locked
problem with mipch2 msmpi hp-mpi in windows Hpc 2008 RRS feed

  • Question

  • hello
    I have configured a cluster with 4 nodes (lmp2m-01,lmp2m-02,lmp2m-03 and lmp2m-04) on windows HPC 2008  with lmp2m-01 as head node.

    when I run   wmpiconfig in MPICH2 directory in the head node, I find only lmp2m-01.
    when I run fluent in parallel with two nodes (lmp2m-01 and lmp2m-02), I have the following message:

    abort: unable to connect to 'lmp2m-02: 8676',
    sock_error: generic socket failure, error stack'
    Mpidu_sock_post_connect (1200) unable to connect to lpmp-02 on port 8676, exhausted all endpoints (errno-1)
    Mpidu_sock_post_connect (1247) unable to connect to lpmp-02 on port 8676, a socket operation was attempted to an unreachable network (errno 10051)
    thanks for your help
    • Moved by parmita mehta Wednesday, February 24, 2010 8:48 AM (From:Windows HPC Server Deployment, Management, and Administration)
    Monday, February 22, 2010 11:01 AM

Answers

  • i have resolved this problem
    • Marked as answer by Don Pattee Wednesday, January 12, 2011 2:50 AM
    Monday, February 22, 2010 6:12 PM

All replies

  • i have resolved this problem
    • Marked as answer by Don Pattee Wednesday, January 12, 2011 2:50 AM
    Monday, February 22, 2010 6:12 PM
  • could you provide details on what was going on and what you needed to do to fix it?
    thanks

    pm
    Tuesday, February 23, 2010 3:25 PM