doc: update document build version for 4.1 branch

This release does not make any changes to the extension database
objects.

HISTORY

@@ -1,3 +1,48 @@
+4.1.0   2018-??-??
+        repmgr: change default log_level to INFO, add documentation; GitHub #470 (Ian)
+        repmgr: add "--missing-slots" check to "repmgr node check" (Ian)
+        repmgr: improve command line error handling; GitHub #464 (Ian)
+        repmgr: fix "standby register --wait-sync" when no timeout provided (Ian)
+        repmgr: "cluster show" returns non-zero value if an issue encountered;
+          GitHub #456 (Ian)
+        repmgr: "node check" and "node status" returns non-zero value if an issue
+           encountered (Ian)
+        repmgr: add CSV output mode to "cluster event"; GitHub #471 (Ian)
+        repmgr: add -q/--quiet option to suppress non-error output; GitHub #468 (Ian)
+        repmgr: "node status" returns non-zero value if an issue encountered (Ian)
+        repmgr: enable "recovery_min_apply_delay" to be 0; GitHub #448 (Ian)
+        repmgr: "cluster cleanup" - add missing help options; GitHub #461/#462 (gclough)
+        repmgr: ensure witness node follows new primary after switchover;
+          GitHub #453 (Ian)
+        repmgr: fix witness node handling in "node check"/"node status";
+          GitHub #451 (Ian)
+        repmgr: fix "primary_slot_name" when using "standby clone" with --recovery-conf-only;
+          GitHub #474 (Ian)
+        repmgr: don't perform a switchover if an exclusive backup is running;
+          GitHub #476 (Martín)
+        repmgr: enable "witness unregister" to be run on any node; GitHub #472 (Ian)
+        repmgrd: create a PID file by default; GitHub #457 (Ian)
+        repmgrd: daemonize process by default; GitHub #458 (Ian)
+
+4.0.6   2018-06-14
+        repmgr: (witness register) prevent registration of a witness server with the
+          same name as an existing node (Ian)
+        repmgr: (standby follow) check node has actually connected to new primary
+          before reporting success; GitHub #444 (Ian)
+        repmgr: (standby clone) improve handling of external configuration file copying,
+          including consideration in --dry-run check; GitHub #443 (Ian)
+        repmgr: (standby clone) don't require presence of "user" parameter in
+          conninfo string; GitHub #437 (Ian)
+        repmgr: (standby clone) improve documentation of --recovery-conf-only
+          mode; GitHub #438 (Ian)
+        repmgr: (node rejoin) fix bug when parsing --config-files parameter;
+          GitHub #442 (Ian)
+        repmgr: when using --dry-run, force log level to INFO to ensure output
+          will always be displayed; GitHub #441 (Ian)
+        repmgr: (cluster matrix/crosscheck) return non-zero exit code if node
+           connection issues detected; GitHub #447 (Ian)
+        repmgrd: ensure local node is counted as quorum member; GitHub #439 (Ian)
+
 4.0.5   2018-05-02
         repmgr: poll demoted primary after restart as a standby during a
           switchover operation; GitHub #408 (Ian)

Makefile.in

@@ -11,7 +11,10 @@ EXTENSION = repmgr
 
 DATA = \
   repmgr--unpackaged--4.0.sql \
-  repmgr--4.0.sql
+  repmgr--4.0.sql \
+  repmgr--4.0--4.1.sql \
+  repmgr--4.1.sql
+
 
 REGRESS = repmgr_extension
 

configfile.c

@@ -29,9 +29,6 @@ static bool config_file_provided = false;
 bool		config_file_found = false;
 
 static void _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *warning_list);
-static bool parse_bool(const char *s,
-		   const char *config_item,
-		   ItemList *error_list);
 
 static void _parse_line(char *buf, char *name, char *value);
 static void parse_event_notifications_list(t_configuration_options *options, const char *arg);
@@ -319,13 +316,26 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 	options->use_primary_conninfo_password = false;
 	memset(options->passfile, 0, sizeof(options->passfile));
 
