Thanks Hui,  is the spawned process on the local host, or the remote host or both?

 

Kurt

 

From: Zhou, Hui <zhouh@anl.gov>
Sent: Friday, April 1, 2022 4:20 PM
To: discuss@mpich.org
Cc: Mccall, Kurt E. (MSFC-EV41) <kurt.e.mccall@nasa.gov>
Subject: [EXTERNAL] Re: Hydra WARNING: too many ssh connections

 

Every time you call MPI_Comm_spawn, hydra will launch a ssh (for each host) to create a proxy. It is certainly not ideal for applications relying on spawning many processes.


From: Mccall, Kurt E. (MSFC-EV41) via discuss <discuss@mpich.org>
Sent: Friday, April 1, 2022 4:08 PM
To: discuss@mpich.org <discuss@mpich.org>
Cc: Mccall, Kurt E. (MSFC-EV41) <kurt.e.mccall@nasa.gov>
Subject: [mpich-discuss] Hydra WARNING: too many ssh connections

 

Hi,  you provided the following information about the warning “too many ssh connections”:

 

The particular warning is issued by hydra, MPICH’s process manager. Following excerpt is the comment in that source code:

 

        /* ssh has many types of security controls that do not allow a

         * user to ssh to the same node multiple times very

         * quickly. If this happens, the ssh daemons disables ssh

         * connections causing the job to fail. This is basically a

         * hack to slow down ssh connections to the same node. We

         * check for offset == 0 before applying this hack, so we only

         * slow down the cases where ssh is being used, and not the

         * cases where we fall back to fork. */

 

Is this just during an initial ssh connection attempt?  I’m trying to figure out where my code is triggering this warning.  Could it be from

 

  1. MPI_Intercomm_create
  2. MPI_Comm_spawn
  3. others?

 

I’m calling mpiexec with the “—launcher ssh” option, MPICH 4.0.1.

 

Thanks,

Kurt