Merge pull request #6968 from rhc54/cmr31/oob

v3.1.x: Be a little less restrictive on interface requirements

Author: jsquyres

orte/mca/oob/tcp/help-oob-tcp.txt

@@ -10,8 +10,8 @@
 #                         University of Stuttgart.  All rights reserved.
 # Copyright (c) 2004-2005 The Regents of the University of California.
 #                         All rights reserved.
-# Copyright (c) 2014-2017 Intel, Inc. All rights reserved.
-# Copyright (c) 2015      Cisco Systems, Inc.  All rights reserved.
+# Copyright (c) 2014-2019 Intel, Inc.  All rights reserved.
+# Copyright (c) 2015-2018 Cisco Systems, Inc.  All rights reserved
 # $COPYRIGHT$
 #
 # Additional copyrights may follow
@@ -113,3 +113,27 @@ The connection was rejected.
 Static ports were requested while orte_fwd_mpirun_port was set.
 Both options cannot be simultaneously set. Please either set
 orte_fwd_mpirun_port=false or remove any static port directives.
+#
+[version mismatch]
+Open MPI detected a mismatch in versions between two processes.  This
+typically means that you executed "mpirun" (or "mpiexec") from one
+version of Open MPI on on node, but your default path on one of the
+other nodes upon which you launched found a different version of Open
+MPI.
+
+Open MPI only supports running exactly the same version between all
+processes in a single job.
+
+This will almost certainly cause unpredictable behavior, and may end
+up aborting your job.
+
+  Local host:             %s
+  Local process name:     %s
+  Local Open MPI version: %s
+  Peer host:              %s
+  Peer process name:      %s
+  Peer Open MPI version:  %s
+#
+[no-listeners]
+No sockets were able to be opened on the available protocols
+(IPv4 and/or IPv6). Please check your network and retry.

orte/mca/oob/tcp/oob_tcp_listener.c

@@ -13,7 +13,7 @@
  *                         All rights reserved.
  * Copyright (c) 2009-2015 Cisco Systems, Inc.  All rights reserved.
  * Copyright (c) 2011      Oak Ridge National Labs.  All rights reserved.
- * Copyright (c) 2013-2017 Intel, Inc.  All rights reserved.
+ * Copyright (c) 2013-2019 Intel, Inc.  All rights reserved.
  * Copyright (c) 2015      Research Organization for Information Science
  *                         and Technology (RIST). All rights reserved.
  * $COPYRIGHT$
@@ -98,7 +98,7 @@ static void connection_event_handler(int sd, short flags, void* cbdata);
  */
 int orte_oob_tcp_start_listening(void)
 {
-    int rc;
+    int rc = ORTE_SUCCESS, rc2 = ORTE_SUCCESS;
     mca_oob_tcp_listener_t *listener;
 
     /* if we don't have any TCP interfaces, we shouldn't be here */
@@ -112,19 +112,19 @@ int orte_oob_tcp_start_listening(void)
     }
 
     /* create listen socket(s) for incoming connection attempts */
-    if (ORTE_SUCCESS != (rc = create_listen())) {
-        ORTE_ERROR_LOG(rc);
-        return rc;
-    }
+    rc = create_listen();
 
 #if OPAL_ENABLE_IPV6
     /* create listen socket(s) for incoming connection attempts */
-    if (ORTE_SUCCESS != (rc = create_listen6())) {
-        ORTE_ERROR_LOG(rc);
-        return rc;
-    }
+    rc2 = create_listen6();
 #endif
 
+    if (ORTE_SUCCESS != rc && ORTE_SUCCESS != rc2) {
+        /* we were unable to open any listening sockets */
+        opal_show_help("help-oob-tcp.txt", "no-listeners", true);
+        return ORTE_ERR_FATAL;
+    }
+
     /* if I am the HNP, start a listening thread so we can
      * harvest connection requests as rapidly as possible
      */