-	/*-----------------------
+	/*-------------------------
 	 * standby promote settings
-	 *------------------------
+	 *-------------------------
 	 */
 	options->promote_check_timeout = DEFAULT_PROMOTE_CHECK_TIMEOUT;
 	options->promote_check_interval = DEFAULT_PROMOTE_CHECK_INTERVAL;
 
+	/*------------------------
+	 * standby follow settings
+	 *------------------------
+	 */
+	options->primary_follow_timeout = DEFAULT_PRIMARY_FOLLOW_TIMEOUT;
+	options->standby_follow_timeout = DEFAULT_STANDBY_FOLLOW_TIMEOUT;
+
+	/*------------------------
+	 * standby switchover settings
+	 *------------------------
+	 */
+	options->standby_reconnect_timeout = DEFAULT_STANDBY_RECONNECT_TIMEOUT;
+
 	/*-----------------
 	 * repmgrd settings
 	 *-----------------
@@ -345,8 +355,8 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 	options->degraded_monitoring_timeout = -1;
 	options->async_query_timeout = DEFAULT_ASYNC_QUERY_TIMEOUT;
 	options->primary_notification_timeout = DEFAULT_PRIMARY_NOTIFICATION_TIMEOUT;
-	options->primary_follow_timeout = DEFAULT_PRIMARY_FOLLOW_TIMEOUT;
-	options->standby_reconnect_timeout = DEFAULT_STANDBY_RECONNECT_TIMEOUT;
+	options->repmgrd_standby_startup_timeout = -1; /* defaults to "standby_reconnect_timeout" if not set */
+	memset(options->repmgrd_pid_file, 0, sizeof(options->repmgrd_pid_file));
 
 	/*-------------
 	 * witness settings
@@ -527,6 +537,20 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 		else if (strcmp(name, "promote_check_interval") == 0)
 			options->promote_check_interval = repmgr_atoi(value, name, error_list, 1);
 
+		/* standby follow settings */
+		else if (strcmp(name, "primary_follow_timeout") == 0)
+			options->primary_follow_timeout = repmgr_atoi(value, name, error_list, 0);
+		else if (strcmp(name, "standby_follow_timeout") == 0)
+			options->standby_follow_timeout = repmgr_atoi(value, name, error_list, 0);
+
+		/* standby switchover settings */
+		else if (strcmp(name, "standby_reconnect_timeout") == 0)
+			options->standby_reconnect_timeout = repmgr_atoi(value, name, error_list, 0);
+
+		/* node rejoin settings */
+		else if (strcmp(name, "node_rejoin_timeout") == 0)
+			options->node_rejoin_timeout = repmgr_atoi(value, name, error_list, 0);
+
 		/* node check settings */
 		else if (strcmp(name, "archive_ready_warning") == 0)
 			options->archive_ready_warning = repmgr_atoi(value, name, error_list, 1);
@@ -576,10 +600,10 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 			options->async_query_timeout = repmgr_atoi(value, name, error_list, 0);
 		else if (strcmp(name, "primary_notification_timeout") == 0)
 			options->primary_notification_timeout = repmgr_atoi(value, name, error_list, 0);
-		else if (strcmp(name, "primary_follow_timeout") == 0)
-			options->primary_follow_timeout = repmgr_atoi(value, name, error_list, 0);
-		else if (strcmp(name, "standby_reconnect_timeout") == 0)
-			options->standby_reconnect_timeout = repmgr_atoi(value, name, error_list, 0);
+		else if (strcmp(name, "repmgrd_standby_startup_timeout") == 0)
+			options->repmgrd_standby_startup_timeout = repmgr_atoi(value, name, error_list, 0);
+		else if (strcmp(name, "repmgrd_pid_file") == 0)
+			strncpy(options->repmgrd_pid_file, value, MAXPGPATH);
 
 		/* witness settings */
 		else if (strcmp(name, "witness_sync_interval") == 0)
@@ -761,6 +785,18 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 		PQconninfoFree(conninfo_options);
 	}
 
