<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><div>Hi Wesley,</div><div>Hi Pavan, </div><div><br>I will do that. Thanks for all your helps. As I mentioned before, I am still beginner with all this. I will read these links to understand very well how to use MPI_COMM_FREE </div>
<div><br>Thanks again,</div><div>Reem</div><div> </div><div> </div><div> </div><blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
Message: 3<br>
Date: Mon, 8 Jul 2013 12:22:55 -0500<br>
From: Wesley Bland <<a href="mailto:wbland@mcs.anl.gov">wbland@mcs.anl.gov</a>><br>
To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
Subject: Re: [mpich-discuss] mpich on Mac os x<br>
Message-ID: <<a href="mailto:0A10C046-D447-45E0-BD4C-CD6CE7F12D92@mcs.anl.gov">0A10C046-D447-45E0-BD4C-CD6CE7F12D92@mcs.anl.gov</a>><br>
Content-Type: text/plain; charset=us-ascii<br>
<br>
Pavan is probably right that this a problem with the application you're trying to run. You should probably talk to them. However, if you're trying to write an MPI problem, you must understand how MPI_COMM_FREE works. Here's a couple of good links:<br>
<br>
<a href="http://www.mcs.anl.gov/research/projects/mpi/www/www3/MPI_Comm_free.html" target="_blank">http://www.mcs.anl.gov/research/projects/mpi/www/www3/MPI_Comm_free.html</a><br>
<a href="http://mpi.deino.net/mpi_functions/MPI_Comm_free.html" target="_blank">http://mpi.deino.net/mpi_functions/MPI_Comm_free.html</a><br>
<br>
On Jul 8, 2013, at 11:56 AM, Jeff Hammond <<a href="mailto:jeff.science@gmail.com">jeff.science@gmail.com</a>> wrote:<br>
<br>
> Do you understand the different between a Fortran procedure/subroutine<br>
> and a variable value?<br>
><br>
> You cannot set an MPI communicator equal to a subroutine. That makes<br>
> no sense and the compiler is rightfully complaining about it.<br>
><br>
> Jeff<br>
><br>
> On Mon, Jul 8, 2013 at 11:32 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>><br>
>> Hi Wesley,<br>
>><br>
>> How could I do that? as I read in the internet. I have to set:<br>
>> include "mpif.h"<br>
>> Call MPI_COMM_FREE<br>
>><br>
>> I found where I can set that in FLASH code ( I have attached the file) So I<br>
>> set:<br>
>> integer, parameter :: FLASH_COMM = MPI_COMM_FREE<br>
>> but I got an error when I compile the code:<br>
>><br>
>> Included at Driver_abortFlash.F90:34:<br>
>><br>
>> integer, parameter :: FLASH_COMM = MPI_COMM_FREE<br>
>> 1<br>
>> Error: Symbol 'flash_comm' at (1) already has basic type of INTEGER<br>
>> make: *** [Driver_abortFlash.o] Error 1<br>
>><br>
>> the code used MPI_COMM_WORLD.<br>
>> So is there a way to fix that?<br>
>><br>
>> Thanks,<br>
>> Reem<br>
>><br>
>><br>
>><br>
>>> Date: Mon, 8 Jul 2013 10:25:38 -0500<br>
>>> From: Wesley Bland <<a href="mailto:wbland@mcs.anl.gov">wbland@mcs.anl.gov</a>><br>
>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>> Subject: Re: [mpich-discuss] discuss Digest, Vol 9, Issue 13<br>
>>> Message-ID: <<a href="mailto:9A8EEBB2-72BB-429B-9368-150983B1C283@mcs.anl.gov">9A8EEBB2-72BB-429B-9368-150983B1C283@mcs.anl.gov</a>><br>
>>> Content-Type: text/plain; charset="iso-8859-1"<br>
>>><br>
>>> That's the correct way to set that environment variable, but I would have<br>
>>> to agree with Pavan here. If you're running into the cap on context id's<br>
>>> (especially with so few processes) it would seem that you're creating a lot<br>
>>> of communicators that you probably don't need. Are you sure that you're<br>
>>> freeing them correctly after use?<br>
>>><br>
>>> Wesley<br>
>>><br>
>>> On Jul 8, 2013, at 10:05 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>><br>
>>>> Hi Wesley.<br>
>>>> I wrote the following:<br>
>>>> mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4<br>
>>>> but the error still the same. Did I do it in the wrong way??<br>
>>>><br>
>>>> Thanks,<br>
>>>> Reem<br>
>>>><br>
>>>> Message: 5<br>
>>>> Date: Mon, 8 Jul 2013 08:14:48 -0500<br>
>>>> From: Wesley Bland <<a href="mailto:wbland@mcs.anl.gov">wbland@mcs.anl.gov</a>><br>
>>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>> Subject: Re: [mpich-discuss] mpich on Mac os x<br>
