[YARN-11148] In federation and security mode, nm recover may fail.

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.2.1
• Fix Version/s: 3.4.0
• Component/s: nodemanager
• Labels: pull-request-available
• Target Version/s: 3.4.0
• Hadoop Flags: Reviewed
• Assignee: Chenyu Zheng
• Reporter: Chenyu Zheng
• Due: 20/May/22
• Created: 13/May/22 09:06
• Updated: 16/Sep/23 02:12
• Resolved: 13/Aug/22 17:33

Description

In federation yarn cluster, security is enable, nm recovery is enable, nm restart may fail.

Exception stack:

2022-05-08 00:44:11,536 WARN org.apache.hadoop.ipc.Client: Exception encountered while connecting to the server : org.apache.hadoop.security.AccessControlException: Client cannot authenticate via:[TOKEN, KERBEROS]
2022-05-08 00:44:11,540 ERROR org.apache.hadoop.yarn.server.nodemanager.amrmproxy.AMRMProxyService: Exception when recovering appattempt_1650635484875_0036_000002, removing it from NMStateStore and move on
org.apache.hadoop.yarn.exceptions.YarnRuntimeException: java.io.IOException: DestHost:destPort host:8032 , LocalHost:localPort node/10.x.x.x:0. Failed on local exception: java.io.IOException: org.apache.hadoop.security.AccessControlException: Client cannot authenticate via:[TOKEN, KERBEROS]
    at org.apache.hadoop.yarn.server.nodemanager.amrmproxy.FederationInterceptor.recover(FederationInterceptor.java:441)
    at org.apache.hadoop.yarn.server.nodemanager.amrmproxy.AMRMProxyService.initializePipeline(AMRMProxyService.java:466)
    at org.apache.hadoop.yarn.server.nodemanager.amrmproxy.AMRMProxyService.recover(AMRMProxyService.java:270)
    at org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl.recover(ContainerManagerImpl.java:389)
    at org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl.serviceInit(ContainerManagerImpl.java:324)
    at org.apache.hadoop.service.AbstractService.init(AbstractService.java:164)
    at org.apache.hadoop.service.CompositeService.serviceInit(CompositeService.java:108)
    at org.apache.hadoop.yarn.server.nodemanager.NodeManager.serviceInit(NodeManager.java:516)
    at org.apache.hadoop.service.AbstractService.init(AbstractService.java:164)
    at org.apache.hadoop.yarn.server.nodemanager.NodeManager.initAndStartNodeManager(NodeManager.java:974)
    at org.apache.hadoop.yarn.server.nodemanager.NodeManager.main(NodeManager.java:1054)
Caused by: java.io.IOException: DestHost:destPort host:8032 , LocalHost:localPort host/10.x.x.x:0. Failed on local exception: java.io.IOException: org.apache.hadoop.security.AccessControlException: Client cannot authenticate via:[TOKEN, KERBEROS]
    at sun.reflect.GeneratedConstructorAccessor30.newInstance(Unknown Source)
    at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
    at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
    at org.apache.hadoop.net.NetUtils.wrapWithMessage(NetUtils.java:833)
    at org.apache.hadoop.net.NetUtils.wrapException(NetUtils.java:808)
    at org.apache.hadoop.ipc.Client.getRpcResponse(Client.java:1558)
    at org.apache.hadoop.ipc.Client.call(Client.java:1492)
    at org.apache.hadoop.ipc.Client.call(Client.java:1389)
    at org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:233)
    at org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:118)
    at com.sun.proxy.$Proxy30.getContainers(Unknown Source)
    at org.apache.hadoop.yarn.api.impl.pb.client.ApplicationClientProtocolPBClientImpl.getContainers(ApplicationClientProtocolPBClientImpl.java:479)
    at sun.reflect.GeneratedMethodAccessor4.invoke(Unknown Source)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.apache.hadoop.io.retry.RetryInvocationHandler.invokeMethod(RetryInvocationHandler.java:422)
    at org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invokeMethod(RetryInvocationHandler.java:165)
    at org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invoke(RetryInvocationHandler.java:157)
    at org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invokeOnce(RetryInvocationHandler.java:95)
    at org.apache.hadoop.io.retry.RetryInvocationHandler.invoke(RetryInvocationHandler.java:359)
    at com.sun.proxy.$Proxy31.getContainers(Unknown Source)
    at org.apache.hadoop.yarn.server.nodemanager.amrmproxy.FederationInterceptor.recover(FederationInterceptor.java:418)
    ... 10 more
Caused by: java.io.IOException: org.apache.hadoop.security.AccessControlException: Client cannot authenticate via:[TOKEN, KERBEROS]
    at org.apache.hadoop.ipc.Client$Connection$1.run(Client.java:771)
    at java.security.AccessController.doPrivileged(Native Method)
    at javax.security.auth.Subject.doAs(Subject.java:422)
    at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1730)
    at org.apache.hadoop.ipc.Client$Connection.handleSaslConnectionFailure(Client.java:734)
    at org.apache.hadoop.ipc.Client$Connection.setupIOstreams(Client.java:828)
    at org.apache.hadoop.ipc.Client$Connection.access$3900(Client.java:422)
    at org.apache.hadoop.ipc.Client.getConnection(Client.java:1615)
    at org.apache.hadoop.ipc.Client.call(Client.java:1436)
    ... 25 more
Caused by: org.apache.hadoop.security.AccessControlException: Client cannot authenticate via:[TOKEN, KERBEROS]
    at org.apache.hadoop.security.SaslRpcClient.selectSaslClient(SaslRpcClient.java:173)
    at org.apache.hadoop.security.SaslRpcClient.saslConnect(SaslRpcClient.java:390)
    at org.apache.hadoop.ipc.Client$Connection.setupSaslConnection(Client.java:628)
    at org.apache.hadoop.ipc.Client$Connection.access$2300(Client.java:422)
    at org.apache.hadoop.ipc.Client$Connection$2.run(Client.java:815)
    at org.apache.hadoop.ipc.Client$Connection$2.run(Client.java:811)
    at java.security.AccessController.doPrivileged(Native Method)
    at javax.security.auth.Subject.doAs(Subject.java:422)
    at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1730)
    at org.apache.hadoop.ipc.Client$Connection.setupIOstreams(Client.java:811)
    ... 28 more
 

Obvious error, this ugi only support simple mode.

Issue Links

• causes: YARN-11254 hadoop-minikdc dependency duplicated in hadoop-yarn-server-nodemanager
• links to: GitHub Pull Request #4308