February 2014 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] turning off MPI abort messages
by Jim Dinan 21 Feb '14

21 Feb '14

A little more detail -- you're actually getting messages from two sources: (1) the MPICH library ("application called MPI_Abort...") and (2) the job launcher ("BAD TERMINATION..."). You can eliminate the messages from the job launcher by providing an error code of 0 in MPI_Abort. ~Jim. On Fri, Feb 21, 2014 at 1:19 PM, Jeff Hammond <jeff.science(a)gmail.com>wrote: > >> Just configure MPICH such that snprintf isn't discovered by configure > >> and you won't see these messages. > >> > >> The other solution is to fix PETSc so that people can't crash it so > easily ;-) > > > > Here we go again. It is not CRASHING; it has detected an error > conditioning and trying to appropriately and cleanly terminate. The reason > it needs to use MPI_Abort() is that often detecting error conditions is not > a uniformly collective thing. > > > > Printing a suitable error message and ending is not crashing. But > with all the badly formatted "error messages" printed by MPICH I can not > control at the end it looks like it is crashing. > > You're returning a non-zero exit code, which I consider crashing. I > apologize if this definition disagrees with yours. If this is just > gentle cleanup, why not exit with code=0 as Jim suggested already? > > Jeff > > >> On Thu, Feb 20, 2014 at 3:19 PM, Jim Dinan <james.dinan(a)gmail.com> > wrote: > >>> If you can find a way to call MPI_Finalize instead, you will portably > >>> eliminate these messages. > >>> > >>> A lesser solution would be to provide an error code of 0 (or > MPI_SUCCESS) to > >>> MPI_Abort, e.g. MPI_Comm_abort(MPI_COMM_WORLD, MPI_SUCCESS). This > would > >>> eliminate the error message that you are getting from the job launcher. > >>> MPICH could be modified to be quiet about the abort when the > application > >>> aborts with an error code of MPI_SUCCESS. > >>> > >>> ~Jim. > >>> > >>> > >>> On Thu, Feb 20, 2014 at 12:33 PM, Barry Smith <bsmith(a)mcs.anl.gov> > wrote: > >>>> > >>>> > >>>> Is there any way to turn off MPICH (and others) printing messages > about > >>>> MPI_Abort? We have already prepared and presented useful error > messages to > >>>> the user about the situation and would like to avoid having these > additional > >>>> messages printed (that often make the situation look worse than it is) > >>>> > >>>> Thanks > >>>> > >>>> Barry > >>>> > >>>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 > >>>> [cli_0]: aborting job: > >>>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 > >>>> > >>>> > >>>> > ==================================================================mailto: > discuss(a)mpich.org================= > >>>> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES > >>>> = EXIT CODE: 56 > >>>> = CLEANING UP REMAINING PROCESSES > >>>> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES > >>>> > >>>> > =================================================================================== > >>>> > >>>> > >>>> > >>>> > >>>> _______________________________________________ > >>>> discuss mailing list discuss(a)mpich.org > >>>> To manage subscription options or unsubscribe: > >>>> https://lists.mpich.org/mailman/listinfo/discuss > >>> > >>> > >>> > >>> _______________________________________________ > >>> discuss mailing list discuss(a)mpich.org > >>> To manage subscription options or unsubscribe: > >>> https://lists.mpich.org/mailman/listinfo/discuss > >> > >> > >> > >> -- > >> Jeff Hammond > >> jeff.science(a)gmail.com > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Re: [mpich-discuss] turning off MPI abort messages
by Jeff Hammond 21 Feb '14

