JungWhoNam / ospray

An Open, Scalable, Portable, Ray Tracing Based Rendering Engine for High-Fidelity Visualization
http://ospray.org
Apache License 2.0
0 stars 0 forks source link

error outputs when running and closing mpi examples in a docker container #8

Open ghost opened 1 year ago

ghost commented 1 year ago

I am trying to run mpi examples inside a docker container using X-Server to show the GUI. When I run provided examples, I get the following errors. Despite the errors, the programs runs fine.

ospray@d49abb85a656:~/build-container/install/ospray/bin$ mpirun -n 2 ospMPIDistribTutorialSpheres 
OSPRay rank 0/2
OSPRay rank 1/2
[d49abb85a656:06401] Read -1, expected 81992, errno = 1
[d49abb85a656:06401] Read -1, expected 81992, errno = 1
[d49abb85a656:06401] Read -1, expected 81992, errno = 1
[d49abb85a656:06401] Read -1, expected 81992, errno = 1

ospMPIDistribTutorialSpheres runs fine despite the errors.

ospray@d49abb85a656:~/build-container/install/ospray/bin$ mpirun -n 3 ospMPIMultiDisplays 
OSPRay rank 1/3
OSPRay rank 2/3
OSPRay rank 0/3
[d49abb85a656:06561] *** Process received signal ***
[d49abb85a656:06561] Signal: Segmentation fault (11)
[d49abb85a656:06561] Signal code: Address not mapped (1)
[d49abb85a656:06561] Failing at address: 0x558446688403
[d49abb85a656:06560] *** Process received signal ***
[d49abb85a656:06560] Signal: Segmentation fault (11)
[d49abb85a656:06560] Signal code: Address not mapped (1)
[d49abb85a656:06560] Failing at address: 0x558446688403
[d49abb85a656:06561] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x12980)[0x7fc7b1951980]
[d49abb85a656:06561] [ 1] /lib/x86_64-linux-gnu/libc.so.6(+0x18ea13)[0x7fc7b01e3a13]
[d49abb85a656:06561] [ 2] [d49abb85a656:06560] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x12980)[0x7f22a0700980]
[d49abb85a656:06560] [ 1] /lib/x86_64-linux-gnu/libc.so.6(+0x18ea13)[0x7f229ef92a13]
[d49abb85a656:06560] [ 2] /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE9_M_assignERKS4_+0x8d)[0x7f229f8cfaad]
/usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE9_M_assignERKS4_+0x8d)[0x7fc7b0b20aad]
[d49abb85a656:06561] [ 3] ospMPIMultiDisplays(_ZN16GLFWOSPRayWindow19startNewOSPRayFrameEv+0x197)[0x563cb7a1f5d7]
[d49abb85a656:06561] [ 4] [d49abb85a656:06560] [ 3] ospMPIMultiDisplays(_ZN16GLFWOSPRayWindow19startNewOSPRayFrameEv+0x197)[0x55b71ce1f5d7]
[d49abb85a656:06560] [ 4] ospMPIMultiDisplays(_ZN16GLFWOSPRayWindow8mainLoopEv+0x75)[0x55b71ce20155]
[d49abb85a656:06560] [ 5] ospMPIMultiDisplays(_ZN16GLFWOSPRayWindow8mainLoopEv+0x75)[0x563cb7a20155]
[d49abb85a656:06561] [ 5] ospMPIMultiDisplays(main+0x101)[0x563cb7a1a541]
[d49abb85a656:06561] [ 6] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7)[0x7fc7b0076c87]
[d49abb85a656:06561] [ 7] ospMPIMultiDisplays(_start+0x2a)[0x563cb7a1aaca]
[d49abb85a656:06561] *** End of error message ***
ospMPIMultiDisplays(main+0x101)[0x55b71ce1a541]
[d49abb85a656:06560] [ 6] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7)[0x7f229ee25c87]
[d49abb85a656:06560] [ 7] ospMPIMultiDisplays(_start+0x2a)[0x55b71ce1aaca]
[d49abb85a656:06560] *** End of error message ***
--------------------------------------------------------------------------
mpirun noticed that process rank 2 with PID 0 on node d49abb85a656 exited on signal 11 (Segmentation fault).
--------------------------------------------------------------------------

ospMPIMultiDisplays also runs fine despite the errors. These errors appear when you close the program.

Docker Image FROM ubuntu:18.04 gcc (Ubuntu 7.5.0-3ubuntu1/~18.04) 7.5.0 g++ (Ubuntu 7.5.0-3ubuntu1/~18.04) 7.5.0

ghost commented 1 year ago

https://github.com/open-mpi/ompi/issues/4948 discusses about Read -1, expected <someNumber>, errno =1. In short, the issue may have been addressed in a new version of MPI.

I have not tested it yet. I need to look into this problem more close. The docker container has mpirun (Open MPI) 2.1.1.