[YARN-4288] NodeManager restart should keep retrying to register to RM while connection exception happens during RM failed over.

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 2.6.0
• Fix Version/s: 2.8.0, 2.7.3, 3.0.0-alpha1
• Component/s: nodemanager
• Target Version/s: 2.8.0
• Hadoop Flags: Reviewed
• Assignee: Junping Du
• Reporter: Junping Du
• Created: 22/Oct/15 12:30
• Updated: 25/Oct/19 20:26
• Resolved: 29/Oct/15 07:05

Description

When NM get restarted, NodeStatusUpdaterImpl will try to register to RM with RPC which could throw following exceptions when RM get restarted at the same time, like following exception shows:

2015-08-17 14:35:59,434 ERROR nodemanager.NodeStatusUpdaterImpl (NodeStatusUpdaterImpl.java:rebootNodeStatusUpdaterAndRegisterWithRM(222)) - Unexpected error rebooting NodeStatusUpdater
java.io.IOException: Failed on local exception: java.io.IOException: Connection reset by peer; Host Details : local host is: "172.27.62.28"; destination host is: "172.27.62.57":8025;
        at org.apache.hadoop.net.NetUtils.wrapException(NetUtils.java:772)
        at org.apache.hadoop.ipc.Client.call(Client.java:1473)
        at org.apache.hadoop.ipc.Client.call(Client.java:1400)
        at org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:232)
        at com.sun.proxy.$Proxy36.registerNodeManager(Unknown Source)
        at org.apache.hadoop.yarn.server.api.impl.pb.client.ResourceTrackerPBClientImpl.registerNodeManager(ResourceTrackerPBClientImpl.java:68)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
        at java.lang.reflect.Method.invoke(Method.java:606)
        at org.apache.hadoop.io.retry.RetryInvocationHandler.invokeMethod(RetryInvocationHandler.java:187)
        at org.apache.hadoop.io.retry.RetryInvocationHandler.invoke(RetryInvocationHandler.java:102)
        at com.sun.proxy.$Proxy37.registerNodeManager(Unknown Source)
        at org.apache.hadoop.yarn.server.nodemanager.NodeStatusUpdaterImpl.registerWithRM(NodeStatusUpdaterImpl.java:257)
        at org.apache.hadoop.yarn.server.nodemanager.NodeStatusUpdaterImpl.rebootNodeStatusUpdaterAndRegisterWithRM(NodeStatusUpdaterImpl.java:215)
        at org.apache.hadoop.yarn.server.nodemanager.NodeManager$2.run(NodeManager.java:304)
Caused by: java.io.IOException: Connection reset by peer
        at sun.nio.ch.FileDispatcherImpl.read0(Native Method)
        at sun.nio.ch.SocketDispatcher.read(SocketDispatcher.java:39)
        at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223)
        at sun.nio.ch.IOUtil.read(IOUtil.java:197)
        at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:379)
        at org.apache.hadoop.net.SocketInputStream$Reader.performIO(SocketInputStream.java:57)
        at org.apache.hadoop.net.SocketIOWithTimeout.doIO(SocketIOWithTimeout.java:142)
        at org.apache.hadoop.net.SocketInputStream.read(SocketInputStream.java:161)
        at org.apache.hadoop.net.SocketInputStream.read(SocketInputStream.java:131)
        at java.io.FilterInputStream.read(FilterInputStream.java:133)
        at java.io.FilterInputStream.read(FilterInputStream.java:133)
        at org.apache.hadoop.ipc.Client$Connection$PingInputStream.read(Client.java:514)
        at java.io.BufferedInputStream.fill(BufferedInputStream.java:235)
        at java.io.BufferedInputStream.read(BufferedInputStream.java:254)
        at java.io.DataInputStream.readInt(DataInputStream.java:387)
        at org.apache.hadoop.ipc.Client$Connection.receiveRpcResponse(Client.java:1072)
        at org.apache.hadoop.ipc.Client$Connection.run(Client.java:967)
2015-08-17 14:35:59,436 FATAL nodemanager.NodeManager (NodeManager.java:run(307)) - Error while rebooting NodeStatusUpdater.
org.apache.hadoop.yarn.exceptions.YarnRuntimeException: java.io.IOException: Failed on local exception: java.io.IOException: Connection reset by peer; Host Details : local host is: "172.27.62.28"; destination host is: "172.27.62.57":8025;
        at org.apache.hadoop.yarn.server.nodemanager.NodeStatusUpdaterImpl.rebootNodeStatusUpdaterAndRegisterWithRM(NodeStatusUpdaterImpl.java:223)
        at org.apache.hadoop.yarn.server.nodemanager.NodeManager$2.run(NodeManager.java:304)