21 Feb '14

>> Just configure MPICH such that snprintf isn't discovered by configure >> and you won't see these messages. >> >> The other solution is to fix PETSc so that people can't crash it so easily ;-) > > Here we go again. It is not CRASHING; it has detected an error conditioning and trying to appropriately and cleanly terminate. The reason it needs to use MPI_Abort() is that often detecting error conditions is not a uniformly collective thing. > > Printing a suitable error message and ending is not crashing. But with all the badly formatted “error messages” printed by MPICH I can not control at the end it looks like it is crashing. You're returning a non-zero exit code, which I consider crashing. I apologize if this definition disagrees with yours. If this is just gentle cleanup, why not exit with code=0 as Jim suggested already? Jeff >> On Thu, Feb 20, 2014 at 3:19 PM, Jim Dinan <james.dinan(a)gmail.com> wrote: >>> If you can find a way to call MPI_Finalize instead, you will portably >>> eliminate these messages. >>> >>> A lesser solution would be to provide an error code of 0 (or MPI_SUCCESS) to >>> MPI_Abort, e.g. MPI_Comm_abort(MPI_COMM_WORLD, MPI_SUCCESS). This would >>> eliminate the error message that you are getting from the job launcher. >>> MPICH could be modified to be quiet about the abort when the application >>> aborts with an error code of MPI_SUCCESS. >>> >>> ~Jim. >>> >>> >>> On Thu, Feb 20, 2014 at 12:33 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: >>>> >>>> >>>> Is there any way to turn off MPICH (and others) printing messages about >>>> MPI_Abort? We have already prepared and presented useful error messages to >>>> the user about the situation and would like to avoid having these additional >>>> messages printed (that often make the situation look worse than it is) >>>> >>>> Thanks >>>> >>>> Barry >>>> >>>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >>>> [cli_0]: aborting job: >>>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >>>> >>>> >>>> ==================================================================mailto:[email protected]================= >>>> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES >>>> = EXIT CODE: 56 >>>> = CLEANING UP REMAINING PROCESSES >>>> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES >>>> >>>> =================================================================================== >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> discuss mailing list discuss(a)mpich.org >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/discuss >>> >>> >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> >> >> -- >> Jeff Hammond >> jeff.science(a)gmail.com >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Balaji, Pavan 20 Feb '14

20 Feb '14

It’s an option for hydra’s configure. But if you pass it to the top-level configure, it’ll pass it downstream to hydra’s configure. — Pavan On 2/20/14, 9:08 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> wrote: >Thanks Pavan. Although, one thing I noticed is that the configure script >help doesn't list --with-pbs ! >Guess the help info is not updated? > > >On Feb 21, 2014, at 2:36 AM, Balaji, Pavan wrote: > >> >> If Torque TM libraries are found at configure time, they’ll be used. If >> not, ssh will be used. You can pass --with-pbs to configure to help it >> find the TM libraries. >> >> — Pavan >> >> On 2/20/14, 7:08 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> >> wrote: >> >>> Hi Pavan, >>> >>> Thanks for your response. For the hosts that are not a part of the >>> original PBS job, will mpich try to spawn jobs there through torque/pbs >>> or will the spawn happen through ssh? >>> >>> Best, >>> Suraj >>> >>> On Feb 21, 2014, at 1:33 AM, Balaji, Pavan wrote: >>> >>>> >>>> I thought I responded to this email (it’s in my sent folder), but it >>>> never >>>> showed up on the mailing list. Here it is again: >>>> >>>> Hello, >>>> >>>> Yes, you can use the info argument “hosts” to specify nodes that are >>>> not a >>>> part of the original PBS job. >>>> >>>> — Pavan >>>> >>>> >>>> On 2/20/14, 4:36 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> >>>> wrote: >>>> >>>>> Hi, >>>>> >>>>> Thats an extension we have made. We hope to release it soon! >>>>> >>>>> Best, >>>>> Suraj >>>>> >>>>> >>>>> On Feb 20, 2014, at 10:42 PM, Reuti wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: >>>>>> >>>>>>> I want use MPICH under the Torque RMS. I can see that mpiexec will >>>>>>> automatically take the nodefile from $PBS_NODEFILE. >>>>>>> I am working on dynamically adding nodes to an existing job in >>>>>>> torque. >>>>>>> So there might be a point where the job actually has more number of >>>>>>> nodes than what was listed in $PBS_NODEFILE. >>>>>> >>>>>> How do you tell Torque that it should increase the number of granted >>>>>> slots/nodes for a particular already running job? >>>>>> >>>>>> -- Reuti >>>>>> >>>>>> >>>>>>> Is it possible to do an MPI_Comm_spawn on these new nodes (which is >>>>>>> not listed in PBS_NODEFILE but is actually a part of job)? For >>>>>>> example >>>>>>> using some info argument such as add-host or something? Or if it >>>>>>> entails that the PBS_NODEFILE must be changed with the newly added >>>>>>> nodes, I can do that too. >>>>>>> But the basic question is, can MPICH spawn processes on these newly >>>>>>> allocated nodes for the same job through some mechanism? >>>>>>> >>>>>>> Best, >>>>>>> Suraj >>>>>>> _______________________________________________ >>>>>>> discuss mailing list discuss(a)mpich.org >>>>>>> To manage subscription options or unsubscribe: >>>>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>>>> >>>>>> _______________________________________________ >>>>>> discuss mailing list discuss(a)mpich.org >>>>>> To manage subscription options or unsubscribe: >>>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>>> >>>>> _______________________________________________ >>>>> discuss mailing list discuss(a)mpich.org >>>>> To manage subscription options or unsubscribe: >>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>> >>>> _______________________________________________ >>>> discuss mailing list discuss(a)mpich.org >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/discuss >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > >_______________________________________________ >discuss mailing list discuss(a)mpich.org >To manage subscription options or unsubscribe: >https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Suraj Prabhakaran 20 Feb '14

