MPI signal handling

Posted by Seth Johnson on Stack Overflow See other posts from Stack Overflow or by Seth Johnson
Published on 2009-07-17T21:18:05Z Indexed on 2010/05/12 11:04 UTC
Read the original article Hit count: 446

Filed under:

mpi

|

signals

|

python

When using mpirun, is it possible to catch signals (for example, the SIGINT generated by ^C) in the code being run?

For example, I'm running a parallelized python code. I can except KeyboardInterrupt to catch those errors when running python blah.py by itself, but I can't when doing mpirun -np 1 python blah.py.

~~Does anyone have a suggestion? Even finding how to catch signals in a C or C++ compiled program would be a helpful start.~~

If I send a signal to the spawned Python processes, they can handle the signals properly; however, signals sent to the parent orterun process (i.e. from exceeding wall time on a cluster, or pressing control-C in a terminal) will kill everything immediately.

© Stack Overflow or respective owner

Related posts about mpi

Message Passing Interface (MPI)

as seen on Daniel Moth - Search for 'Daniel Moth'
So you have installed your cluster and you are done with introductory material on Windows HPC. Now you want to develop an application with the most common programming model: Message Passing Interface.The MPI programming model is a standard with implementations from many vendors. For newbies (like… >>> More
How do I debug an MPI program?

as seen on Stack Overflow - Search for 'Stack Overflow'
I have an MPI program which compiles and runs, but I would like to step through it to make sure nothing bizarre is happening. Ideally, I would like a simple way to attach GDB to any particular process, but I'm not really sure whether that's possible or how to do it. An alternative would be having… >>> More
trying to build Boost MPI, but the lib files are not created. What's going on?

as seen on Stack Overflow - Search for 'Stack Overflow'
I am trying to run a program with Boost MPI, but the thing is I don't have the .lib. So I try to create one by following the instruction at http://www.boost.org/doc/libs/1_43_0/doc/html/mpi/getting_started.html#mpi.config The instruction says "For many users using LAM/MPI, MPICH, or OpenMPI, configuration… >>> More
MPI: is there mpi libraries capable of message compression?

as seen on Stack Overflow - Search for 'Stack Overflow'
Sometimes MPI is used to send low-entropy data in messages. So it can be useful to try to compress messages before sending it. I know that MPI can work on very fast networks (10 Gbit/s and more), but many MPI programs are used with cheap network like 0,1G or 1Gbit/s Ethernet and with cheap (slow,… >>> More
Slides and code for MPI Cluster Debugger

as seen on Daniel Moth - Search for 'Daniel Moth'
I've blogged before about the MPI Cluster Debugger in VS2010 that facilitates launching the application on the cluster and attaching the debugger (btw, a shorter version of the screencast I link to there, is here).There have been requests for the code I use in the screencast, so please find a ZIP… >>> More

Related posts about signals

Is 37 Signals business advice applicable only to those with a captivated audience already eg courtes

as seen on Stack Overflow - Search for 'Stack Overflow'
I think that most of the advice that comes from 37 Signals is fascinating. I ask this question as i can't help also thinking that the some of the principals will only pull in customers if you have a fan base to call on such as the Ruby On Rails evangilists? >>> More
Scanf with Signals

as seen on Stack Overflow - Search for 'Stack Overflow'
I have a signal that blocks SIGINT and basically says "Sorry, you can't quit.\n" The issue is this can occur during a scanf. When this occurs during a scanf, scanf takes in the printf as input. How can I do a printf that will cause scanf to basically hit the enter key automatically. I don't care… >>> More
Functions connected to signals in QtScript (on Qt 4.5.2) are not firing

as seen on Stack Overflow - Search for 'Stack Overflow'
I've injected into a proprietary Qt (4.5.2) application, added my own compatible build of QtScript, and have managed to get access to all the signals I need. However, when connecting to them (via QtScript) my functions are never called. I've come up with a few theories for why this is and I've tested… >>> More
MPI signal handling

as seen on Stack Overflow - Search for 'Stack Overflow'
When using mpirun, is it possible to catch signals (for example, the SIGINT generated by ^C) in the code being run? For example, I'm running a parallelized python code. I can except KeyboardInterrupt to catch those errors when running python blah.py by itself, but I can't when doing mpirun -np 1… >>> More
Operating systems theory -- using minimum number of semaphores

as seen on Programmers - Search for 'Programmers'
This situation is prone to deadlock of processes in an operating system and I'd like to solve it with the minimum of semaphores. Basically there are three cooperating processes that all read data from the same input device. Each process, when it gets the input device, must read two consecutive data… >>> More