAM(app.getAMResource())) {
        

        Take out space after exclamation point.

           @Override
        +  public boolean checkIfAMResourceUsageOverLimit(Resource usage, Resource maxAMResource) {
        +    return Resources.greaterThan(RESOURCE_CALCULATOR, null, usage, maxAMResource);
        +  }
        

        Simpler to just use "usage.getMemory() > maxAMResource.getMemory()".

        +      if (request.getPriority().equals(RMAppAttemptImpl.AM_CONTAINER_PRIORITY)) {
        

        I'm a little nervous about using the priority here because apps could unwittingly submit all requests at that priority. Can we use SchedulerApplicationAttempt.getLiveContainers().isEmpty()?

        Show
        Sandy Ryza added a comment - This is looking good. A small things. AppSchedulingInfo is only used to track pending resources. We should hold amResource in SchedulerApplicationAttempt. + if (! queue.canRunAppAM(app.getAMResource())) { Take out space after exclamation point. @Override + public boolean checkIfAMResourceUsageOverLimit(Resource usage, Resource maxAMResource) { + return Resources.greaterThan(RESOURCE_CALCULATOR, null , usage, maxAMResource); + } Simpler to just use "usage.getMemory() > maxAMResource.getMemory()". + if (request.getPriority().equals(RMAppAttemptImpl.AM_CONTAINER_PRIORITY)) { I'm a little nervous about using the priority here because apps could unwittingly submit all requests at that priority. Can we use SchedulerApplicationAttempt.getLiveContainers().isEmpty()?
        Hide
        Wei Yan added a comment -

        Thanks, Sandy.
        One problem may exist if we use SchedulerApplicationAttempt.getLiveContainers().isEmpty(), if the application is unManagedAM, it will not generate an AM resource request. Thus, the first request would be an actual task, not an AM.
        Correct me if I'm wrong here.

        Show
        Wei Yan added a comment - Thanks, Sandy. One problem may exist if we use SchedulerApplicationAttempt.getLiveContainers().isEmpty(), if the application is unManagedAM, it will not generate an AM resource request. Thus, the first request would be an actual task, not an AM. Correct me if I'm wrong here.
        Hide
        Wei Yan added a comment -

        Update a patch. Use getLiveContainer().size() and unManagedAM to detect the AM container.

        Show
        Wei Yan added a comment - Update a patch. Use getLiveContainer().size() and unManagedAM to detect the AM container.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647968 ]
        Hide
        Sandy Ryza added a comment -

        The "M" un Unmanaged AM shouldn't be capitalized. Otherwise, the patch LGTM.

        Show
        Sandy Ryza added a comment - The "M" un Unmanaged AM shouldn't be capitalized. Otherwise, the patch LGTM.
        Hide
        Wei Yan added a comment -

        Thanks, Sandy. Fixed that problem.

        Show
        Wei Yan added a comment - Thanks, Sandy. Fixed that problem.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647969 ]
        Hide
        Hadoop QA added a comment -

        -1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647968/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        -1 core tests. The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager:

        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestMaxRunningAppsEnforcer
        org.apache.hadoop.yarn.server.resourcemanager.scheduler.TestSchedulerApplicationAttempt
        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFSSchedulerApp

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3890//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3890//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - -1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647968/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. -1 core tests . The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager: org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestMaxRunningAppsEnforcer org.apache.hadoop.yarn.server.resourcemanager.scheduler.TestSchedulerApplicationAttempt org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFSSchedulerApp +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3890//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3890//console This message is automatically generated.
        Hide
        Wei Yan added a comment -

        New patch to fix the test errors.

        Show
        Wei Yan added a comment - New patch to fix the test errors.
        Wei Yan made changes -
        Attachment YARN-1913.patch [ 12647976 ]
        Hide
        Hadoop QA added a comment -

        -1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647969/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        -1 core tests. The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager:

        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestMaxRunningAppsEnforcer
        org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFSSchedulerApp
        org.apache.hadoop.yarn.server.resourcemanager.scheduler.TestSchedulerApplicationAttempt

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3891//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3891//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - -1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647969/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. -1 core tests . The patch failed these unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager: org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestMaxRunningAppsEnforcer org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.TestFSSchedulerApp org.apache.hadoop.yarn.server.resourcemanager.scheduler.TestSchedulerApplicationAttempt +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3891//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3891//console This message is automatically generated.
        Wei Yan made changes -
        Status Patch Available [ 10002 ] Open [ 1 ]
        Wei Yan made changes -
        Status Open [ 1 ] Patch Available [ 10002 ]
        Hide
        Hadoop QA added a comment -

        +1 overall. Here are the results of testing the latest attachment
        http://issues.apache.org/jira/secure/attachment/12647976/YARN-1913.patch
        against trunk revision .

        +1 @author. The patch does not contain any @author tags.

