Your message dated Mon, 11 Apr 2022 08:49:18 +0000
with message-id <e1ndpja-0007bn...@fasolo.debian.org>
and subject line Bug#1008966: fixed in openmpi 4.1.3-2
has caused the Debian Bug report #1008966,
regarding libopenmpi-dev: IO (and UCX) problems causing mpi4py test failures
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
1008966: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1008966
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: libopenmpi-dev
Version: 4.1.3-1
Severity: serious
Justification: debci failure (mpi4py)
Control: affects -1 src:mpi4py
mpi4py is failing IO tests with the new openmpi 4.1.3.
i386 has actually been failing test_io.TestIOSelf for some time, but
now the problem has expanded to amd64 and other arches.
The problem (exposed in mpi4py tests) has been reported at
https://github.com/mpi4py/mpi4py/issues/105
We checked that the same tests are passing with mpich, which I why I'm
filing this bug against openmpi rather than mpi4py
The mpi4py maintainer can reproduce the error (after building on
Fedora without ucx, which was causing other problems) using
OMPI_MCA_io=romio321. He observes that the default OMPIO component
failed to load at runtime, so ROMIO is being used instead.
He provides a backtrace:
$ OMPI_MCA_io=romio321 valgrind -q python test/main.py -q -i test_io -k
TestIOSelf.testIReadIWriteAll -v
[0@kw61149] Python 3.10.4 (/usr/bin/python)
[0@kw61149] numpy 1.21.5 (/usr/lib64/python3.10/site-packages/numpy)
[0@kw61149] MPI 3.1 (Open MPI 4.1.3)
[0@kw61149] mpi4py 4.0.0.dev0
(/home/dalcinl/Devel/mpi4py/build/lib.linux-x86_64-3.10/mpi4py)
testIReadIWriteAll (test_io.TestIOSelf) ... ==3623039== Jump to the invalid
address stated on the next line
==3623039== at 0x0: ???
==3623039== by 0x1DA8E8B3: mca_io_romio_dist_MPI_File_iwrite_all
(iwrite_all.c:58)
==3623039== by 0x1DA8BAF5: mca_io_romio321_file_iwrite_all
(io_romio321_file_write.c:204)
==3623039== by 0x1600D5D5: PMPI_File_iwrite_all (pfile_iwrite_all.c:83)
==3623039== by 0x15DBACA9: PyMPI_File_iwrite_all_c (largecnt.h:2377)
==3623039== by 0x15ECAA01: __pyx_pf_6mpi4py_3MPI_4File_62Iwrite_all
(MPI.c:171765)
==3623039== by 0x15ECA843: __pyx_pw_6mpi4py_3MPI_4File_63Iwrite_all
(MPI.c:171700)
==3623039== by 0x498A50F: method_vectorcall_VARARGS_KEYWORDS
(descrobject.c:344)
==3623039== by 0x497CBA2: UnknownInlinedFun (abstract.h:114)
==3623039== by 0x497CBA2: UnknownInlinedFun (abstract.h:123)
==3623039== by 0x497CBA2: UnknownInlinedFun (ceval.c:5867)
==3623039== by 0x497CBA2: _PyEval_EvalFrameDefault (ceval.c:4198)
==3623039== by 0x497B5FF: UnknownInlinedFun (pycore_ceval.h:46)
==3623039== by 0x497B5FF: _PyEval_Vector (ceval.c:5065)
==3623039== by 0x49918F7: UnknownInlinedFun (call.c:342)
==3623039== by 0x49918F7: UnknownInlinedFun (abstract.h:114)
==3623039== by 0x49918F7: method_vectorcall (classobject.c:53)
==3623039== by 0x497C7C6: UnknownInlinedFun (abstract.h:114)
==3623039== by 0x497C7C6: UnknownInlinedFun (abstract.h:123)
==3623039== by 0x497C7C6: UnknownInlinedFun (ceval.c:5867)
==3623039== by 0x497C7C6: _PyEval_EvalFrameDefault (ceval.c:4213)
==3623039== Address 0x0 is not stack'd, malloc'd or (recently) free'd
==3623039==
[kw61149:3623039] *** Process received signal ***
