I'm using Abaqus 6.14 on a Linux cluster. I am having trouble running a co-simulation using Abaqus using command lines. I have tried both the Co-simulation command and the cse command. The Co-simulation command always seems to give back errors, while the cse command either reports back that it requires a csedirector string (which as far as I know, I don't have), or a port number (which is also not used on this Linux cluster system as far as I know).

Here are the line forms I entered. For Co-simulation command the form was: "abaqus cosimulation cosimjob=cosimjobName job=jobName-fluid,jobName-solid scratch=/scratch,/scratch -configure", and I get back this error:

"

Traceback (most recent call last):

   File "SMAPyaModules/SMAPyaDriverPy.m/src/cosimulation.py", line 1139, in

   File "SMAPyaModules/SMAPyaDriverPy.m/src/analysis.py", line 1135, in run

   File "SMAPyaModules/SMAPyaDriverPy.m/src/cosimulation.py", line 133, in _validateJob

TypeError: 'NoneType' object is unsubscriptable

"

for Abaqus 6.13, or this error:

"

Traceback (most recent call last):

   File "SMAPyaModules/SMAPyaDriverPy.m/src/cosimulation.py", line 1206, in

   File "SMAPyaModules/SMAPyaDriverPy.m/src/analysis.py", line 1174, in run

   File "SMAPyaModules/SMAPyaDriverPy.m/src/cosimulation.py", line 141, in _validateJob

TypeError: 'NoneType' object has no attribute '__getitem__'

"

for Abaqus 6.14.

For the cse command I ran in the form "abaqus job=jobName scratch=/scratch input=inputName", and got back this prompt: "Co-simulation csedirector string:".

Some information about the cluster: The Linux cluster I'm using has a lx-amd64 achetecture for each of it's compute nodes. The nodes use RHEL6. Some use 6.3, but most use 6.6. About half of the nodes use CentOS6.7.

Thanks!