You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-dev@hadoop.apache.org by "Sergey Shelukhin (JIRA)" <ji...@apache.org> on 2015/08/10 20:54:45 UTC
[jira] [Created] (YARN-4042) YARN registry should handle the
absence of ZK node
Sergey Shelukhin created YARN-4042:
--------------------------------------
Summary: YARN registry should handle the absence of ZK node
Key: YARN-4042
URL: https://issues.apache.org/jira/browse/YARN-4042
Project: Hadoop YARN
Issue Type: Bug
Reporter: Sergey Shelukhin
{noformat}
2015-08-10 11:33:46,931 WARN [LlapSchedulerNodeEnabler] rm.LlapTaskSchedulerService: Could not refresh list of active instances
org.apache.hadoop.fs.PathNotFoundException: `/registry/users/huzheng/services/org-apache-hive/llap0/components/workers/worker-0000000025': No such file or directory: KeeperErrorCode = NoNode for /registry/users/huzheng/services/org-apache-hive/llap0/components/workers/worker-0000000025
at org.apache.hadoop.registry.client.impl.zk.CuratorService.operationFailure(CuratorService.java:377)
at org.apache.hadoop.registry.client.impl.zk.CuratorService.operationFailure(CuratorService.java:360)
at org.apache.hadoop.registry.client.impl.zk.CuratorService.zkRead(CuratorService.java:720)
at org.apache.hadoop.registry.client.impl.zk.RegistryOperationsService.resolve(RegistryOperationsService.java:120)
at org.apache.hadoop.registry.client.binding.RegistryUtils.extractServiceRecords(RegistryUtils.java:321)
at org.apache.hadoop.registry.client.binding.RegistryUtils.listServiceRecords(RegistryUtils.java:177)
at org.apache.hadoop.hive.llap.daemon.registry.impl.LlapYarnRegistryImpl$DynamicServiceInstanceSet.refresh(LlapYarnRegistryImpl.java:278)
at org.apache.tez.dag.app.rm.LlapTaskSchedulerService.refreshInstances(LlapTaskSchedulerService.java:584)
at org.apache.tez.dag.app.rm.LlapTaskSchedulerService.access$900(LlapTaskSchedulerService.java:79)
at org.apache.tez.dag.app.rm.LlapTaskSchedulerService$NodeEnablerCallable.call(LlapTaskSchedulerService.java:887)
at org.apache.tez.dag.app.rm.LlapTaskSchedulerService$NodeEnablerCallable.call(LlapTaskSchedulerService.java:855)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: org.apache.zookeeper.KeeperException$NoNodeException: KeeperErrorCode = NoNode for /registry/users/huzheng/services/org-apache-hive/llap0/components/workers/worker-0000000025
at org.apache.zookeeper.KeeperException.create(KeeperException.java:111)
at org.apache.zookeeper.KeeperException.create(KeeperException.java:51)
at org.apache.zookeeper.ZooKeeper.getData(ZooKeeper.java:1155)
at org.apache.curator.framework.imps.GetDataBuilderImpl$4.call(GetDataBuilderImpl.java:302)
at org.apache.curator.framework.imps.GetDataBuilderImpl$4.call(GetDataBuilderImpl.java:291)
at org.apache.curator.RetryLoop.callWithRetry(RetryLoop.java:107)
at org.apache.curator.framework.imps.GetDataBuilderImpl.pathInForeground(GetDataBuilderImpl.java:288)
at org.apache.curator.framework.imps.GetDataBuilderImpl.forPath(GetDataBuilderImpl.java:279)
at org.apache.curator.framework.imps.GetDataBuilderImpl.forPath(GetDataBuilderImpl.java:41)
at org.apache.hadoop.registry.client.impl.zk.CuratorService.zkRead(CuratorService.java:718)
... 12 more
{noformat}
ZK nodes can disappear after listing, for example ephemeral node can be cleaned up. YARN registry should handle that.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)