[mpich-discuss] mpich/hydra error while running my model

Balaji, Pavan balaji at anl.gov
Tue Aug 5 17:46:30 CDT 2014


The error message seems to point out the problem:

> memcpy argument memory ranges overlap, dst_=0x2b1e91fd2010 src_=0x2b1e91fd2024 len_=496128


  — Pavan

On Aug 5, 2014, at 3:38 PM, Chun Meng <miracle.chun.2010 at gmail.com> wrote:

> Hi All,
> I have mpich-3.1.2 and hydra-3.1.2 installed in the /opt directory  on the cluster  which I run a regional climate model on.
> Now ,I got a error message listed here in red.Anyone have any idea about this?
> Thanks a lot!Any help is appreciated! 
> 
> 
> Assertion failed in file src/mpi/datatype/unpack.c at line 60: FALSE
> memcpy argument memory ranges overlap, dst_=0x2b1e91fd2010 src_=0x2b1e91fd2024 len_=496128
> 
> internal ABORT - process 0
> [proxy:0:1 at C006] HYD_pmcd_pmip_control_cmd_cb (pm/pmiserv/pmip_cb.c:885): assert (!closed) failed
> [proxy:0:1 at C006] HYDT_dmxu_poll_wait_for_event (tools/demux/demux_poll.c:76): callback returned error status
> [proxy:0:1 at C006] main (pm/pmiserv/pmip.c:206): demux engine error waiting for event
> [mpiexec at C001] HYDT_bscu_wait_for_completion (tools/bootstrap/utils/bscu_wait.c:75): one of the processes terminated badly; aborting
> [mpiexec at C001] HYDT_bsci_wait_for_completion (tools/bootstrap/src/bsci_wait.c:23): launcher returned error waiting for completion
> [mpiexec at C001] HYD_pmci_wait_for_completion (pm/pmiserv/pmiserv_pmci.c:218): launcher returned error waiting for completion
> 
> Best,
> Chun
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss

--
Pavan Balaji  ✉️
http://www.mcs.anl.gov/~balaji



More information about the discuss mailing list