+
+	/* set values for parameters which default to other parameters */
+
+	/*
+	 * From 4.1, "repmgrd_standby_startup_timeout" replaces "standby_reconnect_timeout"
+	 * in repmgrd; fall back to "standby_reconnect_timeout" if no value explicitly provided
+	 */
+	if (options->repmgrd_standby_startup_timeout == -1)
+	{
+		options->repmgrd_standby_startup_timeout = options->standby_reconnect_timeout;
+	}
+
 	/* add warning about changed "barman_" parameter meanings */
 	if ((options->barman_host[0] == '\0' && options->barman_server[0] != '\0') ||
 		(options->barman_host[0] != '\0' && options->barman_server[0] == '\0'))
@@ -785,6 +821,12 @@ _parse_config(t_configuration_options *options, ItemList *error_list, ItemList *
 		item_list_append(error_list,
 						 _("\replication_lag_critical\" must be greater than  \"replication_lag_warning\""));
 	}
+
+	if (options->standby_reconnect_timeout < options->node_rejoin_timeout)
+	{
+		item_list_append(error_list,
+						 _("\"standby_reconnect_timeout\" must be equal to or greater than \"node_rejoin_timeout\""));
+	}
 }
 
 
@@ -949,12 +991,11 @@ parse_time_unit_parameter(const char *name, const char *value, char *dest, ItemL
 	char	   *ptr = NULL;
 	int			targ = strtol(value, &ptr, 10);
 
-	if (targ < 1)
+	if (targ < 0)
 	{
 		if (errors != NULL)
 		{
-			item_list_append_format(
-									errors,
+			item_list_append_format(errors,
 									_("invalid value provided for \"%s\""),
 									name);
 		}
@@ -1008,6 +1049,7 @@ parse_time_unit_parameter(const char *name, const char *value, char *dest, ItemL
  * - promote_delay
  * - reconnect_attempts
  * - reconnect_interval
+ * - repmgrd_standby_startup_timeout
  * - retry_promote_interval_secs
  *
  * non-changeable options
@@ -1033,17 +1075,36 @@ reload_config(t_configuration_options *orig_options)
 	static ItemList config_errors = {NULL, NULL};
 	static ItemList config_warnings = {NULL, NULL};
 
+	PQExpBufferData errors;
+
 	log_info(_("reloading configuration file"));
 
 	_parse_config(&new_options, &config_errors, &config_warnings);
 
 	if (config_errors.head != NULL)
 	{
-		/* XXX dump errors to log */
+		ItemListCell *cell = NULL;
+
 		log_warning(_("unable to parse new configuration, retaining current configuration"));
+
+		initPQExpBuffer(&errors);
+
+		appendPQExpBuffer(&errors,
+						  "following errors were detected:\n");
+
+		for (cell = config_errors.head; cell; cell = cell->next)
+		{
+			appendPQExpBuffer(&errors,
+							  "  %s\n", cell->string);
+		}
+
+		log_detail("%s", errors.data);
+		termPQExpBuffer(&errors);
 		return false;
 	}
 
+
+
 	/* The following options cannot be changed */
 
 	if (new_options.node_id != orig_options->node_id)
@@ -1224,6 +1285,15 @@ reload_config(t_configuration_options *orig_options)
 		config_changed = true;
 	}
 
+	/* repmgrd_standby_startup_timeout */
+	if (orig_options->repmgrd_standby_startup_timeout != new_options.repmgrd_standby_startup_timeout)
+	{
+		orig_options->repmgrd_standby_startup_timeout = new_options.repmgrd_standby_startup_timeout;
+		log_info(_("\"repmgrd_standby_startup_timeout\" is now \"%i\""), new_options.repmgrd_standby_startup_timeout);
+
+		config_changed = true;
+	}
+
 	/*
 	 * Handle changes to logging configuration
 	 */
@@ -1316,13 +1386,23 @@ exit_with_config_file_errors(ItemList *config_errors, ItemList *config_warnings,
 
 
 void
-exit_with_cli_errors(ItemList *error_list)
+exit_with_cli_errors(ItemList *error_list, const char *repmgr_command)
 {
 	fprintf(stderr, _("The following command line errors were encountered:\n"));
 
 	print_item_list(error_list);
 
