Tsuyoshi OZAWA created TEZ-1457: ----------------------------------- Summary: Taking 10 minutes to start a middle size job after submitting a job Key: TEZ-1457 URL: https://issues.apache.org/jira/browse/TEZ-1457 Project: Apache Tez Issue Type: Bug Reporter: Tsuyoshi OZAWA Attachments: syslog.txt
When I start a job 70GB, it always takes 10 minutes to start like this: <--- log1 - job start up time --> 14/08/18 16:02:36 INFO rpc.DAGClientRPCImpl: DAG: State: RUNNING Progress: 0% TotalTasks: 6 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 14/08/18 16:02:36 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Sorter Progress: 0% TotalTasks: 1 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 14/08/18 16:02:36 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Tokenizer Progress: 0% TotalTasks: -1 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 14/08/18 16:02:36 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Summation Progress: 0% TotalTasks: 6 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 -- next print is 10 minutes later -- 14/08/18 16:15:57 INFO rpc.DAGClientRPCImpl: DAG: State: RUNNING Progress: 1.04% TotalTasks: 96 Succeeded: 1 Running: 52 Failed: 0 Killed: 0 14/08/18 16:15:57 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Sorter Progress: 0% TotalTasks: 1 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 14/08/18 16:15:57 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Tokenizer Progress: 3.37% TotalTasks: 89 Succeeded: 3 Running: 52 Failed: 0 Killed: 0 14/08/18 16:15:57 INFO rpc.DAGClientRPCImpl: VertexStatus: VertexName: Summation Progress: 0% TotalTasks: 6 Succeeded: 0 Running: 0 Failed: 0 Killed: 0 <--- log1 --> -- This message was sent by Atlassian JIRA (v6.2#6252)