I think this is fixed by Nathan's PR:
https://github.com/open-mpi/ompi/pull/653
It's waiting for George's review -- George?
> On Jun 24, 2015, at 7:14 AM, Howard Pritchard <hpprit...@gmail.com> wrote:
>
> Hi Folks,
>
> I'm seeing some MTT failures from last nite in the form of run failures
> and/or timeouts.
> Is anyone else seeing these?
>
> On the ibm dataplex system I'm seeing these kinds of assertion failures in
> ob1:
>
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
> [c1476:07379] *** Process received signal ***
> [c1476:07379] Signal: Aborted (6)
> [c1476:07379] Signal code: (-6)
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
> [c1475:18949] *** Process received signal ***
> [c1475:18949] Signal: Aborted (6)
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
> [c1476:07375] *** Process received signal ***
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
> [c1475:18951] *** Process received signal ***
> [c1475:18951] Signal: Aborted (6)
> [c1477:19137] Signal: Aborted (6)
> [c1477:19137] Signal code: (-6)
> [c1476:07375] Signal: Aborted (6)
> c_ring: pml_ob1_component.c:308: mca_pml_ob1_component_fini: Assertion
> `((0xdeafbeedULL << 32) +
> 0xdeafbeedULL) == ((opal_object_t *) (mca_pml_ob1_recvreq))->obj_magic_id'
> failed.
>
> ---------- Forwarded message ----------
> From: Howard Pritchard <h...@c1479.nersc.gov>
> Date: 2015-06-24 7:04 GMT-06:00
> Subject: MTT test has completed, status: failed
> To: hpprit...@gmail.com
>
>
> Subject: MTT test has completed, status: failed
> hostname: c1479
> uname: Linux c1479 2.6.32-358.18.1.el6.nersc.x86_64 #1 SMP Wed Aug 28
> 02:17:42 PDT 2013 x86_64 x86_64 x86_64 GNU/Linux
> who am i:
>
> +----------+---------+-------------------+----------+------+------+----------+------+--------------------------------------------------------------------------+
> | Phase | Section | MPI Version | Duration | Pass | Fail | Time out
> | Skip | Detailed report
> |
> +----------+---------+-------------------+----------+------+------+----------+------+--------------------------------------------------------------------------+
> | Test Run | trivial | dev-1936-gdb3c59b | 00:39 | 3 | 3 |
> | |
> Test_Run-trivial-ompi-nightly-master-dev-1936-gdb3c59b-pgi_warnings.html |
> +----------+---------+-------------------+----------+------+------+----------+------+--------------------------------------------------------------------------+
>
>
> Total Tests: 6
> Total Failures: 3
> Total Passed: 3
> Total Duration: 39 secs. (00:39)
>
>
>
> Test Scratch Directory is /global/homes/h/hpp/mtt_carver_tmp
>
> _______________________________________________
> devel mailing list
> de...@open-mpi.org
> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> Link to this post:
> http://www.open-mpi.org/community/lists/devel/2015/06/17525.php
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
