search

airair / graphlabapi

Automatically exported from code.google.com/p/graphlabapi
0 stars 0 forks source link

can't execute mpiexec my helloworld app #45

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
mpiexec -n 2 ./my_first_app 
ERROR:    dc_tcp_comm.cpp(new_socket:336): Check failed: all_addrs[id]==addr  
[162113728 == 164997312]
but when:
./my_first_app 
TCP Communication layer constructed.
Hello World!

I don't know what's wrong with it!

Original issue reported on code.google.com by wuyanzan...@gmail.com on 7 Mar 2013 at 7:48

GoogleCodeExporter commented 9 years ago 
mpiexec -n 2 ./my_first_app 
ERROR:    dc_tcp_comm.cpp(new_socket:336): Check failed: all_addrs[id]==addr  
[162113728 == 164997312]
FATAL:    dc_tcp_comm.cpp(connect:418): Failed to establish connection
terminate called after throwing an instance of 'char const*'
rank 1 in job 35  node9_38481   caused collective abort of all ranks
  exit status of rank 1: killed by signal 9

Original comment by wuyanzan...@gmail.com on 7 Mar 2013 at 8:13

GoogleCodeExporter commented 9 years ago 
I've found the problem!
The main problem is that one of the nodes of the mpi cluster have more than one 
ip!
My cluster consists of two nodes,a physical node and a virtual machine,while 
the physical node has 3 ip, so the init function in dc_tcp_comm.cpp check 
failed!

Original comment by wuyanzan...@gmail.com on 8 Mar 2013 at 3:41

GoogleCodeExporter commented 9 years ago 
I've found the problem!
The main problem is that one of the nodes of the mpi cluster have more than one 
ip!
My cluster consists of two nodes,a physical node and a virtual machine,while 
the physical node has 3 ip, so the init function in dc_tcp_comm.cpp check 
failed!

Original comment by wuyanzan...@gmail.com on 8 Mar 2013 at 6:43