Segfault错误发生在R启动后（在新的conda环境中）。

huangapple
go评论100阅读模式
英文:

Segfault error directly after R startup (within fresh conda env)

问题

以下是翻译的部分内容：

这个问题似乎涉及到R，conda安装问题和可能的内存映射。

以下是一个已经出现问题的最小示例：

运行mamba create -n R_test r-base -c r似乎工作正常。但是当尝试从此环境中运行R时，我会得到以下错误（重要部分）：

Segfault错误发生在R启动后（在新的conda环境中）。

当查找此错误消息时，它经常与不同的包和库不匹配相关的问题一起出现（例如，https://stackoverflow.com/questions/49190251/caught-segfault-memory-not-mapped-error-in-r），这些问题似乎可以在R中进行检查... 除非它直接在启动时就发生，就像对我而言那样。

在另一个已经存在的R 4.1.2环境中运行R是有效的。指定新环境的版本也导致了相同的问题。所以似乎问题不在于版本，而是在于创建新环境时发生的事情。

我尝试使用conda clean --all来移除潜在有问题的未使用的包，以下是需要重启以移除文件句柄的一些包的示例，例如removing conda-env-2.6.0-1 WARNING conda.gateways.disk.delete:unlink_or_rename_to_trash(143): Could not remove or rename /opt/anaconda/anaconda3/pkgs/conda-env-2.6.0-1/info/recipe/meta.yaml. Please remove this file manually (you may need to reboot to free file handles)。

最小示例在我们计算集群的另一位用户那里可以正常工作，所以我排除了一些“全局”原因作为问题的原因。另外：我不能自己重启计算集群，所以我希望我能够自己解决这个问题。

有人有任何关于这里可能是什么问题或者（因为我不是IT专家）如何进一步调查正在发生的事情的想法吗？非常感谢您的建议！

编辑：根据@merv的评论进行了一些编辑和添加：

创建新环境的命令：mamba create -c conda-forge r-base -n R_test_cf

以下是为新conda环境安装的所有软件包：

  1. （未翻译的软件包列表）

这里是mamba env export的输出：

  1. （未翻译的输出）

conda info的详细信息：

  1. （未翻译的信息）

LD_LIBRARY_PATH没有值

ldd ~/.conda/envs/R_test_cf/bin/R 返回not a dynamic executable

另一位用户在集群上似乎能够简单地从我的环境中运行R... -->也许问题不在于安装方式，而在于如何激活环境或者我意外地全局安装了一些正在访问的内容。

英文:

The following issue seems to touch R, conda installation issues and possibly memory mapping.

Here is a minimal example that already results in the problem:

Running mamba create -n R_test r-base -c r seems to work just fine. But when trying to run R from within this env I get the following error (important part):

Segfault错误发生在R启动后（在新的conda环境中）。

When looking up this error message online, it often popped up in relation to distinct packages with library mismatches (e.g. https://stackoverflow.com/questions/49190251/caught-segfault-memory-not-mapped-error-in-r) which seem to be examinable from within R ... unless it happens directly at startup like it does for me.

Running R in another, already existing, R 4.1.2 env worked. Also scpecifying that version for a new env resulted in the same problem. So it seems, rather than the version, the issue is with whatever happens when I icreate a new envirmonment.

I tried removing potentially problematic unused packages with conda clean --all and the follwing packages seem to require reboot to remove file handle, e.g.removing conda-env-2.6.0-1
WARNING conda.gateways.disk.delete:unlink_or_rename_to_trash(143): Could not remove or rename /opt/anaconda/anaconda3/pkgs/conda-env-2.6.0-1/info/recipe/meta.yaml. Please remove this file manually (you may need to reboot to free file handles):