20 Feb '14

Thanks Pavan. Although, one thing I noticed is that the configure script help doesn't list --with-pbs ! Guess the help info is not updated? On Feb 21, 2014, at 2:36 AM, Balaji, Pavan wrote: > > If Torque TM libraries are found at configure time, they’ll be used. If > not, ssh will be used. You can pass --with-pbs to configure to help it > find the TM libraries. > > — Pavan > > On 2/20/14, 7:08 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> > wrote: > >> Hi Pavan, >> >> Thanks for your response. For the hosts that are not a part of the >> original PBS job, will mpich try to spawn jobs there through torque/pbs >> or will the spawn happen through ssh? >> >> Best, >> Suraj >> >> On Feb 21, 2014, at 1:33 AM, Balaji, Pavan wrote: >> >>> >>> I thought I responded to this email (it’s in my sent folder), but it >>> never >>> showed up on the mailing list. Here it is again: >>> >>> Hello, >>> >>> Yes, you can use the info argument “hosts” to specify nodes that are >>> not a >>> part of the original PBS job. >>> >>> — Pavan >>> >>> >>> On 2/20/14, 4:36 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> >>> wrote: >>> >>>> Hi, >>>> >>>> Thats an extension we have made. We hope to release it soon! >>>> >>>> Best, >>>> Suraj >>>> >>>> >>>> On Feb 20, 2014, at 10:42 PM, Reuti wrote: >>>> >>>>> Hi, >>>>> >>>>> Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: >>>>> >>>>>> I want use MPICH under the Torque RMS. I can see that mpiexec will >>>>>> automatically take the nodefile from $PBS_NODEFILE. >>>>>> I am working on dynamically adding nodes to an existing job in >>>>>> torque. >>>>>> So there might be a point where the job actually has more number of >>>>>> nodes than what was listed in $PBS_NODEFILE. >>>>> >>>>> How do you tell Torque that it should increase the number of granted >>>>> slots/nodes for a particular already running job? >>>>> >>>>> -- Reuti >>>>> >>>>> >>>>>> Is it possible to do an MPI_Comm_spawn on these new nodes (which is >>>>>> not listed in PBS_NODEFILE but is actually a part of job)? For >>>>>> example >>>>>> using some info argument such as add-host or something? Or if it >>>>>> entails that the PBS_NODEFILE must be changed with the newly added >>>>>> nodes, I can do that too. >>>>>> But the basic question is, can MPICH spawn processes on these newly >>>>>> allocated nodes for the same job through some mechanism? >>>>>> >>>>>> Best, >>>>>> Suraj >>>>>> _______________________________________________ >>>>>> discuss mailing list discuss(a)mpich.org >>>>>> To manage subscription options or unsubscribe: >>>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>>> >>>>> _______________________________________________ >>>>> discuss mailing list discuss(a)mpich.org >>>>> To manage subscription options or unsubscribe: >>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>> >>>> _______________________________________________ >>>> discuss mailing list discuss(a)mpich.org >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/discuss >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Balaji, Pavan 20 Feb '14