-	fprintf(stderr, _("Try \"%s --help\" for more information.\n"), progname());
+	if (repmgr_command != NULL)
+	{
+		fprintf(stderr, _("Try \"%s --help\" or \"%s %s --help\" for more information.\n"),
+				progname(),
+				progname(),
+				repmgr_command);
+	}
+	else
+	{
+		fprintf(stderr, _("Try \"%s --help\" for more information.\n"), progname());
+	}
 
 	exit(ERR_BAD_CONFIG);
 }
@@ -1427,7 +1507,7 @@ repmgr_atoi(const char *value, const char *config_item, ItemList *error_list, in
  *
  *   https://www.postgresql.org/docs/current/static/config-setting.html
  */
-static bool
+bool
 parse_bool(const char *s, const char *config_item, ItemList *error_list)
 {
 	PQExpBufferData errors;
@@ -1713,6 +1793,9 @@ free_parsed_argv(char ***argv_array)
 }
 
 
+
+
+
 bool
 parse_pg_basebackup_options(const char *pg_basebackup_options, t_basebackup_options *backup_options, int server_version_num, ItemList *error_list)
 {

configfile.h

@@ -98,6 +98,16 @@ typedef struct
 	int			promote_check_timeout;
 	int			promote_check_interval;
 
+	/* standby follow settings */
+	int			primary_follow_timeout;
+	int			standby_follow_timeout;
+
+	/* standby switchover settings */
+	int			standby_reconnect_timeout;
+
+	/* node rejoin settings */
+	int			node_rejoin_timeout;
+
 	/* node check settings */
 	int			archive_ready_warning;
 	int			archive_ready_critical;
@@ -120,8 +130,8 @@ typedef struct
 	int			degraded_monitoring_timeout;
 	int			async_query_timeout;
 	int			primary_notification_timeout;
-	int			primary_follow_timeout;
-	int			standby_reconnect_timeout;
+	int			repmgrd_standby_startup_timeout;
+	char		repmgrd_pid_file[MAXPGPATH];
 
 	/* BDR settings */
 	bool		bdr_local_monitoring_only;
@@ -167,6 +177,13 @@ typedef struct
 		false, "", "", { NULL, NULL }, "", false, "", false, "", \
 		/* standby promote settings */ \
 		DEFAULT_PROMOTE_CHECK_TIMEOUT, DEFAULT_PROMOTE_CHECK_INTERVAL, \
+		/* standby follow settings */ \
+		DEFAULT_PRIMARY_FOLLOW_TIMEOUT,	\
+		DEFAULT_STANDBY_FOLLOW_TIMEOUT,	\
+		/* standby switchover settings */ \
+		DEFAULT_STANDBY_RECONNECT_TIMEOUT, \
+		/* node rejoin settings */ \
+		DEFAULT_NODE_REJOIN_TIMEOUT, \
 		/* node check settings */ \
 		DEFAULT_ARCHIVE_READY_WARNING, DEFAULT_ARCHIVE_READY_CRITICAL, \
 		DEFAULT_REPLICATION_LAG_WARNING, DEFAULT_REPLICATION_LAG_CRITICAL, \
@@ -180,8 +197,7 @@ typedef struct
         false, -1, \
 		DEFAULT_ASYNC_QUERY_TIMEOUT, \
 		DEFAULT_PRIMARY_NOTIFICATION_TIMEOUT,	\
-		DEFAULT_PRIMARY_FOLLOW_TIMEOUT,	\
-		DEFAULT_STANDBY_RECONNECT_TIMEOUT,	\
+		-1, "", \
 		/* BDR settings */ \
 		false, DEFAULT_BDR_RECOVERY_TIMEOUT, \
 		/* service settings */ \
@@ -267,6 +283,10 @@ bool		reload_config(t_configuration_options *orig_options);
 
 bool		parse_recovery_conf(const char *data_dir, t_recovery_conf *conf);
 
+bool		parse_bool(const char *s,
+					   const char *config_item,
+					   ItemList *error_list);
+
 int repmgr_atoi(const char *s,
 			const char *config_item,
 			ItemList *error_list,
@@ -282,7 +302,7 @@ void free_parsed_argv(char ***argv_array);
 
