[ https://issues.apache.org/jira/browse/YARN-11649?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17809878#comment-17809878 ]
ASF GitHub Bot commented on YARN-11649: --------------------------------------- jchanggg commented on code in PR #6481: URL: https://github.com/apache/hadoop/pull/6481#discussion_r1463069638 ########## hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/conf/YarnConfiguration.java: ########## @@ -4081,6 +4081,24 @@ public static boolean isAclEnabled(Configuration conf) { public static final long DEFAULT_FEDERATION_AMRMPROXY_SUBCLUSTER_TIMEOUT = 60000; // one minute + public static final String FEDERATION_OVERRIDE_MAX_CLUSTER_CAPABILITY = + FEDERATION_PREFIX + "override.maximum.cluster.capability.enabled"; Review Comment: Revised and added. > YARN Federation getNewApplication returns different maxresourcecapability > ------------------------------------------------------------------------- > > Key: YARN-11649 > URL: https://issues.apache.org/jira/browse/YARN-11649 > Project: Hadoop YARN > Issue Type: Bug > Reporter: Jeffrey Chang > Assignee: Jeffrey Chang > Priority: Minor > Labels: pull-request-available > > When getNewApplication is called against YARN Router with Federation on, its > possible we get different maxResourceCapabilities in its response on > different calls. This is because getNewApplication is called against a random > cluster on each call, which may return different maxResourceCapability based > on the cluster that the call is executed on. ie in response in HTTP > [https://hadoop.apache.org/docs/stable/hadoop-yarn/hadoop-yarn-site/ResourceManagerRest.html#Cluster_New_Application_API] > -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org