20 Feb '14

If Torque TM libraries are found at configure time, they’ll be used. If not, ssh will be used. You can pass --with-pbs to configure to help it find the TM libraries. — Pavan On 2/20/14, 7:08 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> wrote: >Hi Pavan, > >Thanks for your response. For the hosts that are not a part of the >original PBS job, will mpich try to spawn jobs there through torque/pbs >or will the spawn happen through ssh? > >Best, >Suraj > >On Feb 21, 2014, at 1:33 AM, Balaji, Pavan wrote: > >> >> I thought I responded to this email (it’s in my sent folder), but it >>never >> showed up on the mailing list. Here it is again: >> >> Hello, >> >> Yes, you can use the info argument “hosts” to specify nodes that are >>not a >> part of the original PBS job. >> >> — Pavan >> >> >> On 2/20/14, 4:36 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> >> wrote: >> >>> Hi, >>> >>> Thats an extension we have made. We hope to release it soon! >>> >>> Best, >>> Suraj >>> >>> >>> On Feb 20, 2014, at 10:42 PM, Reuti wrote: >>> >>>> Hi, >>>> >>>> Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: >>>> >>>>> I want use MPICH under the Torque RMS. I can see that mpiexec will >>>>> automatically take the nodefile from $PBS_NODEFILE. >>>>> I am working on dynamically adding nodes to an existing job in >>>>>torque. >>>>> So there might be a point where the job actually has more number of >>>>> nodes than what was listed in $PBS_NODEFILE. >>>> >>>> How do you tell Torque that it should increase the number of granted >>>> slots/nodes for a particular already running job? >>>> >>>> -- Reuti >>>> >>>> >>>>> Is it possible to do an MPI_Comm_spawn on these new nodes (which is >>>>> not listed in PBS_NODEFILE but is actually a part of job)? For >>>>>example >>>>> using some info argument such as add-host or something? Or if it >>>>> entails that the PBS_NODEFILE must be changed with the newly added >>>>> nodes, I can do that too. >>>>> But the basic question is, can MPICH spawn processes on these newly >>>>> allocated nodes for the same job through some mechanism? >>>>> >>>>> Best, >>>>> Suraj >>>>> _______________________________________________ >>>>> discuss mailing list discuss(a)mpich.org >>>>> To manage subscription options or unsubscribe: >>>>> https://lists.mpich.org/mailman/listinfo/discuss >>>> >>>> _______________________________________________ >>>> discuss mailing list discuss(a)mpich.org >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/discuss >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > >_______________________________________________ >discuss mailing list discuss(a)mpich.org >To manage subscription options or unsubscribe: >https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Suraj Prabhakaran 20 Feb '14

