通过Python在MPI_Init中启动打开MPI时出错 [英] Error when starting Open MPI in MPI_Init via Python

问题描述

我试图通过python访问与OpenMPI共享库，但由于某种原因，我得到以下错误信息：

  [Geo00433：01196] mca：base：component_find：无法打开/ usr / li / openmpi / lib / openmpi / mca_paffinity_hwloc：可能是一个缺少的符号或是为不同版本的Open MPI编译的？ （忽略）
 [Geo00433：01196] mca：base：component_find：无法打开/ usr / lib / openmpi / lib / openmpi / mca_carto_auto_detect：可能缺少符号或为不同版本的Open MPI编译？ （忽略）
 [Geo00433：01196] mca：base：component_find：无法打开/ usr / lib / openmpi / lib / openmpi / mca_carto_file：可能缺少符号或为不同版本的Open MPI编译？ （忽略）
 [Geo00433：01196] mca：base：component_find：无法打开/ usr / lib / openmpi / lib / openmpi / mca_shmem_mmap：可能缺少符号或为不同版本的Open MPI编译？ （忽略）
 [Geo00433：01196] mca：base：component_find：无法打开/ usr / lib / openmpi / lib / openmpi / mca_shmem_posix：可能缺少符号或为不同版本的Open MPI编译？ （忽略）
 [Geo00433：01196] mca：base：component_find：无法打开/ usr / lib / openmpi / lib / openmpi / mca_shmem_sysv：可能是一个缺少的符号或者为不同版本的Open MPI编译？ （忽略）
 ------------------------------------------- ------------------------------ 
看起来像opal_init由于某种原因失败;你的并行过程是
可能中止。有很多原因导致并行进程可能在opal_init期间
失败;其中一些是由于配置或
环境问题。这种故障似乎是内部故障; 
这里是一些附加信息（它可能只与
 Open MPI开发人员相关）：
 
 opal_shmem_base_select failed 
  - >返回值-1而不是OPAL_SUCCESS 
 --------------------------------------- ----------------------------------- 
 [Geo00433：01196] [[INVALID]，无效] ORTE_ERROR_LOG：在第79行的文件运行时/ orte_init.c中出错
 -------------------------------- ------------------------------------------ 
它看起来像MPI_INIT由于某种原因失败;你的并行过程是
可能中止。有许多原因导致并行进程可能在MPI_INIT期间
失败;其中一些是由于配置或环境
的问题。这种故障似乎是内部故障;这里是一些
的附加信息（它可能只与一个Open MPI 
开发者相关）：
 
 ompi_mpi_init：orte_init failed 
  - >返回错误（-1）而不是成功（0）
 ------------------------------ -------------------------------------------- 
 ** * MPI_Init中出现错误
 ***在NULL通信器上
 *** MPI_ERRORS_ARE_FATAL：您的MPI作业现在将中止
 [Geo00433：1196]在MPI_INIT成功完成之前本地中止;无法聚合错误消息，并且无法保证所有其他进程被杀死！ 
  

任何线索是什么原因？我已经检查了很多网页，但是无法找到我的问题的解决方案。

我安装了Ubuntu 15.10和mpich以及open-mpi。

解决方案

我有同样的问题非常相似与稍微不同的错误消息）在Ubuntu 16.04，即使只安装打开MPI。从我可以知道有一个问题，如何从Ubuntu的mpi4py包建成，但不知道到底是什么。

繁殖： strong>因为这个问题不能完全清楚错误消息是如何产生的（我没有声誉来编辑它），这里是我得到它。首先，安装Ubuntu的mpi4py包，然后进入python环境：

  $ sudo apt-get install mpi 
 $ python 
  

在python内，尝试以下操作：

 >>> from mpi4py import MPI 
  

然后，您应该会收到一条类似OP的错误信息。

解决方案：这是我的工作方式。首先卸载Ubuntu的软件包：

  $ sudo apt-get remove mpi4py 
  

然后安装Open MPI头（下一步涉及构建mpi4py）和pip：

  $ sudo apt-get install libopenmpi-dev python-pip 
  

最后安装mpi4py ：

  $ sudo pip install mpi4py 
  

如果你尝试上面的python命令，它现在应该可以正常工作。

I am trying to access a shared library with OpenMPI via python, but for some reason I get the following error message:

[Geo00433:01196] mca: base: component_find: unable to open /usr/li/openmpi/lib/openmpi/mca_paffinity_hwloc: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
[Geo00433:01196] mca: base: component_find: unable to open /usr/lib/openmpi/lib/openmpi/mca_carto_auto_detect: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
[Geo00433:01196] mca: base: component_find: unable to open /usr/lib/openmpi/lib/openmpi/mca_carto_file: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
[Geo00433:01196] mca: base: component_find: unable to open /usr/lib/openmpi/lib/openmpi/mca_shmem_mmap: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
[Geo00433:01196] mca: base: component_find: unable to open /usr/lib/openmpi/lib/openmpi/mca_shmem_posix: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
[Geo00433:01196] mca: base: component_find: unable to open /usr/lib/openmpi/lib/openmpi/mca_shmem_sysv: perhaps a missing symbol, or compiled for a different version of Open MPI? (ignored)
-------------------------------------------------------------------------
It looks like opal_init failed for some reason; your parallel process is
likely to abort.  There are many reasons that a parallel process can
fail during opal_init; some of which are due to configuration or
environment problems.  This failure appears to be an internal failure;
here is some additional information (which may only be relevant to an
Open MPI developer):

  opal_shmem_base_select failed
    --> Returned value -1 instead of OPAL_SUCCESS
--------------------------------------------------------------------------
[Geo00433:01196] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file runtime/orte_init.c at line 79
--------------------------------------------------------------------------
It looks like MPI_INIT failed for some reason; your parallel process is
likely to abort.  There are many reasons that a parallel process can
fail during MPI_INIT; some of which are due to configuration or environment
problems.  This failure appears to be an internal failure; here is some
additional information (which may only be relevant to an Open MPI
developer):

  ompi_mpi_init: orte_init failed
  --> Returned "Error" (-1) instead of "Success" (0)
--------------------------------------------------------------------------
*** An error occurred in MPI_Init
*** on a NULL communicator
*** MPI_ERRORS_ARE_FATAL: your MPI job will now abort
[Geo00433:1196] Local abort before MPI_INIT completed successfully; not able to aggregate error messages, and not able to guarantee that all other processes were killed!

Any clue what's the reason? I checked many webpages already, but somehow couldn't find a solution for my problem yet.

I have Ubuntu 15.10 and mpich as well as open-mpi installed.

Thanks a lot guys!

解决方案

I had the same problem (or very similar with slightly different error message) on Ubuntu 16.04, even with only Open MPI installed. From what I can tell there is a problem with how the mpi4py package from Ubuntu was built, but am not sure what exactly it is.

Reproduction: Since the question doesn't make it entirely clear how the error message was produced (I don't have the reputation to edit it), here's how I got it. First, install Ubuntu's mpi4py package and then enter the python environment:

$ sudo apt-get install mpi
$ python

Inside python, try the following:

>>> from mpi4py import MPI

You should then get an error message like the OP had.

Solution: Here is how I got it working. First uninstall Ubuntu's package:

$ sudo apt-get remove mpi4py

Then install the Open MPI headers (the next step involves building mpi4py) and pip:

$ sudo apt-get install libopenmpi-dev python-pip

Finally install mpi4py:

$ sudo pip install mpi4py

If you try the python command above, it should now work fine.

这篇关于通过Python在MPI_Init中启动打开MPI时出错的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！