  1. blas-1.0-mkl                                  15 KB
  2. conda-env-2.6.0-1                              6 KB
  3. _libgcc_mutex-0.1-main                         7 KB
  4. path.py-12.5.0-0                               8 KB
  5. get_terminal_size-1.0.0-haa9412d_0             7 KB
  6. importlib_metadata-3.10.0-hd3eb1b0_0          37 KB
  7. prompt_toolkit-3.0.17-hd3eb1b0_0              39 KB
  8. numpy-1.20.1-py38h93e21f0_0                   70 KB
  9. matplotlib-3.3.4-py38h06a4308_0               92 KB
  10. anaconda-2021.05-py38_0                       81 KB
  11. dask-2021.4.0-pyhd3eb1b0_0                    12 KB
  12. _openmp_mutex-4.5-1_gnu                       93 KB
  13. openssl-1.1.1k-h7f98852_0                    6.0 MB
  14. conda-4.10.1-py38h06a4308_1                 11.7 MB
  15. python_abi-3.9-2_cp39                         11 KB
  16. _libgcc_mutex-0.1-conda_forge                  6 KB
  17. python_abi-3.8-2_cp38                         11 KB
  18. libgfortran-ng-7.5.0-ha8ba4b0_17              92 KB
  19. numpy-1.21.5-py38he7a7128_2                   26 KB

The minimal example works for someone else on our computing cluster, so I'd exclude some "global" reasons as the culprit. Btw: I cannot reboot the cluster myself, so I am hoing there might be something I can do on my own to resolve the issue.

Does anyone have an idea what could be the problem here or (since I am not an IT expert) how to further investigate what's going on? Thanks a lot for your suggestions!

---------------------------------EDIT--------------------------------------

Some edits and additions based on comments by @merv :

New environment creation command: mamba create -c conda-forge r-base -n R_test_cf

Here's all packages installed for the new conda env:

  1. _libgcc_mutex                       0.1  conda_forge          conda-forge/linux-64     Cached
  2. _openmp_mutex                       4.5  2_gnu                conda-forge/linux-64     Cached
  3. _r-mutex                          1.0.1  anacondar_1          conda-forge/noarch       Cached
  4. binutils_impl_linux-64             2.40  hf600244_0           conda-forge/linux-64     Cached
  5. blas                                1.1  openblas             conda-forge/linux-64     Cached
  6. bwidget                          1.9.14  ha770c72_1           conda-forge/linux-64     Cached
  7. bzip2                             1.0.8  h7f98852_4           conda-forge/linux-64     Cached
  8. c-ares                           1.19.0  h5eee18b_0           pkgs/main/linux-64       Cached
  9. ca-certificates              2023.01.10  h06a4308_0           pkgs/main/linux-64       Cached
  10. cairo                            1.16.0  h35add3b_1015        conda-forge/linux-64     Cached
  11. curl                             7.88.1  hdc1c0ab_1           conda-forge/linux-64     Cached
  12. expat                             2.5.0  hcb278e6_1           conda-forge/linux-64     Cached
  13. font-ttf-dejavu-sans-mono          2.37  hab24e00_0           conda-forge/noarch       Cached
  14. font-ttf-inconsolata              3.000  h77eed37_0           conda-forge/noarch       Cached
  15. font-ttf-source-code-pro          2.038  h77eed37_0           conda-forge/noarch       Cached
  16. font-ttf-ubuntu                    0.83  hab24e00_0           conda-forge/noarch       Cached
  17. fontconfig                       2.14.2  h14ed4e7_0           conda-forge/linux-64     Cached
  18. fonts-conda-ecosystem                 1  0                    conda-forge/noarch       Cached
  19. fonts-conda-forge                     1  0                    conda-forge/noarch       Cached
  20. freetype                         2.12.1  hca18f0e_1           conda-forge/linux-64     Cached
  21. fribidi                          1.0.10  h516909a_0           conda-forge/linux-64     Cached
  22. gcc_impl_linux-64                12.2.0  hcc96c02_19          conda-forge/linux-64     Cached
  23. gettext                          0.21.1  h27087fc_0           conda-forge/linux-64     Cached
  24. gfortran_impl_linux-64           12.2.0  h55be85b_19          conda-forge/linux-64     Cached
  25. graphite2                        1.3.14  h295c915_1           pkgs/main/linux-64       Cached
  26. gsl                               2.7.1  h917eeb2_1           pkgs/main/linux-64       Cached
  27. gxx_impl_linux-64                12.2.0  hcc96c02_19          conda-forge/linux-64     Cached
  28. harfbuzz                          6.0.0  h3ff4399_1           conda-forge/linux-64     Cached
  29. icu                                72.1  hcb278e6_0           conda-forge/linux-64     Cached
  30. kernel-headers_linux-64          2.6.32  he073ed8_15          conda-forge/noarch       Cached
  31. keyutils                          1.6.1  h166bdaf_0           conda-forge/linux-64     Cached
  32. krb5                             1.20.1  h81ceb04_0           conda-forge/linux-64     Cached
  33. ld_impl_linux-64                   2.40  h41732ed_0           conda-forge/linux-64     Cached
  34. lerc                              4.0.0  h27087fc_0           conda-forge/linux-64     Cached
  35. libblas                           3.9.0  16_linux64_openblas  conda-forge/linux-64     Cached
  36. libcurl                          7.88.1  hdc1c0ab_1           conda-forge/linux-64     Cached
  37. libdeflate                         1.18  h0b41bf4_0           conda-forge/linux-64     Cached
  38. libedit                    3.1.20221030  h5eee18b_0           pkgs/main/linux-64       Cached
  39. libev                              4.33  h516909a_1           conda-forge/linux-64     Cached
  40. libexpat                          2.5.0  hcb278e6_1           conda-forge/linux-64     Cached
  41. libffi                            3.4.2  h7f98852_5           conda-forge/linux-64     Cached
  42. libgcc-devel_linux-64            12.2.0  h3b97bd3_19          conda-forge/linux-64     Cached
  43. libgcc-ng                        12.2.0  h65d4601_19          conda-forge/linux-64     Cached
  44. libgfortran-ng                   12.2.0  h69a702a_19          conda-forge/linux-64     Cached
  45. libgfortran5                     12.2.0  h337968e_19          conda-forge/linux-64     Cached
  46. libglib                          2.74.1  h606061b_1           conda-forge/linux-64     Cached
  47. libgomp                          12.2.0  h65d4601_19          conda-forge/linux-64     Cached