>>>> Message-ID: <<a href="mailto:8DC984B2-4E4B-4BFE-806E-203463A7A4E4@mcs.anl.gov">8DC984B2-4E4B-4BFE-806E-203463A7A4E4@mcs.anl.gov</a>><br>
>>>> Content-Type: text/plain; charset=iso-8859-1<br>
>>>><br>
>>>> It seems that you're creating more communicators than MPICH can handle.<br>
>>>> It's possible that you might be able to get around this by setting the<br>
>>>> environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller than<br>
>>>> its default (which is 2). That frees up a few more communicators, but there<br>
>>>> is a pathological case where even with fewer communicators than the max,<br>
>>>> MPICH won't be able to agree on a new communicator id when needed. Try<br>
>>>> changing that environment variable and see if that fixes things.<br>
>>>><br>
>>>> Wesley<br>
>>>><br>
>>>> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>>><br>
>>>>> Hi all,<br>
>>>>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code.<br>
>>>>> It works fine in the beginning of the run and then I got the following<br>
>>>>> error:<br>
>>>>><br>
>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>> MPI_Comm_create(600).................:<br>
>>>>> MPI_Comm_create(comm=0x84000002, group=0xc8001349, new_comm=0x7fff606a8614)<br>
>>>>> failed<br>
>>>>> MPI_Comm_create(577).................:<br>
>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>> because of fragmentation (169/2048 free on this process; ignore_id=0)<br>
>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>> MPI_Comm_create(600).................:<br>
>>>>> MPI_Comm_create(comm=0x84000002, group=0xc80012b6, new_comm=0x7fff670cc614)<br>
>>>>> failed<br>
>>>>> MPI_Comm_create(577).................:<br>
>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>> because of fragmentation (316/2048 free on this process; ignore_id=0)<br>
>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>> MPI_Comm_create(600).................:<br>
>>>>> MPI_Comm_create(comm=0x84000004, group=0xc800000e, new_comm=0x7fff629d5614)<br>
>>>>> failed<br>
>>>>> MPI_Comm_create(577).................:<br>
>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>> because of fragmentation (2020/2048 free on this process; ignore_id=0)<br>
>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>> MPI_Comm_create(600).................:<br>
>>>>> MPI_Comm_create(comm=0x84000002, group=0xc8000020, new_comm=0x7fff639ae614)<br>
>>>>> failed<br>
>>>>> MPI_Comm_create(577).................:<br>
>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>> because of fragmentation (2002/2048 free on this process; ignore_id=0<br>
>>>>><br>
>>>>> Is there a way to fix that ?<br>
>>>>><br>
>>>>> Thanks,<br>
>>>>> Reem<br>
>><br>
>><br>
>>><br>
>>> On Mon, Jul 8, 2013 at 9:17 AM, Pavan Balaji <<a href="mailto:balaji@mcs.anl.gov">balaji@mcs.anl.gov</a>> wrote:<br>
>>>><br>
>>>> Also make sure you are freeing unused communicators.<br>
>>>><br>
>>>> -- Pavan<br>
>>>><br>
>>>> On 07/08/2013 08:14 AM, Wesley Bland wrote:<br>
>>>>><br>
>>>>> It seems that you're creating more communicators than MPICH can handle.<br>
>>>>> It's possible that you might be able to get around this by setting the<br>
>>>>> environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller<br>
>>>>> than<br>
>>>>> its default (which is 2). That frees up a few more communicators, but<br>
>>>>> there<br>
>>>>> is a pathological case where even with fewer communicators than the<br>
>>>>> max,<br>
>>>>> MPICH won't be able to agree on a new communicator id when needed. Try<br>
>>>>> changing that environment variable and see if that fixes things.<br>
>>>>><br>
>>>>> Wesley<br>
>>>>><br>
>>>>> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>>>><br>
>>>>>> Hi all,<br>
>>>>>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code.<br>
>>>>>> It<br>
>>>>>> works fine in the beginning of the run and then I got the following<br>
>>>>>> error:<br>
>>>>>><br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>>>> group=0xc8001349, new_comm=0x7fff606a8614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (169/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>>>> group=0xc80012b6, new_comm=0x7fff670cc614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (316/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000004,<br>
