[jira] [Commented] (AMBARI-7023) Incorrect ATS metric request for non-HDP stack with version 2.1

["Alexander Denissov (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/AMBARI-7023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14118726#comment-14118726 ] 

Alexander Denissov commented on AMBARI-7023:
--------------------------------------------

We will likely use PHD stack with version greater than 2.1, and we do not yet have Ambari trunk compatible stack (we used 1.6.1 branch for now), so I can't verify the fix at this time, please apply the change and we will report if we encounter anything similar again.

> Incorrect ATS metric request for non-HDP stack with version 2.1
> ---------------------------------------------------------------
>
>                 Key: AMBARI-7023
>                 URL: https://issues.apache.org/jira/browse/AMBARI-7023
>             Project: Ambari
>          Issue Type: Bug
>    Affects Versions: 1.6.1
>            Reporter: Alexander Denissov
>            Assignee: Jaimin D Jetly
>            Priority: Critical
>             Fix For: 1.7.0
>
>
> Define a non-HDP stack based on hadoop 2.2, such as PHD 2.1.0 with HDFS+YARN+ZK services.
> After cluster deployment when a user presses "Complete" and UI tries to navigate to dashboard, a "Server Error" popup comes up with the message and the UI is stuck on loading bar of http://c6401.ambari.apache.org:8080/#/main/dashboard/metrics:
> The popup has the following error message:
> 500 status code received on GET method for API: /api/v1/clusters/test/components/?ServiceComponentInfo/component_name=APP_TIMELINE_SERVER|ServiceComponentInfo/component_name=JOURNALNODE|ServiceComponentInfo/category=MASTER&fields=ServiceComponentInfo/Version,ServiceComponentInfo/StartTime,ServiceComponentInfo/HeapMemoryUsed,ServiceComponentInfo/HeapMemoryMax,ServiceComponentInfo/service_name,host_components/HostRoles/host_name,host_components/HostRoles/state,host_components/HostRoles/maintenance_state,host_components/HostRoles/stale_configs,host_components/metrics/jvm/memHeapUsedM,host_components/metrics/jvm/HeapMemoryMax,host_components/metrics/jvm/HeapMemoryUsed,host_components/metrics/jvm/memHeapCommittedM,host_components/metrics/mapred/jobtracker/trackers_decommissioned,host_components/metrics/cpu/cpu_wio,host_components/metrics/rpc/RpcQueueTime_avg_time,host_components/metrics/dfs/FSNamesystem/*,host_components/metrics/dfs/namenode/Version,host_components/metrics/dfs/namenode/DecomNodes,host_components/metrics/dfs/namenode/TotalFiles,host_components/metrics/dfs/namenode/UpgradeFinalized,host_components/metrics/dfs/namenode/Safemode,host_components/metrics/runtime/StartTime,host_components/metrics/yarn/Queue,ServiceComponentInfo/rm_metrics/cluster/activeNMcount,ServiceComponentInfo/rm_metrics/cluster/unhealthyNMcount,ServiceComponentInfo/rm_metrics/cluster/rebootedNMcount,ServiceComponentInfo/rm_metrics/cluster/decommissionedNMcount&minimal_response=true 
> Error message: org.apache.ambari.server.controller.spi.SystemException: An internal system exception occurred: Could not find service for component, componentName=APP_TIMELINE_SERVER, clusterName=test, stackInfo=PHD-2.1.0 
> The problem, I believe is in ambari-web/app/controllers/global/update_controller.js lines:
> isATSInstalled = App.cache['services'].mapProperty('ServiceInfo.service_name').contains('YARN') && App.get('isHadoop21Stack'),
>       flumeHandlerParam = isFlumeInstalled ? 'ServiceComponentInfo/component_name=FLUME_HANDLER|' : '',
>       atsHandlerParam = isATSInstalled ? 'ServiceComponentInfo/component_name=APP_TIMELINE_SERVER|' : '',
> and ambari-web/app/app.js lines:
> isHadoop21Stack: function () {
>     return (stringUtils.compareVersions(this.get('currentStackVersionNumber'), "2.1") === 1 ||
>       stringUtils.compareVersions(this.get('currentStackVersionNumber'), "2.1") === 0)
>   }.property('currentStackVersionNumber'),
> Since the stack version number is 2.1 and YARN is installed, the UI assumes the stack is Hadoop 2.4 compatible (as is the case with HDP), which is not the case with non-HDP stacks.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)