 
 /* called by repmgr-client and repmgrd */
-void		exit_with_cli_errors(ItemList *error_list);
+void		exit_with_cli_errors(ItemList *error_list, const char *repmgr_command);
 void		print_item_list(ItemList *item_list);
 
 #endif							/* _REPMGR_CONFIGFILE_H_ */

configure

@@ -1,6 +1,6 @@
 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.69 for repmgr 4.0.5.
+# Generated by GNU Autoconf 2.69 for repmgr 4.1.
 #
 # Report bugs to <pgsql-bugs@postgresql.org>.
 #
@@ -582,8 +582,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='repmgr'
 PACKAGE_TARNAME='repmgr'
-PACKAGE_VERSION='4.0.5'
-PACKAGE_STRING='repmgr 4.0.5'
+PACKAGE_VERSION='4.1'
+PACKAGE_STRING='repmgr 4.1'
 PACKAGE_BUGREPORT='pgsql-bugs@postgresql.org'
 PACKAGE_URL='https://2ndquadrant.com/en/resources/repmgr/'
 
@@ -1178,7 +1178,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures repmgr 4.0.5 to adapt to many kinds of systems.
+\`configure' configures repmgr 4.1 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1239,7 +1239,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of repmgr 4.0.5:";;
+     short | recursive ) echo "Configuration of repmgr 4.1:";;
    esac
   cat <<\_ACEOF
 
@@ -1313,7 +1313,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-repmgr configure 4.0.5
+repmgr configure 4.1
 generated by GNU Autoconf 2.69
 
 Copyright (C) 2012 Free Software Foundation, Inc.
@@ -1332,7 +1332,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by repmgr $as_me 4.0.5, which was
+It was created by repmgr $as_me 4.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
@@ -2359,7 +2359,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by repmgr $as_me 4.0.5, which was
+This file was extended by repmgr $as_me 4.1, which was
 generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -2422,7 +2422,7 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-repmgr config.status 4.0.5
+repmgr config.status 4.1
 configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 

configure.in

@@ -1,4 +1,4 @@
-AC_INIT([repmgr], [4.0.5], [pgsql-bugs@postgresql.org], [repmgr], [https://2ndquadrant.com/en/resources/repmgr/])
+AC_INIT([repmgr], [4.1], [pgsql-bugs@postgresql.org], [repmgr], [https://2ndquadrant.com/en/resources/repmgr/])
 
 AC_COPYRIGHT([Copyright (c) 2010-2018, 2ndQuadrant Ltd.])
 

dbutils.c

@@ -23,6 +23,7 @@
 #include <sys/time.h>
 #include <sys/stat.h>
 #include <dirent.h>
+#include <arpa/inet.h>
 
 #include "repmgr.h"
 #include "dbutils.h"
@@ -32,6 +33,12 @@
 /* mainly for use by repmgrd */
 int			server_version_num = UNKNOWN_SERVER_VERSION_NUM;
 
+/*
+ * This is set by is_bdr_db(), which is called by every BDR-related
+ * action anyway; this is required to be able to generate appropriate
+ * queries for versions 2 and 3.
+ */
+int			bdr_version_num = UNKNOWN_BDR_VERSION_NUM;
 
 static PGconn *_establish_db_connection(const char *conninfo,
 						 const bool exit_on_error,
@@ -83,7 +90,10 @@ wrap_ddl_query(PQExpBufferData *query_buf, int replication_type, const char *fmt
 
 	if (replication_type == REPLICATION_TYPE_BDR)
 	{
-		appendPQExpBuffer(query_buf, "SELECT bdr.bdr_replicate_ddl_command($repmgr$");
+		if (bdr_version_num < 3)
+			appendPQExpBuffer(query_buf, "SELECT bdr.bdr_replicate_ddl_command($repmgr$");
+		else
+			appendPQExpBuffer(query_buf, "SELECT bdr.replicate_ddl_command($repmgr$");
 	}
 
 	va_start(arglist, fmt);
@@ -370,6 +380,37 @@ get_conninfo_value(const char *conninfo, const char *keyword, char *output)
 }
 