20 Feb '14

Hi Pavan, Thanks for your response. For the hosts that are not a part of the original PBS job, will mpich try to spawn jobs there through torque/pbs or will the spawn happen through ssh? Best, Suraj On Feb 21, 2014, at 1:33 AM, Balaji, Pavan wrote: > > I thought I responded to this email (it’s in my sent folder), but it never > showed up on the mailing list. Here it is again: > > Hello, > > Yes, you can use the info argument “hosts” to specify nodes that are not a > part of the original PBS job. > > — Pavan > > > On 2/20/14, 4:36 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> > wrote: > >> Hi, >> >> Thats an extension we have made. We hope to release it soon! >> >> Best, >> Suraj >> >> >> On Feb 20, 2014, at 10:42 PM, Reuti wrote: >> >>> Hi, >>> >>> Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: >>> >>>> I want use MPICH under the Torque RMS. I can see that mpiexec will >>>> automatically take the nodefile from $PBS_NODEFILE. >>>> I am working on dynamically adding nodes to an existing job in torque. >>>> So there might be a point where the job actually has more number of >>>> nodes than what was listed in $PBS_NODEFILE. >>> >>> How do you tell Torque that it should increase the number of granted >>> slots/nodes for a particular already running job? >>> >>> -- Reuti >>> >>> >>>> Is it possible to do an MPI_Comm_spawn on these new nodes (which is >>>> not listed in PBS_NODEFILE but is actually a part of job)? For example >>>> using some info argument such as add-host or something? Or if it >>>> entails that the PBS_NODEFILE must be changed with the newly added >>>> nodes, I can do that too. >>>> But the basic question is, can MPICH spawn processes on these newly >>>> allocated nodes for the same job through some mechanism? >>>> >>>> Best, >>>> Suraj >>>> _______________________________________________ >>>> discuss mailing list discuss(a)mpich.org >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/discuss >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Balaji, Pavan 20 Feb '14

20 Feb '14

I thought I responded to this email (it’s in my sent folder), but it never showed up on the mailing list. Here it is again: Hello, Yes, you can use the info argument “hosts” to specify nodes that are not a part of the original PBS job. — Pavan On 2/20/14, 4:36 PM, "Suraj Prabhakaran" <suraj.prabhakaran(a)gmail.com> wrote: >Hi, > >Thats an extension we have made. We hope to release it soon! > >Best, >Suraj > > >On Feb 20, 2014, at 10:42 PM, Reuti wrote: > >> Hi, >> >> Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: >> >>> I want use MPICH under the Torque RMS. I can see that mpiexec will >>>automatically take the nodefile from $PBS_NODEFILE. >>> I am working on dynamically adding nodes to an existing job in torque. >>>So there might be a point where the job actually has more number of >>>nodes than what was listed in $PBS_NODEFILE. >> >> How do you tell Torque that it should increase the number of granted >>slots/nodes for a particular already running job? >> >> -- Reuti >> >> >>> Is it possible to do an MPI_Comm_spawn on these new nodes (which is >>>not listed in PBS_NODEFILE but is actually a part of job)? For example >>>using some info argument such as add-host or something? Or if it >>>entails that the PBS_NODEFILE must be changed with the newly added >>>nodes, I can do that too. >>> But the basic question is, can MPICH spawn processes on these newly >>>allocated nodes for the same job through some mechanism? >>> >>> Best, >>> Suraj >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > >_______________________________________________ >discuss mailing list discuss(a)mpich.org >To manage subscription options or unsubscribe: >https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] turning off MPI abort messages
by Barry Smith 20 Feb '14

