Hi,
   I encounter a problem with Yarn's fair scheduler. The thing is that,  I 
first set a queue by configuring fair-scheduler.xml below. Next I try to submit 
a job to that queue by designating queue name via "mapreduce.job.queuename= 
amelie".
fair-scheduler.xml:
<allocations>
  <queue name="amelie">
    <minResources>10000 mb,1vcores</minResources>
    <maxResources>90000 mb,1vcores</maxResources>
    <maxRunningApps>50</maxRunningApps>
    <weight>2.0</weight>
    <schedulingPolicy>fair</schedulingPolicy>
  </queue>
</allocations>
However, the job is stuck in that queue without running, and the 
yarn-amelie-resourcemanager-pdcbl3-log file shows  errors that,
2013-12-06 10:16:20,551 ERROR org.apache.hadoop.security.UserGroupInformation: 
PriviledgedActionException as:amelie (auth:SIMPLE) 
cause:org.apache.hadoop.yarn.exceptions.ApplicationNotFoundException: 
Application with id 'application_1386295642784_0001' doesn't exist in RM.
2013-12-06 10:16:20,552 INFO org.apache.hadoop.ipc.Server: IPC Server handler 4 
on 8032, call 
org.apache.hadoop.yarn.api.ApplicationClientProtocolPB.getApplicationReport 
from 155.69.148.21:60571 Call#1416 Retry#0: error: 
org.apache.hadoop.yarn.exceptions.ApplicationNotFoundException: Application 
with id 'application_1386295642784_0001' doesn't exist in RM.
org.apache.hadoop.yarn.exceptions.ApplicationNotFoundException: Application 
with id 'application_1386295642784_0001' doesn't exist in RM.
        at 
org.apache.hadoop.yarn.server.resourcemanager.ClientRMService.getApplicationReport(ClientRMService.java:247)
        at 
org.apache.hadoop.yarn.api.impl.pb.service.ApplicationClientProtocolPBServiceImpl.getApplicationReport(ApplicationClientProtocolPBServiceImpl.java:120)
        at 
org.apache.hadoop.yarn.proto.ApplicationClientProtocol$ApplicationClientProtocolService$2.callBlockingMethod(ApplicationClientProtocol.java:241)
        at 
org.apache.hadoop.ipc.ProtobufRpcEngine$Server$ProtoBufRpcInvoker.call(ProtobufRpcEngine.java:585)
        at org.apache.hadoop.ipc.RPC$Server.call(RPC.java:928)
        at org.apache.hadoop.ipc.Server$Handler$1.run(Server.java:2048)
        at org.apache.hadoop.ipc.Server$Handler$1.run(Server.java:2044)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at 
org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1491)
        at org.apache.hadoop.ipc.Server$Handler.run(Server.java:2042)

Moreover, the strange point is that, if I don't set the queue 'amelie' in 
fair-scheduler.xml and run my job directly. The problem can run successfully by 
creating a queue called 'amelie' automatically. It seems that Yarn's fair 
scheduler do not support the job running on user's queue configuration in 
fair-scheduler.xml

Reply via email to