  48. libiconv                           1.17  h166bdaf_0           conda-forge/linux-64     Cached
  49. libjpeg-turbo                   2.1.5.1  h0b41bf4_0           conda-forge/linux-64     Cached
  50. liblapack                         3.9.0  16_linux64_openblas  conda-forge/linux-64     Cached
  51. libnghttp2                       1.52.0  h61bc06f_0           conda-forge/linux-64     Cached
  52. libopenblas                      0.3.21  pthreads_h78a6416_3  conda-forge/linux-64     Cached
  53. libpng                           1.6.39  h753d276_0           conda-forge/linux-64     Cached
  54. libsanitizer                     12.2.0  h46fd767_19          conda-forge/linux-64     Cached
  55. libssh2                          1.10.0  hf14f497_3           conda-forge/linux-64     Cached
  56. libstdcxx-devel_linux-64         12.2.0  h3b97bd3_19          conda-forge/linux-64     Cached
  57. libstdcxx-ng                     12.2.0  h46fd767_19          conda-forge/linux-64     Cached
  58. libtiff                           4.5.0  ha587672_6           conda-forge/linux-64     Cached
  59. libuuid                          2.38.1  h0b41bf4_0           conda-forge/linux-64     Cached
  60. libwebp-base                      1.3.0  h0b41bf4_0           conda-forge/linux-64     Cached
  61. libxcb                             1.13  h7f98852_1004        conda-forge/linux-64     Cached
  62. libxml2                          2.10.3  hfdac1af_6           conda-forge/linux-64     Cached
  63. libzlib                          1.2.13  h166bdaf_4           conda-forge/linux-64     Cached
  64. lz4-c                             1.9.4  hcb278e6_0           conda-forge/linux-64     Cached
  65. make                                4.3  hd18ef5c_1           conda-forge/linux-64     Cached
  66. ncurses                             6.4  h6a678d5_0           pkgs/main/linux-64       Cached
  67. openblas                         0.3.21  pthreads_h320a7e8_3  conda-forge/linux-64     Cached
  68. openssl                           3.1.0  h0b41bf4_0           conda-forge/linux-64     Cached
  69. pango                           1.50.14  hd33c08f_0           conda-forge/linux-64     Cached
  70. pcre2                             10.40  hc3806b6_0           conda-forge/linux-64     Cached
  71. pixman                           0.40.0  h36c2ea0_0           conda-forge/linux-64     Cached
  72. pthread-stubs                       0.4  h36c2ea0_1001        conda-forge/linux-64     Cached
  73. r-base                            4.2.3  h4a03800_2           conda-forge/linux-64     Cached
  74. readline                            8.2  h8228510_1           conda-forge/linux-64     Cached
  75. sed                                 4.8  he412f7d_0           conda-forge/linux-64     Cached
  76. sysroot_linux-64                   2.12  he073ed8_15          conda-forge/noarch       Cached
  77. tk                               8.6.12  h27826a3_0           conda-forge/linux-64     Cached
  78. tktable                            2.10  hb7b940f_3           conda-forge/linux-64     Cached
  79. xorg-kbproto                      1.0.7  h14c3975_1002        conda-forge/linux-64     Cached
  80. xorg-libice                      1.0.10  h516909a_0           conda-forge/linux-64     Cached
  81. xorg-libsm                        1.2.3  hd9c2040_1000        conda-forge/linux-64     Cached
  82. xorg-libx11                       1.8.4  h0b41bf4_0           conda-forge/linux-64     Cached
  83. xorg-libxau                       1.0.9  h14c3975_0           conda-forge/linux-64     Cached
  84. xorg-libxdmcp                     1.1.3  h516909a_0           conda-forge/linux-64     Cached
  85. xorg-libxext                      1.3.4  h0b41bf4_2           conda-forge/linux-64     Cached
  86. xorg-libxrender                  0.9.10  h7f98852_1003        conda-forge/linux-64     Cached
  87. xorg-libxt                        1.2.1  h7f98852_2           conda-forge/linux-64     Cached
  88. xorg-renderproto                 0.11.1  h14c3975_1002        conda-forge/linux-64     Cached
  89. xorg-xextproto                    7.3.0  h0b41bf4_1003        conda-forge/linux-64     Cached
  90. xorg-xproto                      7.0.31  h14c3975_1007        conda-forge/linux-64     Cached
  91. xz                               5.2.10  h5eee18b_1           pkgs/main/linux-64       Cached
  92. zlib                             1.2.13  h166bdaf_4           conda-forge/linux-64     Cached
  93. zstd                              1.5.4  hc292b87_0           pkgs/main/linux-64       Cached
  94. Summary:
  95. Install: 93 packages
  96. Total download: 0  B