20 Feb '14

On Feb 20, 2014, at 4:40 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: > Just configure MPICH such that snprintf isn't discovered by configure > and you won't see these messages. > > The other solution is to fix PETSc so that people can't crash it so easily ;-) Here we go again. It is not CRASHING; it has detected an error conditioning and trying to appropriately and cleanly terminate. The reason it needs to use MPI_Abort() is that often detecting error conditions is not a uniformly collective thing. Printing a suitable error message and ending is not crashing. But with all the badly formatted “error messages” printed by MPICH I can not control at the end it looks like it is crashing. Barry > > Jeff > > On Thu, Feb 20, 2014 at 3:19 PM, Jim Dinan <james.dinan(a)gmail.com> wrote: >> If you can find a way to call MPI_Finalize instead, you will portably >> eliminate these messages. >> >> A lesser solution would be to provide an error code of 0 (or MPI_SUCCESS) to >> MPI_Abort, e.g. MPI_Comm_abort(MPI_COMM_WORLD, MPI_SUCCESS). This would >> eliminate the error message that you are getting from the job launcher. >> MPICH could be modified to be quiet about the abort when the application >> aborts with an error code of MPI_SUCCESS. >> >> ~Jim. >> >> >> On Thu, Feb 20, 2014 at 12:33 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: >>> >>> >>> Is there any way to turn off MPICH (and others) printing messages about >>> MPI_Abort? We have already prepared and presented useful error messages to >>> the user about the situation and would like to avoid having these additional >>> messages printed (that often make the situation look worse than it is) >>> >>> Thanks >>> >>> Barry >>> >>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >>> [cli_0]: aborting job: >>> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >>> >>> >>> ==================================================================mailto:[email protected]================= >>> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES >>> = EXIT CODE: 56 >>> = CLEANING UP REMAINING PROCESSES >>> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES >>> >>> =================================================================================== >>> >>> >>> >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] turning off MPI abort messages
by Jeff Hammond 20 Feb '14

20 Feb '14

Just configure MPICH such that snprintf isn't discovered by configure and you won't see these messages. The other solution is to fix PETSc so that people can't crash it so easily ;-) Jeff On Thu, Feb 20, 2014 at 3:19 PM, Jim Dinan <james.dinan(a)gmail.com> wrote: > If you can find a way to call MPI_Finalize instead, you will portably > eliminate these messages. > > A lesser solution would be to provide an error code of 0 (or MPI_SUCCESS) to > MPI_Abort, e.g. MPI_Comm_abort(MPI_COMM_WORLD, MPI_SUCCESS). This would > eliminate the error message that you are getting from the job launcher. > MPICH could be modified to be quiet about the abort when the application > aborts with an error code of MPI_SUCCESS. > > ~Jim. > > > On Thu, Feb 20, 2014 at 12:33 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: >> >> >> Is there any way to turn off MPICH (and others) printing messages about >> MPI_Abort? We have already prepared and presented useful error messages to >> the user about the situation and would like to avoid having these additional >> messages printed (that often make the situation look worse than it is) >> >> Thanks >> >> Barry >> >> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >> [cli_0]: aborting job: >> application called MPI_Abort(MPI_COMM_WORLD, 56) - process 0 >> >> >> ==================================================================mailto:[email protected]================= >> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES >> = EXIT CODE: 56 >> = CLEANING UP REMAINING PROCESSES >> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES >> >> =================================================================================== >> >> >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com

1 0

Re: [mpich-discuss] MPICH under torque - adding new nodes
by Suraj Prabhakaran 20 Feb '14

20 Feb '14

Hi, Thats an extension we have made. We hope to release it soon! Best, Suraj On Feb 20, 2014, at 10:42 PM, Reuti wrote: > Hi, > > Am 20.02.2014 um 18:10 schrieb Suraj Prabhakaran: > >> I want use MPICH under the Torque RMS. I can see that mpiexec will automatically take the nodefile from $PBS_NODEFILE. >> I am working on dynamically adding nodes to an existing job in torque. So there might be a point where the job actually has more number of nodes than what was listed in $PBS_NODEFILE. > > How do you tell Torque that it should increase the number of granted slots/nodes for a particular already running job? > > -- Reuti > > >> Is it possible to do an MPI_Comm_spawn on these new nodes (which is not listed in PBS_NODEFILE but is actually a part of job)? For example using some info argument such as add-host or something? Or if it entails that the PBS_NODEFILE must be changed with the newly added nodes, I can do that too. >> But the basic question is, can MPICH spawn processes on these newly allocated nodes for the same job through some mechanism? >> >> Best, >> Suraj >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0