[ahmadbakri]

Hi all,


I have a problem, I need to run simulation under linux system and using loadleveler queuing system. I have to use the batch mode and some run file.cmd and then use llsubmit. Any idea how would the batch ( run ) file would look like. I was trying similar scripts like the one below, but got no luck. I am able to run local parallel like this:



#@output = $(jobid).out
#@ error = $(jobid).err
#@ job_type = parallel
#@ node = 1
##@ tasks_per_node = 2
#@ total_tasks=4
##@ node_usage = shared
#@ class = medium
##@ cpu_limit=900:00:00, 900:00:00
#@ queue
NP=$(cat $LOADL_HOSTFILE | wc -l)

cfx5solve -part $NP -start-method "PVM Local Parallel" -def "NormalMeshCase1.def"




However, with distributed parallel using following script, I am not able

#@output = $(jobid).out
#@ error = $(jobid).err
#@ job_type = parallel
#@ node = 2
##@ tasks_per_node = 2
#@ total_tasks=4
##@ node_usage = shared
#@ class = medium
##@ cpu_limit=900:00:00, 900:00:00
#@ queue
NP=$(cat $LOADL_PROCESSOR_LIST | wc -l)



cfx5solve -part $NP -start-method "PVM Distributed Parallel" -par-dist $LOADL_PROCESSOR_LIST -def "NormalMeshCase1.def"


I tried to run directly through solver but I got error msg



An error has occurred in cfx5solve: |
| |
| Remote connection to kuauhccn06 (ku-auh-ccn06) exited with return |
| code 1. It gave the following output: |
| |
| connect to address 172.30.102.6 port 544: Connection refused |
| connect to address 172.30.102.6 port 544: Connection refused |
| trying normal rsh (/usr/bin/rsh) |
| ku-auh-ccn06.kustar.ac.ae: Connection refused |
| |
| Check that you have typed the hostname correctly, that you have an |
| account "ahmad.albakri" on the specified host with permission to |
| rsh from this host, and that (particularly for Windows hosts) it |
| is running an rsh daemon. You can use the following command to |
| check the connection to a UNIX machine: |
| |
| rsh ku-auh-ccn06 uname |
| |
| or the following command if it is a Windows machine: |
| |
| rsh ku-auh-ccn06 cmd /c echo working |
+--------------------------------------------------------------------+


+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| The architecture string for host kuauhccn06 (ku-auh-ccn06) could |
| not be determined. |
+--------------------------------------------------------------------+


+-------------------------------------------------------------------

The second question I have, if am not able to run directly through solver ( interactive method) , would I still have chance to run using batch file mode ?

ANY HELP PLS

[ghorrocks]

You might need to talk to somebody more familiar with loadleveller to fix this.

But one thing I can think of which is worth looking at is that loadleveller might be running using a different user account name, and that account might not have permissions to run rsh. I would check what account loadlleveller is trying to use when it starts a job.