Fix comm_spawn. The problem stems from our use of the existence of an attribute as equivalent to a boolean "true" - in other words, we only confirm the existence of an attribute on a list to indicate something as opposed to looking at its specific value. Hence, we create the attribute with a type of ORTE_UNDEF - which is fine...until we then attempt to store/retrieve that attribute from the registry. In that case, the DSS barks because it treats ORTE_UNDEF as an error.

The only place where we attempt to store/retrieve attributes is in the RMAPS framework in support of comm_spawn. So this is where things broke down. The fix was simply to say "if the attribute data type is ORTE_UNDEF, then treat it like a boolean with value true". Trivial fix - solves problem.

This commit was SVN r14910.

orte/mca/rmaps/base/rmaps_base_registry_fns.c

@@ -596,6 +596,7 @@ int orte_rmaps_base_store_mapping_plan(orte_jobid_t job, opal_list_t *attr_list)
         ORTE_RMAPS_BOOKMARK
     };
     orte_std_cntr_t num_attrs_defd;
+    bool tval = true;
 
     OPAL_TRACE(2);
 
@@ -637,7 +638,7 @@ int orte_rmaps_base_store_mapping_plan(orte_jobid_t job, opal_list_t *attr_list)
                 }
             } else {
                 if (ORTE_SUCCESS != (rc = orte_gpr.create_keyval(&(value->keyvals[j]), attr->key,
-                                                                 ORTE_UNDEF, NULL))) {
+                                                                 ORTE_BOOL, &tval))) {
                     ORTE_ERROR_LOG(rc);
                     OBJ_RELEASE(value);
                     return rc;
@@ -716,24 +717,13 @@ int orte_rmaps_base_get_mapping_plan(orte_jobid_t job, opal_list_t *attr_list)
     for (i=0; i < value->cnt; i++) {
         kval = value->keyvals[i];
 
-        if (NULL != kval->value) {
-            if (ORTE_SUCCESS != (rc = orte_rmgr.add_attribute(attr_list, kval->key,
-                                                              kval->value->type,
-                                                              kval->value->data,
-                                                              ORTE_RMGR_ATTR_OVERRIDE))) {
-                ORTE_ERROR_LOG(rc);
-                OBJ_RELEASE(value);
-                return rc;
-            }
-        } else {
-            if (ORTE_SUCCESS != (rc = orte_rmgr.add_attribute(attr_list, kval->key,
-                                                              ORTE_UNDEF,
-                                                              NULL,
-                                                              ORTE_RMGR_ATTR_OVERRIDE))) {
-                ORTE_ERROR_LOG(rc);
-                OBJ_RELEASE(value);
-                return rc;
-            }
+        if (ORTE_SUCCESS != (rc = orte_rmgr.add_attribute(attr_list, kval->key,
+                                                          kval->value->type,
+                                                          kval->value->data,
+                                                          ORTE_RMGR_ATTR_OVERRIDE))) {
+            ORTE_ERROR_LOG(rc);
+            OBJ_RELEASE(value);
+            return rc;
         }
     }