Here's the output of mamba env export:

  1. name: R_test_cf
  2. channels:
  3. - conda-forge
  4. - defaults
  5. dependencies:
  6. - _libgcc_mutex=0.1=conda_forge
  7. - _openmp_mutex=4.5=2_gnu
  8. - _r-mutex=1.0.1=anacondar_1
  9. - binutils_impl_linux-64=2.40=hf600244_0
  10. - blas=1.1=openblas
  11. - bwidget=1.9.14=ha770c72_1
  12. - bzip2=1.0.8=h7f98852_4
  13. - c-ares=1.19.0=h5eee18b_0
  14. - ca-certificates=2023.01.10=h06a4308_0
  15. - cairo=1.16.0=h35add3b_1015
  16. - curl=7.88.1=hdc1c0ab_1
  17. - expat=2.5.0=hcb278e6_1
  18. - font-ttf-dejavu-sans-mono=2.37=hab24e00_0
  19. - font-ttf-inconsolata=3.000=h77eed37_0
  20. - font-ttf-source-code-pro=2.038=h77eed37_0
  21. - font-ttf-ubuntu=0.83=hab24e00_0
  22. - fontconfig=2.14.2=h14ed4e7_0
  23. - fonts-conda-ecosystem=1=0
  24. - fonts-conda-forge=1=0
  25. - freetype=2.12.1=hca18f0e_1
  26. - fribidi=1.0.10=h516909a_0
  27. - gcc_impl_linux-64=12.2.0=hcc96c02_19
  28. - gettext=0.21.1=h27087fc_0
  29. - gfortran_impl_linux-64=12.2.0=h55be85b_19
  30. - graphite2=1.3.14=h295c915_1
  31. - gsl=2.7.1=h917eeb2_1
  32. - gxx_impl_linux-64=12.2.0=hcc96c02_19
  33. - harfbuzz=6.0.0=h3ff4399_1
  34. - icu=72.1=hcb278e6_0
  35. - kernel-headers_linux-64=2.6.32=he073ed8_15
  36. - keyutils=1.6.1=h166bdaf_0
  37. - krb5=1.20.1=h81ceb04_0
  38. - ld_impl_linux-64=2.40=h41732ed_0
  39. - lerc=4.0.0=h27087fc_0
  40. - libblas=3.9.0=16_linux64_openblas
  41. - libcurl=7.88.1=hdc1c0ab_1
  42. - libdeflate=1.18=h0b41bf4_0
  43. - libedit=3.1.20221030=h5eee18b_0
  44. - libev=4.33=h516909a_1
  45. - libexpat=2.5.0=hcb278e6_1
  46. - libffi=3.4.2=h7f98852_5
  47. - libgcc-devel_linux-64=12.2.0=h3b97bd3_19
  48. - libgcc-ng=12.2.0=h65d4601_19
  49. - libgfortran-ng=12.2.0=h69a702a_19
  50. - libgfortran5=12.2.0=h337968e_19
  51. - libglib=2.74.1=h606061b_1
  52. - libgomp=12.2.0=h65d4601_19
  53. - libiconv=1.17=h166bdaf_0
  54. - libjpeg-turbo=2.1.5.1=h0b41bf4_0
  55. - liblapack=3.9.0=16_linux64_openblas
  56. - libnghttp2=1.52.0=h61bc06f_0
  57. - libopenblas=0.3.21=pthreads_h78a6416_3
  58. - libpng=1.6.39=h753d276_0
  59. - libsanitizer=12.2.0=h46fd767_19
  60. - libssh2=1.10.0=hf14f497_3
  61. - libstdcxx-devel_linux-64=12.2.0=h3b97bd3_19
  62. - libstdcxx-ng=12.2.0=h46fd767_19
  63. - libtiff=4.5.0=ha587672_6
  64. - libuuid=2.38.1=h0b41bf4_0
  65. - libwebp-base=1.3.0=h0b41bf4_0
  66. - libxcb=1.13=h7f98852_1004
  67. - libxml2=2.10.3=hfdac1af_6
  68. - libzlib=1.2.13=h166bdaf_4
  69. - lz4-c=1.9.4=hcb278e6_0
  70. - make=4.3=hd18ef5c_1
  71. - ncurses=6.4=h6a678d5_0
  72. - openblas=0.3.21=pthreads_h320a7e8_3
  73. - openssl=3.1.0=h0b41bf4_0
  74. - pango=1.50.14=hd33c08f_0
  75. - pcre2=10.40=hc3806b6_0
  76. - pixman=0.40.0=h36c2ea0_0
  77. - pthread-stubs=0.4=h36c2ea0_1001
  78. - r-base=4.2.3=h4a03800_2
  79. - readline=8.2=h8228510_1
  80. - sed=4.8=he412f7d_0
  81. - sysroot_linux-64=2.12=he073ed8_15
  82. - tk=8.6.12=h27826a3_0
  83. - tktable=2.10=hb7b940f_3
  84. - xorg-kbproto=1.0.7=h14c3975_1002
  85. - xorg-libice=1.0.10=h516909a_0
  86. - xorg-libsm=1.2.3=hd9c2040_1000
  87. - xorg-libx11=1.8.4=h0b41bf4_0
  88. - xorg-libxau=1.0.9=h14c3975_0
  89. - xorg-libxdmcp=1.1.3=h516909a_0
  90. - xorg-libxext=1.3.4=h0b41bf4_2
  91. - xorg-libxrender=0.9.10=h7f98852_1003
  92. - xorg-libxt=1.2.1=h7f98852_2
  93. - xorg-renderproto=0.11.1=h14c3975_1002
  94. - xorg-xextproto=7.3.0=h0b41bf4_1003
  95. - xorg-xproto=7.0.31=h14c3975_1007
  96. - xz=5.2.10=h5eee18b_1
  97. - zlib=1.2.13=h166bdaf_4
  98. - zstd=1.5.4=hc292b87_0