 
+/*
+ * Get a default conninfo value for the provided parameter, and copy
+ * it to the 'output' buffer.
+ *
+ * Returns true on success, or false on failure (provided keyword not found).
+ *
+ */
+bool
+get_conninfo_default_value(const char *param, char *output, int maxlen)
+{
+	PQconninfoOption *defs = NULL;
+	PQconninfoOption *def = NULL;
+	bool found = false;
+
+	defs = PQconndefaults();
+
+	for (def = defs; def->keyword; def++)
+	{
+		if (strncmp(def->keyword, param, maxlen) == 0)
+		{
+			strncpy(output, def->val, maxlen);
+			found = true;
+		}
+	}
+
+	PQconninfoFree(defs);
+
+	return found;
+}
+
+
 void
 initialize_conninfo_params(t_conninfo_param_list *param_list, bool set_defaults)
 {
@@ -1560,6 +1601,39 @@ repmgrd_get_local_node_id(PGconn *conn)
 }
 
 
+/*
+ * Function that checks if the primary is in exclusive backup mode.
+ * We'll use this when executing an action can conflict with an exclusive
+ * backup.
+ */
+BackupState
+server_in_exclusive_backup_mode(PGconn *conn)
+{
+	BackupState backup_state = BACKUP_STATE_UNKNOWN;
+	PGresult   *res = PQexec(conn, "SELECT pg_catalog.pg_is_in_backup()");
+
+	if (PQresultStatus(res) != PGRES_TUPLES_OK)
+	{
+		log_error(_("unable to retrieve information regarding backup mode of node"));
+		log_detail("%s", PQerrorMessage(conn));
+		PQclear(res);
+		return BACKUP_STATE_UNKNOWN;
+	}
+
+	if (atobool(PQgetvalue(res, 0, 0)) == true)
+	{
+		backup_state = BACKUP_STATE_IN_BACKUP;
+	}
+	else
+	{
+		backup_state = BACKUP_STATE_NO_BACKUP;
+	}
+
+	PQclear(res);
+
+	return backup_state;
+}
+
 
 /* ================ */
 /* result functions */
@@ -1733,7 +1807,7 @@ _populate_node_record(PGresult *res, t_node_info *node_info, int row)
 	strncpy(node_info->location, PQgetvalue(res, row, 7), MAXLEN);
 	node_info->priority = atoi(PQgetvalue(res, row, 8));
 	node_info->active = atobool(PQgetvalue(res, row, 9));
-	strncpy(node_info->config_file, PQgetvalue(res, row, 10), MAXLEN);
+	strncpy(node_info->config_file, PQgetvalue(res, row, 10), MAXPGPATH);
 
 	/* This won't normally be set */
 	strncpy(node_info->upstream_node_name, PQgetvalue(res, row, 11), MAXLEN);
@@ -2146,8 +2220,9 @@ get_downstream_nodes_with_missing_slot(PGconn *conn, int this_node_id, NodeInfoL
 					  "LEFT JOIN pg_catalog.pg_replication_slots rs "
 					  "       ON rs.slot_name = n.slot_name "
 					  "    WHERE n.slot_name IS NOT NULL"
-                      "      AND rs.slot_name IS NULL "
-                      "      AND n.upstream_node_id = %i ",
+					  "      AND rs.slot_name IS NULL "
+					  "      AND n.upstream_node_id = %i "
+					  "      AND n.type = 'standby'",
 					  this_node_id);
 
 	log_verbose(LOG_DEBUG, "get_all_node_records_with_missing_slot():\n%s", query.data);
@@ -2207,6 +2282,7 @@ _create_update_node_record(PGconn *conn, char *action, t_node_info *node_info)
 	const char *param_values[param_count];
 