>>>>>> group=0xc800000e, new_comm=0x7fff629d5614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (2020/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>>>> group=0xc8000020, new_comm=0x7fff639ae614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (2002/2048 free on this process; ignore_id=0<br>
>>>>>><br>
>>>>>> Is there a way to fix that ?<br>
>>>>>><br>
>>>>>> Thanks,<br>
>>>>>> Reem<br>
>>>>>> _______________________________________________<br>
>>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>>> To manage subscription options or unsubscribe:<br>
>>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>>><br>
>>>>> _______________________________________________<br>
>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>> To manage subscription options or unsubscribe:<br>
>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>><br>
>>>> --<br>
>>>> Pavan Balaji<br>
>>>> <a href="http://www.mcs.anl.gov/~balaji" target="_blank">http://www.mcs.anl.gov/~balaji</a><br>
>>>> _______________________________________________<br>
>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>> To manage subscription options or unsubscribe:<br>
>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> Jeff Hammond<br>
>>> <a href="mailto:jeff.science@gmail.com">jeff.science@gmail.com</a><br>
>>><br>
>>><br>
>>> ------------------------------<br>
>>><br>
>>> Message: 3<br>
>>> Date: Mon, 8 Jul 2013 16:39:03 +0100<br>
>>> From: Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>><br>
>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>> Subject: [mpich-discuss] mpich on Mac os x<br>
>>> Message-ID:<br>
>>><br>
>>> <<a href="mailto:CANKsntEJR0b_TA-tbdBO3fOV2O8PkwTG82a718qZPd3FnEBT3w@mail.gmail.com">CANKsntEJR0b_TA-tbdBO3fOV2O8PkwTG82a718qZPd3FnEBT3w@mail.gmail.com</a>><br>
>>> Content-Type: text/plain; charset="iso-8859-1"<br>
>>><br>
>>> Hi Pavan,<br>
>>><br>
>>> Could you explain more what do you mean by that? I am sorry I am still<br>
>>> beginner in MPICH.<br>
>>><br>
>>> Hi Wesley,<br>
>>> apologize about my first confused reply I have forgotten to change the<br>
>>> subject. However, as I mentioned I wrote the following line:<br>
>>><br>
>>> mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4<br>
>>><br>
>>> I am wondering if I made any mistake here as the problem is still the same<br>
>>><br>
>>> Thanks,<br>
>>> Reem<br>
>>><br>
>>><br>
>>><br>
>>> Message: 1<br>
>>>> Date: Mon, 08 Jul 2013 09:17:00 -0500<br>
>>>> From: Pavan Balaji <<a href="mailto:balaji@mcs.anl.gov">balaji@mcs.anl.gov</a>><br>
>>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>> Subject: Re: [mpich-discuss] mpich on Mac os x<br>
>>>> Message-ID: <<a href="mailto:51DAC9DC.5000502@mcs.anl.gov">51DAC9DC.5000502@mcs.anl.gov</a>><br>
>>>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
>>>><br>
>>>><br>
>>>> Also make sure you are freeing unused communicators.<br>
>>>><br>
>>>> -- Pavan<br>
>>>><br>
>>>> On 07/08/2013 08:14 AM, Wesley Bland wrote:<br>
>>>>> It seems that you're creating more communicators than MPICH can<br>
>>>>> handle.<br>
>>>> It's possible that you might be able to get around this by setting the<br>
>>>> environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller<br>
>>>> than<br>
>>>> its default (which is 2). That frees up a few more communicators, but<br>
>>>> there<br>
>>>> is a pathological case where even with fewer communicators than the max,<br>
>>>> MPICH won't be able to agree on a new communicator id when needed. Try<br>
>>>> changing that environment variable and see if that fixes things.<br>
>>>>><br>
>>>>> Wesley<br>
>>>>><br>
>>>>> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>>>><br>
>>>>>> Hi all,<br>
>>>>>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code.<br>
>>>>>> It<br>
>>>> works fine in the beginning of the run and then I got the following<br>
>>>> error:<br>
>>>>>><br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>> group=0xc8001349, new_comm=0x7fff606a8614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>> because of fragmentation (169/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>> group=0xc80012b6, new_comm=0x7fff670cc614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>> because of fragmentation (316/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000004,<br>
