hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jie Tao <jie....@gameforge.com>
Subject problem with mapreduce: map output flush
Date Fri, 17 Jun 2016 15:19:58 GMT
Hi,

I ran a MR job in context of Kylin cube building and the job was stuck 
in the step of flushing map output buffer to disk:

INFO [main] org.apache.hadoop.mapred.MapTask: Starting flush of map output
INFO [main] org.apache.hadoop.mapred.MapTask: Spilling map output
INFO [main] org.apache.hadoop.mapred.MapTask: bufstart = 0; bufend = 
190884986; bufvoid = 536870912
INFO [main] org.apache.hadoop.mapred.MapTask: kvstart = 
134217724(536870896); kvend = 116938212(467752848); length = 
17279513/33554432

then the following happens till map task timeout (one hour). It says 
that the client keeps connection to the server. with netstat I have seen 
that the connection has been built between the two machines. But I 
cannot see any data at the directory where the map output shall be 
flushed to.

Maybe any Hadoop configuration or /etc/hosts wrong? I checked but cannot 
figure out what shall be wrong. I use Hadoop 2.4.1. Our cluster has 3 
worker nodes.

Help is needed!

016-06-16 16:39:08,622 DEBUG [IPC Client (1991820533) connection to 
/ip:49222 from job_1466061903809_0024] org.apache.hadoop.ipc.Client: IPC 
Client (1991820533) connection to /ip:49222 from job_1466061903809_0024 
got value #211
2016-06-16 16:39:08,622 DEBUG [communication thread] 
org.apache.hadoop.ipc.RPC: Call: ping 1
2016-06-16 16:39:11,623 DEBUG [IPC Parameter Sending Thread #0] 
org.apache.hadoop.ipc.Client: IPC Client (1991820533) connection to 
/ip:49222 from job_1466061903809_0024 sending #212
2016-06-16 16:39:11,624 DEBUG [IPC Client (1991820533) connection to 
/ip:49222 from job_1466061903809_0024] org.apache.hadoop.ipc.Client: IPC 
Client (1991820533) connection to /ip:49222 from job_1466061903809_0024 
got value #212
2016-06-16 16:39:11,624 DEBUG [communication thread] 
org.apache.hadoop.ipc.RPC: Call: ping 1
2016-06-16 16:39:14,625 DEBUG [IPC Parameter Sending Thread #0] 
org.apache.hadoop.ipc.Client: IPC Client (1991820533) connection to 
/ip:49222 from job_1466061903809_0024 sending #213
2016-06-16 16:39:14,626 DEBUG [IPC Client (1991820533) connection to 
/ip:49222 from job_1466061903809_0024] org.apache.hadoop.ipc.Client: IPC 
Client (1991820533) connection to /ip:49222 from job_1466061903809_0024 
got value #213
2016-06-16 16:39:14,626 DEBUG [communication thread] 
org.apache.hadoop.ipc.RPC: Call: ping 1


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@hadoop.apache.org
For additional commands, e-mail: user-help@hadoop.apache.org


Mime
View raw message