Debian Bug report logs - #814451
openmpi 1.10 has stderr output by default, breaking autopkg tests

Reply or subscribe to this bug.

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Matthias Klose <doko@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: openmpi 1.10 has stderr output by default, breaking autopkg tests

Date: Thu, 11 Feb 2016 17:56:47 +0100

Package: src:openmpi
Version: 1.10.2-7
Severity: important

Seen at least for lammps, esys-particle, liggghts:

adt-run [15:36:24]: test dipole: -----------------------]
adt-run [15:36:24]: test dipole:  - - - - - - - - - - results - - - - - - - - - -
dipole               FAIL stderr: [adt:03802] 1 more process has sent help 
message help-mpi-btl-base.txt / btl:no-nics
adt-run [15:36:25]: test dipole:  - - - - - - - - - - stderr - - - - - - - - - -
[adt:03802] 1 more process has sent help message help-mpi-btl-base.txt / btl:no-nics
[adt:03802] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / 
error messages
adt-run [15:36:25]: test ellipse: preparing testbed

so maybe disable that?  Or else these tests should expect stderr output. In this 
case, please clone/reassign the issue.

Message #10 received at 814451@bugs.debian.org (full text, mbox, reply):

From: Alastair McKinstry <mckinstry@debian.org>

To: Matthias Klose <doko@debian.org>, 814451@bugs.debian.org

Subject: openmpi: stderr output is an error message

Date: Mon, 22 Feb 2016 17:00:33 +0000

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Hi,

Upstream considers this output to be a proper error message, and hence
reports
it to stderr (as openmpi will not run).
What platform did you see this message on?

best regards
Alastair

- -- 
Alastair McKinstry, <alastair@sceal.ie>, <mckinstry@debian.org>,
https://diaspora.sceal.ie/u/amckinstry
Misentropy: doubting that the Universe is becoming more disordered.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQIcBAEBCAAGBQJWyz6xAAoJEMvmu05dmtOl2j4P/jD4zbGUYwTznBmmo+3Az5Tk
D9la6rZVYedcQoz4fx3PNLsQC/a64Xulyt9Ug5V9EazOh9MDoTmA8itS6ynx35FQ
9knvuqAplPkR0hfC6VQU2eXpVxfxm/uA80rdeyy3gy/uxi3anS2vCMJhxbt21cqe
RHNdGMiUEgqzA5ZreS3gFFY1w38F+e/vikw9iqojXpbAmGI+eR5f3MW7q4ZlBVsH
WO5Dk/Q7tDyuoO+DqY/f/F3n0+GUSMsQc3d2uJHSIZLVwh0eNgFc05xBT1oEWxTZ
6EbR6+Mzqe0KceNgErTVzqC7LVIct+vu14vVyg1wuzOQTWXcfQ6azjh67mzkb90l
FWNgxu3UtjlsaeGKUzmGcjd2q5tmIKi4xpnh+FoYRrNcNpfbDMkKp6ey43mX0C63
t4sl86Jfu2vfMgjOd9PRSbLTt37BKzvn0BwRz4napEsFZplyOVSE/whgdOGNOvK6
rNlFKPG421/iIHTbouENE1Rsfqs4uT/9JVDPsLWhJwsTfSxlcNzTv3hN8Wf6aMhm
9b9N0ihU7HmKGx4unlLZ+7YFEunBZWf2hJlrQ/FGZ5nDQd5EsbF2fuXbsnYnh0pm
32+cfWcaG2T6643oVnfkzgChdM0fDViEjI748WVI8BMXe2xeRV58sV6JUEtt8SgD
XNxaqVjzdQ7LTGDsBeX3
=7fCz
-----END PGP SIGNATURE-----

Message #15 received at 814451@bugs.debian.org (full text, mbox, reply):

From: Matthias Klose <doko@debian.org>

To: Alastair McKinstry <mckinstry@debian.org>, 814451@bugs.debian.org, Martin Pitt <mpitt@debian.org>

Subject: Re: openmpi: stderr output is an error message

Date: Mon, 22 Feb 2016 18:25:17 +0100

On 22.02.2016 18:00, Alastair McKinstry wrote:
> Upstream considers this output to be a proper error message, and hence
> reports
> it to stderr (as openmpi will not run).
> What platform did you see this message on?

afaicr, on all, where I was running the autopkg tests on the Ubuntu autopkg test 
infrastructure. Now looking at the Debian infrastructure, I can't see these 
tests failing there ... Martin, any idea?

Matthias

Message #20 received at 814451@bugs.debian.org (full text, mbox, reply):

From: Alastair McKinstry <mckinstry@debian.org>

To: Matthias Klose <doko@debian.org>, 814451@bugs.debian.org, Martin Pitt <mpitt@debian.org>

Subject: Re: Bug#814451: openmpi: stderr output is an error message

Date: Wed, 24 Feb 2016 12:32:57 +0000

[Message part 1 (text/plain, inline)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Hi,


Gilles Gouaillardet points out that this error would occur if, eg.
openmpi detects inifiniband
but it can't be used.

as a workaround, you can disable the openib btl
you can
|mpirun --mca btl ^openib ...|
or you can
|export OMPI_MCA_btl=^openib|
and then invoke mpirun as usual

regards
Alastair



On 22/02/2016 17:25, Matthias Klose wrote:
> On 22.02.2016 18:00, Alastair McKinstry wrote:
>> Upstream considers this output to be a proper error message, and hence
>> reports
>> it to stderr (as openmpi will not run).
>> What platform did you see this message on?
>
> afaicr, on all, where I was running the autopkg tests on the Ubuntu
autopkg test infrastructure. Now looking at the Debian infrastructure, I
can't see these tests failing there ... Martin, any idea?
>
> Matthias

- -- 
Alastair McKinstry, <alastair@sceal.ie>, <mckinstry@debian.org>,
https://diaspora.sceal.ie/u/amckinstry
Misentropy: doubting that the Universe is becoming more disordered.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=TlHo
-----END PGP SIGNATURE-----

[Message part 2 (text/html, inline)]

Message #25 received at 814451@bugs.debian.org (full text, mbox, reply):

From: Gilles Gouaillardet <gilles.gouaillardet@gmail.com>

To: 814451@bugs.debian.org

Subject: openmpi: stderr output is an error message

Date: Thu, 25 Feb 2016 14:11:42 +0900

Folks,

i was able to reproduce this behaviour.

here is what the openib (e.g. infiniband) btl does :
- if running on Linux, and there is no /sys/class/infiniband, then the
btl disqualifies itself (see ompi_common_verbs_check_basics()(
- if the btl did not previously disqualified itself, and there is no
infiniband port available, then a warning message is issued (we
consider there is likely something wrong)

i am able to reproduce this scenario on a VM with no infiniband
hardware by loading the mlx4_core kernel module.

can you confirm you are running into this very same scenario ?
is the mlx4_core.ko or similar module loaded ?
if yes, is there any reason for it to be loaded ?
is the __linux__ macro defined on debian ?
does the /sys/class/infiniband directory exists ?
is this directory empty ?

from an OpenMPI point of view, it could be possible to disqualify the
openib btl with no warning message if /sys/class/infiniband is empty,
would that be helpful ?

Cheers,

Gilles

Send a report that this bug log contains spam.