[YARN-8465] Dshell docker container gets marked as lost after NM restart

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.2.0, 3.1.1
• Fix Version/s: 3.2.0, 3.1.1
• Component/s: yarn-native-services
• Labels: Docker
• Target Version/s: 3.2.0, 3.1.1
• Assignee: Shane Kumpf
• Reporter: Yesha Vora
• Created: 26/Jun/18 23:18
• Updated: 02/Jul/18 18:04
• Resolved: 02/Jul/18 17:41

Description

scenario:
1) launch dshell application

yarn  jar /usr/hdp/current/hadoop-yarn-client/hadoop-yarn-applications-distributedshell.jar  -shell_command "sleep 500" -num_containers 2 -shell_env YARN_CONTAINER_RUNTIME_TYPE=docker -shell_env YARN_CONTAINER_RUNTIME_DOCKER_IMAGE=xx/httpd:0.1 -jar /usr/hdp/current/hadoop-yarn-client/hadoop-yarn-applications-distributedshell.jar

2) wait for app to be in stable state ( container_e01_1529968198450_0001_01_000002 is running on host7 and container_e01_1529968198450_0001_01_000003 is running on host5)
3) restart NM (host7)

Here, dshell application fails with below error

18/06/25 23:35:30 INFO distributedshell.Client: Got application report from ASM for, appId=1, clientToAMToken=Token { kind: YARN_CLIENT_TOKEN, service:  }, appDiagnostics=, appMasterHost=host9/xxx, appQueue=default, appMasterRpcPort=-1, appStartTime=1529969211776, yarnAppState=RUNNING, distributedFinalState=UNDEFINED, appTrackingUrl=https://host4:8090/proxy/application_1529968198450_0001/, appUser=hbase
18/06/25 23:35:31 INFO distributedshell.Client: Got application report from ASM for, appId=1, clientToAMToken=null, appDiagnostics=Application Failure: desired = 2, completed = 2, allocated = 2, failed = 1, diagnostics = [2018-06-25 23:35:28.000]Container exited with a non-zero exit code 154
[2018-06-25 23:35:28.001]Container exited with a non-zero exit code 154
, appMasterHost=host9/xxx, appQueue=default, appMasterRpcPort=-1, appStartTime=1529969211776, yarnAppState=FINISHED, distributedFinalState=FAILED, appTrackingUrl=https://host4:8090/proxy/application_1529968198450_0001/, appUser=hbase
18/06/25 23:35:31 INFO distributedshell.Client: Application did finished unsuccessfully. YarnState=FINISHED, DSFinalStatus=FAILED. Breaking monitoring loop
18/06/25 23:35:31 ERROR distributedshell.Client: Application failed to complete successfully

Here, the docker container marked as LOST after completion

2018-06-25 23:35:27,970 WARN  runtime.DockerLinuxContainerRuntime (DockerLinuxContainerRuntime.java:signalContainer(1034)) - Signal docker container failed. Exception:
org.apache.hadoop.yarn.server.nodemanager.containermanager.runtime.ContainerExecutionException: Liveliness check failed for PID: 423695. Container may have already completed.
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.DockerLinuxContainerRuntime.executeLivelinessCheck(DockerLinuxContainerRuntime.java:1208)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.DockerLinuxContainerRuntime.signalContainer(DockerLinuxContainerRuntime.java:1026)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.DelegatingLinuxContainerRuntime.signalContainer(DelegatingLinuxContainerRuntime.java:159)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.signalContainer(LinuxContainerExecutor.java:755)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.isContainerAlive(LinuxContainerExecutor.java:905)
        at org.apache.hadoop.yarn.server.nodemanager.ContainerExecutor.reacquireContainer(ContainerExecutor.java:284)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.reacquireContainer(LinuxContainerExecutor.java:721)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.RecoveredContainerLaunch.call(RecoveredContainerLaunch.java:84)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.RecoveredContainerLaunch.call(RecoveredContainerLaunch.java:47)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
        at java.lang.Thread.run(Thread.java:748)
2018-06-25 23:35:27,975 WARN  nodemanager.LinuxContainerExecutor (LinuxContainerExecutor.java:signalContainer(762)) - Error in signalling container 423695 with NULL; exit = -1
org.apache.hadoop.yarn.server.nodemanager.containermanager.runtime.ContainerExecutionException: Signal docker container failed
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.DockerLinuxContainerRuntime.signalContainer(DockerLinuxContainerRuntime.java:1036)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.DelegatingLinuxContainerRuntime.signalContainer(DelegatingLinuxContainerRuntime.java:159)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.signalContainer(LinuxContainerExecutor.java:755)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.isContainerAlive(LinuxContainerExecutor.java:905)
        at org.apache.hadoop.yarn.server.nodemanager.ContainerExecutor.reacquireContainer(ContainerExecutor.java:284)
        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.reacquireContainer(LinuxContainerExecutor.java:721)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.RecoveredContainerLaunch.call(RecoveredContainerLaunch.java:84)
        at org.apache.hadoop.yarn.server.nodemanager.containermanager.launcher.RecoveredContainerLaunch.call(RecoveredContainerLaunch.java:47)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
        at java.lang.Thread.run(Thread.java:748)

Issue Links

• is caused by: YARN-8259 Revisit liveliness checks for Docker containers