>>>> group=0xc800000e, new_comm=0x7fff629d5614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>> because of fragmentation (2020/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002,<br>
>>>> group=0xc8000020, new_comm=0x7fff639ae614) failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>> because of fragmentation (2002/2048 free on this process; ignore_id=0<br>
>>>>>><br>
>>>>>> Is there a way to fix that ?<br>
>>>>>><br>
>>>>>> Thanks,<br>
>>>>>> Reem<br>
>>>>>> _______________________________________________<br>
>>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>>> To manage subscription options or unsubscribe:<br>
>>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>>> _______________________________________________<br>
>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>> To manage subscription options or unsubscribe:<br>
>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>><br>
>>>> --<br>
>>>> Pavan Balaji<br>
>>>> <a href="http://www.mcs.anl.gov/~balaji" target="_blank">http://www.mcs.anl.gov/~balaji</a><br>
>>>><br>
>>> -------------- next part --------------<br>
>>> An HTML attachment was scrubbed...<br>
>>> URL:<br>
>>> <<a href="http://lists.mpich.org/pipermail/discuss/attachments/20130708/e3a9be0d/attachment-0001.html" target="_blank">http://lists.mpich.org/pipermail/discuss/attachments/20130708/e3a9be0d/attachment-0001.html</a>><br>
>>><br>
>>> ------------------------------<br>
>>><br>
>>> Message: 4<br>
>>> Date: Mon, 8 Jul 2013 10:53:48 -0500<br>
>>> From: Wesley Bland <<a href="mailto:wbland@mcs.anl.gov">wbland@mcs.anl.gov</a>><br>
>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>> Subject: Re: [mpich-discuss] mpich on Mac os x<br>
>>> Message-ID: <<a href="mailto:CEEC8B51-FFDF-40FE-BE56-E25F9C05BB27@mcs.anl.gov">CEEC8B51-FFDF-40FE-BE56-E25F9C05BB27@mcs.anl.gov</a>><br>
>>> Content-Type: text/plain; charset="iso-8859-1"<br>
>>><br>
>>> On Jul 8, 2013, at 10:39 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>><br>
>>>> Hi Pavan,<br>
>>>><br>
>>>> Could you explain more what do you mean by that? I am sorry I am still<br>
>>>> beginner in MPICH.<br>
>>><br>
>>> This isn't specific to MPICH. This would be the case with any MPI<br>
>>> implementation. What he means is that you need to make sure that when you<br>
>>> are done using a communicator, you call MPI_Comm_free(communicator) so MPICH<br>
>>> can free the internal resources used by the communicator. If you never free<br>
>>> your communicators, then you will eventually run out. There is a limited<br>
>>> number of communicators available internally.<br>
>>><br>
>>>><br>
>>>> Hi Wesley,<br>
>>>> apologize about my first confused reply I have forgotten to change the<br>
>>>> subject. However, as I mentioned I wrote the following line:<br>
>>>><br>
>>>> mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4<br>
>>>><br>
>>>> I am wondering if I made any mistake here as the problem is still the<br>
>>>> same<br>
>>><br>
>>> Yes, that's the correct way to use it, so that means your problem is<br>
>>> probably what we described above.<br>
>>><br>
>>>><br>
>>>> Thanks,<br>
>>>> Reem<br>
>>>><br>
>>>><br>
>>>><br>
>>>> Message: 1<br>
>>>> Date: Mon, 08 Jul 2013 09:17:00 -0500<br>
>>>> From: Pavan Balaji <<a href="mailto:balaji@mcs.anl.gov">balaji@mcs.anl.gov</a>><br>
>>>> To: <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>> Subject: Re: [mpich-discuss] mpich on Mac os x<br>
>>>> Message-ID: <<a href="mailto:51DAC9DC.5000502@mcs.anl.gov">51DAC9DC.5000502@mcs.anl.gov</a>><br>
>>>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
>>>><br>
>>>><br>
>>>> Also make sure you are freeing unused communicators.<br>
>>>><br>
>>>> -- Pavan<br>
>>>><br>
>>>> On 07/08/2013 08:14 AM, Wesley Bland wrote:<br>
>>>>> It seems that you're creating more communicators than MPICH can<br>
>>>>> handle. It's possible that you might be able to get around this by setting<br>
>>>>> the environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller<br>