Caused by: java.io.IOException: Failed on local exception: java.io.IOException: Connection reset by peer; Host Details : local host is: "ebdp-ch2-172.27.62.28"; destination host is: "172.27.62.57":8025;
        at org.apache.hadoop.net.NetUtils.wrapException(NetUtils.java:772)
        at org.apache.hadoop.ipc.Client.call(Client.java:1473)
        at org.apache.hadoop.ipc.Client.call(Client.java:1400)
        at org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:232)
        at com.sun.proxy.$Proxy36.registerNodeManager(Unknown Source)
        at org.apache.hadoop.yarn.server.api.impl.pb.client.ResourceTrackerPBClientImpl.registerNodeManager(ResourceTrackerPBClientImpl.java:68)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
        at java.lang.reflect.Method.invoke(Method.java:606)
        at org.apache.hadoop.io.retry.RetryInvocationHandler.invokeMethod(RetryInvocationHandler.java:187)
        at org.apache.hadoop.io.retry.RetryInvocationHandler.invoke(RetryInvocationHandler.java:102)
        at com.sun.proxy.$Proxy37.registerNodeManager(Unknown Source)
        at org.apache.hadoop.yarn.server.nodemanager.NodeStatusUpdaterImpl.registerWithRM(NodeStatusUpdaterImpl.java:257)
        at org.apache.hadoop.yarn.server.nodemanager.NodeStatusUpdaterImpl.rebootNodeStatusUpdaterAndRegisterWithRM(NodeStatusUpdaterImpl.java:215)
        ... 1 more
Caused by: java.io.IOException: Connection reset by peer
        at sun.nio.ch.FileDispatcherImpl.read0(Native Method)
        at sun.nio.ch.SocketDispatcher.read(SocketDispatcher.java:39)
        at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223)
        at sun.nio.ch.IOUtil.read(IOUtil.java:197)
        at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:379)
        at org.apache.hadoop.net.SocketInputStream$Reader.performIO(SocketInputStream.java:57)
        at org.apache.hadoop.net.SocketIOWithTimeout.doIO(SocketIOWithTimeout.java:142)
        at org.apache.hadoop.net.SocketInputStream.read(SocketInputStream.java:161)
        at org.apache.hadoop.net.SocketInputStream.read(SocketInputStream.java:131)
        at java.io.FilterInputStream.read(FilterInputStream.java:133)
        at java.io.FilterInputStream.read(FilterInputStream.java:133)
        at org.apache.hadoop.ipc.Client$Connection$PingInputStream.read(Client.java:514)
        at java.io.BufferedInputStream.fill(BufferedInputStream.java:235)
        at java.io.BufferedInputStream.read(BufferedInputStream.java:254)
        at java.io.DataInputStream.readInt(DataInputStream.java:387)
        at org.apache.hadoop.ipc.Client$Connection.receiveRpcResponse(Client.java:1072)
        at org.apache.hadoop.ipc.Client$Connection.run(Client.java:967)
2015-08-17 14:35:59,445 INFO  mortbay.log (Slf4jLog.java:info(67)) - Stopped HttpServer2$SelectChannelConnectorWithSafeStartup@0.0.0.0:8042
2015-08-17 14:35:59,547 INFO  containermanager.ContainerManagerImpl (ContainerManagerImpl.java:cleanUpApplicationsOnNMShutDown(512)) - Applications still running : [application_1439417357296_45357, application_1439417357296_45403, application_1439417357296_45355, application_1439417357296_45111, application_1439417357296_45452, application_1439417357296_45350, application_1439417357296_45499, application_1439417357296_45205, application_1439417357296_21009]
2015-08-17 14:35:59,548 INFO  ipc.Server (Server.java:stop(2469)) - Stopping server on 45454
2015-08-17 14:35:59,551 INFO  ipc.Server (Server.java:run(717)) - Stopping IPC Server listener on 45454
2015-08-17 14:35:59,551 INFO  logaggregation.LogAggregationService (LogAggregationService.java:serviceStop(141)) - org.apache.hadoop.yarn.server.nodemanager.containermanager.logaggregation.LogAggregationService waiting for pending aggregation during exit
2015-08-17 14:35:59,552 INFO  ipc.Server (Server.java:run(843)) - Stopping IPC Server Responder

It will make NM restart get failed. We should have a simple fix to allow this register to RM can retry with connection failures.

Issue Links

• relates to: YARN-4274 NodeStatusUpdaterImpl should register to RM again after a non-fatal exception happen before