Hi,
My parallel script fails after upgrading from Neuron version 7.3 and 7.4 to 7.5 with both mpich and openmpi. The error does not happen with the test script or when a single node is used, or when the time step is not fixed. Any ideas what could be causing it? The error output is bellow:
mpiexec -np 16 nrniv -mpi init.hoc
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 7 out of range
internal error: list index 7 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
internal error: list index 7 out of range
internal error: list index 7 out of range
internal error: list index 6 out of range
internal error: list index 6 out of range
===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 5719 RUNNING AT martynas-S5520SC
= EXIT CODE: 134
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Aborted (signal 6)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions
Thanks,
Martynas
Parallel jobs fail after upgrading to Neuron version 7.5
Moderator: hines
-
- Posts: 3
- Joined: Thu Jun 13, 2013 11:00 am
-
- Posts: 3
- Joined: Thu Jun 13, 2013 11:00 am
Re: Parallel jobs fail after upgrading to Neuron version 7.5
Happens with gap junctions only
Re: Parallel jobs fail after upgrading to Neuron version 7.5
To fix this, I would need to reproduce on my machine. Please send a zip file with all the necessary mod, hoc, py, ses, files needed to run a sim.
Email is best for this at michael.hines@yale.edu
Email is best for this at michael.hines@yale.edu
Re: Parallel jobs fail after upgrading to Neuron version 7.5
There was an internal error fixed by the latest commit
https://github.com/nrnhines/nrn/commit/ ... 055a54edfa
Fixed ParallelContext transfer index error when output buffer size
is greater than number of sgids. Possibly also an error if extracellular
was inserted.
This bug was created during a major change to the partrans.cpp implementation that allowed non-voltage variables to be transferred.
That happened on July 30, 2016.
https://github.com/nrnhines/nrn/commit/ ... 055a54edfa
Fixed ParallelContext transfer index error when output buffer size
is greater than number of sgids. Possibly also an error if extracellular
was inserted.
This bug was created during a major change to the partrans.cpp implementation that allowed non-voltage variables to be transferred.
That happened on July 30, 2016.