[HDFS-17250] EditLogTailer#triggerActiveLogRoll should handle thread Interrupted

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.4.0
• Fix Version/s: 3.4.0
• Component/s: hdfs
• Labels: pull-request-available
• Target Version/s: 3.4.0
• Hadoop Flags: Reviewed
• Assignee: Haiyang Hu
• Reporter: Haiyang Hu
• Created: 09/Nov/23 11:46
• Updated: 28/Jan/24 01:13
• Resolved: 04/Dec/23 13:20

Description

Issue:
When the NameNode attempts to trigger a log roll and the cachedActiveProxy is a "the machine has been shut down of the namenode," it is unable to establish a network connection. This results in a timeout during the socket connection phase, which has a set timeout of 90 seconds. Since the asynchronous call for "Triggering log roll" has a waiting time of 60 seconds, it triggers a timeout and initiates a "cancel" operation, causing the executing thread to receive an "Interrupted" signal and throwing a "java.io.InterruptedIOException" exception.

Currently, the logic not to handle interrupted signal, and the "getActiveNodeProxy" method hasn't reached the maximum retry limit, the overall execution process doesn't exit and it continues to attempt to
call the "rollEditLog" on the next NameNode in the list. However when a socket connection is established, it throws a "java.nio.channels.ClosedByInterruptException" exception due to the thread being in an "Interrupted" state.
this cycle repeats until it reaches the maximum retry limit (nnCount * maxRetries) will exits.

However in the next cycle of "Triggering log roll," it continues to traverse the NameNode list and encounters the same issue and the cachedActiveProxy is still a "shut down NameNode."
This eventually results in the NameNode being unable to successfully complete the "Triggering log roll" operation.

To optimize this, we need to handle the thread being interrupted and exit the execution.

Detailed logs such as:

the Observer node "ob1" will execute "Triggering log roll" is as follows:
nns list is [ob2(shut down machine),nn1(active),nn2(standy)]

• The Observer node "ob1" periodically executes "triggerActiveLogRoll" and asynchronously calls "getNameNodeProxy" to request "ob2" for the "rollEditLog" operation, since the "ob2" machine is shut down, it cannot establish a network connection, this results in a timeout during the socket connection phase (here set timeout is 90 seconds).

  2023-11-03 10:27:41,734 INFO ha.EditLogTailer (EditLogTailer.java:triggerActiveLogRoll(465)) [Edit log tailer] - Triggering log roll on remote NameNode

  2023-11-03 10:28:41,734 WARN ha.EditLogTailer (EditLogTailer.java:triggerActiveLogRoll(478)) [Edit log tailer] - Unable to finish rolling edits in 60000 ms

• As the asynchronous call for "Triggering log roll" has a waiting time of 60 seconds, it triggers a timeout and initiates a "cancel" operation, causing the executing thread to receive an "Interrupted" signal and will throw "java.io.InterruptedIOException".

  2023-11-03 10:28:41,753 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
  2023-11-03 10:28:41,972 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=ob2, ipcAddress=xxx, httpAddress=http://xxx], try next.
  java.io.InterruptedIOException: DestHost:destPort xxx , LocalHost:localPort xxx. Failed on local exception: java.io.InterruptedIOException: Interrupted while waiting for IO on channel java.nio.channels.SocketChannel[connection-pending remote=xxx:8040]. Total timeout mills is 90000, 30038 millis timeout left.
  at sun.reflect.GeneratedConstructorAccessor39.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:931)
  at org.apache.hadoop.net.NetUtils.wrapException(NetUtils.java:906)
  at org.apache.hadoop.ipc.Client.getRpcResponse(Client.java:1583)
  at org.apache.hadoop.ipc.Client.call(Client.java:1511)
  at org.apache.hadoop.ipc.Client.call(Client.java:1402)
  at org.apache.hadoop.ipc.ProtobufRpcEngine2$Invoker.invoke(ProtobufRpcEngine2.java:261)
  at org.apache.hadoop.ipc.ProtobufRpcEngine2$Invoker.invoke(ProtobufRpcEngine2.java:141)
  at com.sun.proxy.$Proxy21.rollEditLog(Unknown Source)
  at org.apache.hadoop.hdfs.protocolPB.NamenodeProtocolTranslatorPB.rollEditLog(NamenodeProtocolTranslatorPB.java:152)
  at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$2.doWork(EditLogTailer.java:454)
  at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$2.doWork(EditLogTailer.java:451)
  at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$MultipleNameNodeProxy.call(EditLogTailer.java:615)
  at java.util.concurrent.FutureTask.run(FutureTask.java:266)
  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
  at java.lang.Thread.run(Thread.java:748)
  Caused by: java.io.InterruptedIOException: Interrupted while waiting for IO on channel java.nio.channels.SocketChannel[connection-pending remote=xxx:8040]. Total timeout mills is 90000, 30038 millis timeout left.
  at org.apache.hadoop.net.SocketIOWithTimeout$SelectorPool.select(SocketIOWithTimeout.java:351)
  at org.apache.hadoop.net.SocketIOWithTimeout.connect(SocketIOWithTimeout.java:203)
  at org.apache.hadoop.net.NetUtils.connect(NetUtils.java:604)
  at org.apache.hadoop.ipc.Client$Connection.setupConnection(Client.java:662)
  at org.apache.hadoop.ipc.Client$Connection.setupIOstreams(Client.java:783)
  at org.apache.hadoop.ipc.Client$Connection.access$3800(Client.java:346)
  at org.apache.hadoop.ipc.Client.getConnection(Client.java:1653)