>>>>> than its default (which is 2). That frees up a few more communicators, but<br>
>>>>> there is a pathological case where even with fewer communicators than the<br>
>>>>> max, MPICH won't be able to agree on a new communicator id when needed. Try<br>
>>>>> changing that environment variable and see if that fixes things.<br>
>>>>><br>
>>>>> Wesley<br>
>>>>><br>
>>>>> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <<a href="mailto:raba500@york.ac.uk">raba500@york.ac.uk</a>> wrote:<br>
>>>>><br>
>>>>>> Hi all,<br>
>>>>>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code.<br>
>>>>>> It works fine in the beginning of the run and then I got the following<br>
>>>>>> error:<br>
>>>>>><br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002, group=0xc8001349, new_comm=0x7fff606a8614)<br>
>>>>>> failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (169/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002, group=0xc80012b6, new_comm=0x7fff670cc614)<br>
>>>>>> failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (316/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000004, group=0xc800000e, new_comm=0x7fff629d5614)<br>
>>>>>> failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (2020/2048 free on this process; ignore_id=0)<br>
>>>>>> Fatal error in MPI_Comm_create: Other MPI error, error stack:<br>
>>>>>> MPI_Comm_create(600).................:<br>
>>>>>> MPI_Comm_create(comm=0x84000002, group=0xc8000020, new_comm=0x7fff639ae614)<br>
>>>>>> failed<br>
>>>>>> MPI_Comm_create(577).................:<br>
>>>>>> MPIR_Comm_create_intra(241)..........:<br>
>>>>>> MPIR_Get_contextid(799)..............:<br>
>>>>>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID<br>
>>>>>> because of fragmentation (2002/2048 free on this process; ignore_id=0<br>
>>>>>><br>
>>>>>> Is there a way to fix that ?<br>
>>>>>><br>
>>>>>> Thanks,<br>
>>>>>> Reem<br>
>>>>>> _______________________________________________<br>
>>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>>> To manage subscription options or unsubscribe:<br>
>>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>>> _______________________________________________<br>
>>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>>> To manage subscription options or unsubscribe:<br>
>>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>>>><br>
>>>><br>
>>>> --<br>
>>>> Pavan Balaji<br>
>>>> <a href="http://www.mcs.anl.gov/~balaji" target="_blank">http://www.mcs.anl.gov/~balaji</a><br>
>>>> _______________________________________________<br>
>>>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>>> To manage subscription options or unsubscribe:<br>
>>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>><br>
>>> -------------- next part --------------<br>
>>> An HTML attachment was scrubbed...<br>
>>> URL:<br>
>>> <<a href="http://lists.mpich.org/pipermail/discuss/attachments/20130708/5b36571f/attachment.html" target="_blank">http://lists.mpich.org/pipermail/discuss/attachments/20130708/5b36571f/attachment.html</a>><br>
>>><br>
>>> ------------------------------<br>
>>><br>
>>> _______________________________________________<br>
>>> discuss mailing list<br>
>>> <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
>>><br>
>>> End of discuss Digest, Vol 9, Issue 15<br>
>>> **************************************<br>
>><br>
>><br>
>><br>
>> _______________________________________________<br>
>> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
>> To manage subscription options or unsubscribe:<br>
>> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
><br>
><br>
><br>
> --<br>
> Jeff Hammond<br>
> <a href="mailto:jeff.science@gmail.com">jeff.science@gmail.com</a><br>
> _______________________________________________<br>
> discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
> To manage subscription options or unsubscribe:<br>
> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
<br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
discuss mailing list<br>
<a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
End of discuss Digest, Vol 9, Issue 19<br>
**************************************<br>
</blockquote></div><br></div></div>