The conda info details:

  1.      active environment : R_test_cf
  2. active env location : /home/rstrasser/.conda/envs/R_test_cf
  3. shell level : 1
  4. user config file : /home/rstrasser/.condarc
  5. populated config files : /home/rstrasser/.condarc
  6. conda version : 4.12.0
  7. conda-build version : 3.21.4
  8. python version : 3.8.8.final.0
  9. virtual packages : __linux=3.10.0=0
  10. __glibc=2.17=0
  11. __unix=0=0
  12. __archspec=1=x86_64
  13. base environment : /opt/anaconda/anaconda3  (read only)
  14. conda av data dir : /opt/anaconda/anaconda3/etc/conda
  15. conda av metadata url : None
  16. channel URLs : https://conda.anaconda.org/conda-forge/linux-64
  17. https://conda.anaconda.org/conda-forge/noarch
  18. https://repo.anaconda.com/pkgs/main/linux-64
  19. https://repo.anaconda.com/pkgs/main/noarch
  20. https://repo.anaconda.com/pkgs/r/linux-64
  21. https://repo.anaconda.com/pkgs/r/noarch
  22. package cache : /opt/anaconda/anaconda3/pkgs
  23. /home/rstrasser/.conda/pkgs
  24. envs directories : /home/rstrasser/.conda/envs
  25. /opt/anaconda/anaconda3/envs
  26. platform : linux-64
  27. user-agent : conda/4.12.0 requests/2.25.1 CPython/3.8.8 Linux/3.10.0-1160.53.1.el7.x86_64 centos/7.9.2009 glibc/2.17
  28. UID:GID : 1003:1004
  29. netrc file : None
  30. offline mode : False

LD_LIBRARY_PATH does not have a value

ldd ~/.conda/envs/R_test_cf/bin/R returns not a dynamic executable

-------------------EDIT 2-----------------------------------------

Another user on the cluster seems to be able to simply run R from my environment... --> Maybe the issue is not how things are installed, but how the env is activated or that I accidentally installed something globally that is being accessed now.

答案1

得分: 2

  1. 我们在我们的大学集群遇到了相同的问题。我们在.bashrc文件的末尾添加了以下行：
  2. LANG=en_US.UTF-8
英文:

We faced the same problem at our university cluster. We added the line

  1. LANG=en_US.UTF-8

at the end of the .bashrc file and it solved the problem.

答案2

得分: 0

我遇到了和你类似的问题。而且我发现在核心转储之前打印了一些警告消息。
警告消息
所以我觉得可能是地区设置是这种奇怪现象的根源。我执行了以下命令并通过conda重新安装了R。

  1. sudo apt-get install language-pack-en

对我来说有效。
结果

英文:

I faced a similar problem to you.
And I found that some Warning messages are printed before the core dump.
Warning messages
So I feel maybe the locale setting is the source of this strange phenomenon. I executed the following command and reinstalled R through conda.

  1. sudo apt-get install language-pack-en

And it works for me.
Result

huangapple
  • 本文由 发表于 2023年3月31日 18:37:21
  • 转载请务必保留本文链接：https://go.coder-hub.com/75897562.html
  • conda
  • r
  • segmentation-fault
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定