[kw61149:3623039] Signal: Segmentation fault (11)
[kw61149:3623039] Signal code: Invalid permissions (2)
[kw61149:3623039] Failing at address: (nil)
[kw61149:3623039] [ 0] /lib64/libc.so.6(+0x42750)[0x4bff750]
[kw61149:3623039] *** End of error message ***
Segmentation fault (core dumped)
-- System Information:
Debian Release: bookworm/sid
APT prefers unstable
APT policy: (500, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 5.16.0-6-amd64 (SMP w/8 CPU threads; PREEMPT)
Kernel taint flags: TAINT_OOT_MODULE
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8),
LANGUAGE=en_AU:en
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled
Versions of packages libopenmpi-dev depends on:
ii gfortran [gfortran-mod-15] 4:11.2.0-2
ii gfortran-11 [gfortran-mod-15] 11.2.0-19
ii gfortran-9 [gfortran-mod-15] 9.4.0-5
ii libevent-dev 2.1.12-stable-1
ii libhwloc-dev 2.7.1-1
ii libibverbs-dev 39.0-1+b1
ii libjs-jquery 3.6.0+dfsg+~3.5.13-1
ii libjs-jquery-ui 1.13.1+dfsg-1
ii libopenmpi3 4.1.3-1
ii libpmix-dev 4.1.2-2
ii openmpi-bin 4.1.3-1
ii openmpi-common 4.1.3-1
ii zlib1g-dev 1:1.2.11.dfsg-4
Versions of packages libopenmpi-dev recommends:
ii libcoarrays-openmpi-dev 2.9.3-1
Versions of packages libopenmpi-dev suggests:
pn openmpi-doc <none>
-- no debconf information
--- End Message ---
--- Begin Message ---
Source: openmpi
Source-Version: 4.1.3-2
Done: Alastair McKinstry <mckins...@debian.org>
We believe that the bug you reported is fixed in the latest version of
openmpi, which is due to be installed in the Debian FTP archive.
A summary of the changes between this version and the previous one is
attached.
Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to 1008...@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.
Debian distribution maintenance software
pp.
Alastair McKinstry <mckins...@debian.org> (supplier of updated openmpi package)
(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmas...@ftp-master.debian.org)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
Format: 1.8
Date: Mon, 11 Apr 2022 08:32:17 +0100
Source: openmpi
Architecture: source
Version: 4.1.3-2
Distribution: unstable
Urgency: medium
Maintainer: Alastair McKinstry <mckins...@debian.org>
Changed-By: Alastair McKinstry <mckins...@debian.org>
Closes: 1008966 1009263
Changes:
openmpi (4.1.3-2) unstable; urgency=medium
.
* Update links:
- libmca_common_ompio.so.41.29.2 -> libmca_common_ompio.so.41.29.3
- libmca_common_ucx.so.40.30.1 -> libmca_common_ucx.so.40.30.2
Closes: #1008966
* Fix autopkgtest that was incorrectly silent: Closes: #1009263
Checksums-Sha1:
2d7cb71c9c44aa1c58d2e7a274ff6d32fc095876 2743 openmpi_4.1.3-2.dsc
62b479958c676c8d093ff492133cfdceec9faa2e 68692 openmpi_4.1.3-2.debian.tar.xz
Checksums-Sha256:
1f6eeb3e7ee1c2b85c4fc308b201c67c0e76aefdc50ac4e3b65123b716d0f89e 2743
openmpi_4.1.3-2.dsc
7d2cbeae26c0d18c861370bdbad43912f1a0562c84689abe6d8b1554af537a36 68692
openmpi_4.1.3-2.debian.tar.xz
Files:
ee26b0204f2b31e0501418372fbc9b25 2743 net optional openmpi_4.1.3-2.dsc
f2d11386a896053420d73e4e148317f3 68692 net optional
openmpi_4.1.3-2.debian.tar.xz
-----BEGIN PGP SIGNATURE-----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=hY5M
-----END PGP SIGNATURE-----
--- End Message ---