• Since current logic not to handle interrupted signal and hasn't reached the maximum retry limit, the overall execution process doesn't exit and it continues to attempt to call the "rollEditLog" on the next NameNode in the list. However when a socket connection is established, it throws a "java.nio.channels.ClosedByInterruptException" exception due to the thread being in an "Interrupted" state. this cycle repeats until it reaches 9 retry limit (3*3) and then exits.

2023-11-03 10:28:41,991 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:41,991 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn1, ipcAddress=xxx:8040, httpAddress=http://xxx:50070], try next.
java.nio.channels.ClosedByInterruptException
at java.nio.channels.spi.AbstractInterruptibleChannel.end(AbstractInterruptibleChannel.java:202)
at sun.nio.ch.SocketChannelImpl.connect(SocketChannelImpl.java:656)
at org.apache.hadoop.net.SocketIOWithTimeout.connect(SocketIOWithTimeout.java:192)
at org.apache.hadoop.net.NetUtils.connect(NetUtils.java:604)
at org.apache.hadoop.ipc.Client$Connection.setupConnection(Client.java:662)
at org.apache.hadoop.ipc.Client$Connection.setupIOstreams(Client.java:783)
at org.apache.hadoop.ipc.Client$Connection.access$3800(Client.java:346)
at org.apache.hadoop.ipc.Client.getConnection(Client.java:1653)
at org.apache.hadoop.ipc.Client.call(Client.java:1449)
at org.apache.hadoop.ipc.Client.call(Client.java:1402)
at org.apache.hadoop.ipc.ProtobufRpcEngine2$Invoker.invoke(ProtobufRpcEngine2.java:261)
at org.apache.hadoop.ipc.ProtobufRpcEngine2$Invoker.invoke(ProtobufRpcEngine2.java:141)
at com.sun.proxy.$Proxy21.rollEditLog(Unknown Source)
at org.apache.hadoop.hdfs.protocolPB.NamenodeProtocolTranslatorPB.rollEditLog(NamenodeProtocolTranslatorPB.java:152)
at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$2.doWork(EditLogTailer.java:454)
at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$2.doWork(EditLogTailer.java:451)
at org.apache.hadoop.hdfs.server.namenode.ha.EditLogTailer$MultipleNameNodeProxy.call(EditLogTailer.java:615)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

2023-11-03 10:28:42,095 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,095 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn2, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

2023-11-03 10:28:42,104 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,104 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=ob2, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

2023-11-03 10:28:42,118 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,118 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn1, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.

2023-11-03 10:28:42,119 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,119 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn2, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

2023-11-03 10:28:42,148 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,148 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=ob2, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

2023-11-03 10:28:42,148 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,148 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn1, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

2023-11-03 10:28:42,149 WARN ipc.Client (Client.java:handleConnectionFailure(930)) [pool-33-thread-1] - Interrupted while trying for connection
2023-11-03 10:28:42,149 WARN ha.EditLogTailer (EditLogTailer.java:call(618)) [pool-33-thread-1] - Exception from remote name node RemoteNameNodeInfo [nnId=nn2, ipcAddress=xxx:8040, httpAddress=xxx:50070], try next.
java.nio.channels.ClosedByInterruptException

• In the next cycle of "Triggering log roll," it continues to traverse the NameNode list and proceeds to "ob2," causing the same issue to recur. This eventually results in the NameNode being unable to successfully complete the "Triggering log roll" operation.

Issue Links

• links to: GitHub Pull Request #6266