        +1 tests included. The patch appears to include 3 new or modified test files.

        +1 javac. The applied patch does not increase the total number of javac compiler warnings.

        +1 javadoc. There were no new javadoc warning messages.

        +1 eclipse:eclipse. The patch built with eclipse:eclipse.

        +1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

        +1 release audit. The applied patch does not increase the total number of release audit warnings.

        +1 core tests. The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager.

        +1 contrib tests. The patch passed contrib unit tests.

        Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3892//testReport/
        Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3892//console

        This message is automatically generated.

        Show
        Hadoop QA added a comment - +1 overall . Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12647976/YARN-1913.patch against trunk revision . +1 @author . The patch does not contain any @author tags. +1 tests included . The patch appears to include 3 new or modified test files. +1 javac . The applied patch does not increase the total number of javac compiler warnings. +1 javadoc . There were no new javadoc warning messages. +1 eclipse:eclipse . The patch built with eclipse:eclipse. +1 findbugs . The patch does not introduce any new Findbugs (version 1.3.9) warnings. +1 release audit . The applied patch does not increase the total number of release audit warnings. +1 core tests . The patch passed unit tests in hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager. +1 contrib tests . The patch passed contrib unit tests. Test results: https://builds.apache.org/job/PreCommit-YARN-Build/3892//testReport/ Console output: https://builds.apache.org/job/PreCommit-YARN-Build/3892//console This message is automatically generated.
        Sandy Ryza made changes -
        Status Patch Available [ 10002 ] Resolved [ 5 ]
        Fix Version/s 2.5.0 [ 12326262 ]
        Resolution Fixed [ 1 ]
        Hide
        Hudson added a comment -

        SUCCESS: Integrated in Hadoop-trunk-Commit #5646 (See https://builds.apache.org/job/Hadoop-trunk-Commit/5646/)
        YARN-1913. With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400)

        • /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Show
        Hudson added a comment - SUCCESS: Integrated in Hadoop-trunk-Commit #5646 (See https://builds.apache.org/job/Hadoop-trunk-Commit/5646/ ) YARN-1913 . With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400 ) /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Hide
        Hudson added a comment -

        FAILURE: Integrated in Hadoop-Yarn-trunk #572 (See https://builds.apache.org/job/Hadoop-Yarn-trunk/572/)
        YARN-1913. With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400)

        • /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Show
        Hudson added a comment - FAILURE: Integrated in Hadoop-Yarn-trunk #572 (See https://builds.apache.org/job/Hadoop-Yarn-trunk/572/ ) YARN-1913 . With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400 ) /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Hide
        Hudson added a comment -

        SUCCESS: Integrated in Hadoop-Hdfs-trunk #1763 (See https://builds.apache.org/job/Hadoop-Hdfs-trunk/1763/)
        YARN-1913. With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400)

        • /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Show
        Hudson added a comment - SUCCESS: Integrated in Hadoop-Hdfs-trunk #1763 (See https://builds.apache.org/job/Hadoop-Hdfs-trunk/1763/ ) YARN-1913 . With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400 ) /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Hide
        Hudson added a comment -

        FAILURE: Integrated in Hadoop-Mapreduce-trunk #1790 (See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1790/)
        YARN-1913. With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400)

        • /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java
        • /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Show
        Hudson added a comment - FAILURE: Integrated in Hadoop-Mapreduce-trunk #1790 (See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1790/ ) YARN-1913 . With Fair Scheduler, cluster can logjam when all resources are consumed by AMs (Wei Yan via Sandy Ryza) (sandy: http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1599400 ) /hadoop/common/trunk/hadoop-yarn-project/CHANGES.txt /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/SchedulerApplicationAttempt.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationConfiguration.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/AppSchedulable.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FSLeafQueue.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/SchedulingPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FairSharePolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/FifoPolicy.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/FairSchedulerTestBase.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestAllocationFileLoaderService.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/TestFairScheduler.java /hadoop/common/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-site/src/site/apt/FairScheduler.apt.vm
        Karthik Kambatla (Inactive) made changes -
        Status Resolved [ 5 ] Closed [ 6 ]
        Transition Time In Source Status Execution Times Last Executer Last Execution Date
        Patch Available Patch Available Open Open
        11d 19h 53m 2 Wei Yan 02/Jun/14 21:55
        Open Open Patch Available Patch Available
        43d 8h 42m 3 Wei Yan 02/Jun/14 21:56
        Patch Available Patch Available Resolved Resolved
        4h 2m 1 Sandy Ryza 03/Jun/14 01:59
        Resolved Resolved Closed Closed
        73d 4h 45m 1 Karthik Kambatla (Inactive) 15/Aug/14 06:44

          People

          • Assignee:
            Wei Yan
            Reporter:
            bc Wong
          • Votes:
            0 Vote for this issue
            Watchers:
            20 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development