[
https://issues.apache.org/jira/browse/AMBARI-21898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16157143#comment-16157143
]
Jonathan Hurley commented on AMBARI-21898:
------------------------------------------
This is causing NPEs when loading the UI. Should we revert this until its fixed?
{code}
2017-09-07 11:59:27,940 ERROR [ambari-client-thread-37]
(ContainerResponse.java:419) mapMappableContainerException() - The
RuntimeException could not be mapped to a response, re-throwing to the HTTP
container
java.lang.NullPointerException
at
org.apache.ambari.server.controller.internal.AbstractProviderModule.initProviderMaps(AbstractProviderModule.java:875)
at
org.apache.ambari.server.controller.internal.AbstractProviderModule.checkInit(AbstractProviderModule.java:849)
at
org.apache.ambari.server.controller.internal.AbstractProviderModule.getMetricsServiceType(AbstractProviderModule.java:344)
at
org.apache.ambari.server.controller.metrics.MetricsReportPropertyProviderProxy.checkPropertyIds(MetricsReportPropertyProviderProxy.java:95)
at
org.apache.ambari.server.controller.internal.ClusterControllerImpl.providesRequestProperties(ClusterControllerImpl.java:556)
at
org.apache.ambari.server.controller.internal.ClusterControllerImpl.populateResources(ClusterControllerImpl.java:154)
{code}
> Property provider in-memory maps are refreshed too slowly after config updates
> ------------------------------------------------------------------------------
>
> Key: AMBARI-21898
> URL: https://issues.apache.org/jira/browse/AMBARI-21898
> Project: Ambari
> Issue Type: Bug
> Components: ambari-server
> Affects Versions: 2.0.0
> Reporter: Siddharth Wagle
> Assignee: Siddharth Wagle
> Priority: Critical
> Fix For: 2.6.0
>
> Attachments: AMBARI-21898.patch
>
>
> Building the hostComponentMap map for ProviderModule seems does unnecessary
> computation going through the ResourceProvider layer. On every configuration
> update, we call the reset on these maps which takes a long time to refresh.
> Instead, relying on
> org.apache.ambari.server.state.cluster.ClusterImpl#serviceComponentHosts will
> lead to faster computation without overhead like stale config calculations
> for every SCH.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
