Orterun creates a "clean" copy of its environment for use in launching procs. This includes properly setting LD_LIBRARY_PATH and PATH, among other things. Unfortunately, our PLM modules were using the local environ instead of the saved copy, thus missing a number of things that really should have been included. From what I see, we got away with the error because the PLM's were duplicating all that setup logic themselves - I'll clean this up over the next few days.

Meantime, correct the PLM's so they use the correct environ for launching.

This commit was SVN r18713.

orte/mca/plm/alps/plm_alps_module.c

@@ -328,7 +328,7 @@ static int plm_alps_launch_job(orte_job_t *jdata)
     }
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
     
     /* add the nodelist */
     var = mca_base_param_environ_variable("orte", "alps", "nodelist");

orte/mca/plm/ccp/plm_ccp_module.c

@@ -255,7 +255,7 @@ GETMAP:
     bin_base = opal_basename(opal_install_dirs.bindir);
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
 
     /* add our umask -- see big note in orted.c */
     current_umask = umask(0);

orte/mca/plm/lsf/plm_lsf_module.c

@@ -282,7 +282,7 @@ static int plm_lsf_launch_job(orte_job_t *jdata)
     }
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
 
     if (mca_plm_lsf_component.timing) {
         if (0 != gettimeofday(&launchstart, NULL)) {

orte/mca/plm/process/plm_process_module.c

@@ -780,7 +780,7 @@ int orte_plm_process_launch(orte_job_t *jdata)
             set_handler_default(SIGCHLD);
             
             /* setup environment */
-            env = opal_argv_copy(environ);
+            env = opal_argv_copy(orte_launch_environ);
             
             /* exec the daemon */
             if (0 < opal_output_get_verbosity(orte_plm_globals.output)) {

orte/mca/plm/rsh/plm_rsh_module.c

@@ -545,7 +545,7 @@ static void ssh_child(int argc, char **argv,
     sigset_t sigs;
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
     
     /* ensure that only the ssh plm is selected on the remote daemon */
     var = mca_base_param_environ_variable("plm", NULL, NULL);

orte/mca/plm/slurm/plm_slurm_module.c

@@ -340,7 +340,7 @@ static int plm_slurm_launch_job(orte_job_t *jdata)
     }
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
 
     /* add the nodelist */
     var = mca_base_param_environ_variable("orte", "slurm", "nodelist");

orte/mca/plm/tm/plm_tm_module.c

@@ -230,7 +230,7 @@ static int plm_tm_launch_job(orte_job_t *jdata)
     bin_base = opal_basename(opal_install_dirs.bindir);
 
     /* setup environment */
-    env = opal_argv_copy(environ);
+    env = opal_argv_copy(orte_launch_environ);
 
     /* add our umask -- see big note in orted.c */
     current_umask = umask(0);