 	PGresult   *res;
+	bool		success = true;
 
 	maxlen_snprintf(node_id, "%i", node_info->node_id);
 	maxlen_snprintf(priority, "%i", node_info->priority);
@@ -2293,13 +2369,13 @@ _create_update_node_record(PGconn *conn, char *action, t_node_info *node_info)
 				  node_info->node_name,
 				  node_info->node_id);
 		log_detail("%s", PQerrorMessage(conn));
-		PQclear(res);
-		return false;
+
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -2308,6 +2384,7 @@ update_node_record_set_active(PGconn *conn, int this_node_id, bool active)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
@@ -2326,13 +2403,13 @@ update_node_record_set_active(PGconn *conn, int this_node_id, bool active)
 	{
 		log_error(_("unable to update node record:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
-		return false;
+
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -2341,6 +2418,7 @@ update_node_record_set_active_standby(PGconn *conn, int this_node_id)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
@@ -2360,13 +2438,13 @@ update_node_record_set_active_standby(PGconn *conn, int this_node_id)
 	{
 		log_error(_("unable to update node record:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
-		return false;
+
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -2435,11 +2513,13 @@ update_node_record_set_primary(PGconn *conn, int this_node_id)
 	return commit_transaction(conn);
 }
 
+
 bool
 update_node_record_set_upstream(PGconn *conn, int this_node_id, int new_upstream_node_id)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	log_debug(_("update_node_record_set_upstream(): Updating node %i's upstream node to %i"),
 			  this_node_id, new_upstream_node_id);
@@ -2461,14 +2541,13 @@ update_node_record_set_upstream(PGconn *conn, int this_node_id, int new_upstream
 	{
 		log_error(_("unable to set new upstream node id:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
 
-		return false;
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -2481,6 +2560,7 @@ update_node_record_status(PGconn *conn, int this_node_id, char *type, int upstre
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
@@ -2504,14 +2584,13 @@ update_node_record_status(PGconn *conn, int this_node_id, char *type, int upstre
 	{
 		log_error(_("unable to update node record:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
 
-		return false;
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -2524,6 +2603,7 @@ update_node_record_conn_priority(PGconn *conn, t_configuration_options *options)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
@@ -2541,13 +2621,12 @@ update_node_record_conn_priority(PGconn *conn, t_configuration_options *options)
 
 	if (PQresultStatus(res) != PGRES_COMMAND_OK)
 	{
-
-		PQclear(res);
-		return false;
+		success = false;
 	}
 
 	PQclear(res);
-	return true;
+
+	return success;
 }
 
 
@@ -2611,6 +2690,7 @@ delete_node_record(PGconn *conn, int node)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
@@ -2628,19 +2708,20 @@ delete_node_record(PGconn *conn, int node)
 	{
 		log_error(_("unable to delete node record:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
-		return false;
+
+		success = false;
 	}
 
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 bool
 truncate_node_records(PGconn *conn)
 {
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	res = PQexec(conn, "TRUNCATE TABLE repmgr.nodes");
 
@@ -2648,12 +2729,13 @@ truncate_node_records(PGconn *conn)
 	{
 		log_error(_("unable to truncate node record table:\n  %s"),
 				  PQerrorMessage(conn));
-		PQclear(res);
-		return false;
+
+		success = false;
 	}
 
 	PQclear(res);
-	return true;
+
+	return success;
 }
 
 
@@ -2884,8 +2966,7 @@ get_datadir_configuration_files(PGconn *conn, KeyValueList *list)
 
 	for (i = 0; i < PQntuples(res); i++)
 	{
-		key_value_list_set(
-						   list,
+		key_value_list_set(list,
 						   PQgetvalue(res, i, 1),
 						   PQgetvalue(res, i, 0));
 	}
@@ -3110,6 +3191,8 @@ _create_event(PGconn *conn, t_configuration_options *options, int node_id, char
 	char		event_timestamp[MAXLEN] = "";
 	bool		success = true;
 
+	log_verbose(LOG_DEBUG, "_create_event(): event is \"%s\" for node %i", event, node_id);
+
 	/*
 	 * Only attempt to write a record if a connection handle was provided.
 	 * Also check that the repmgr schema has been properly initialised - if
@@ -3620,7 +3703,7 @@ get_slot_record(PGconn *conn, char *slot_name, t_replication_slot *record)
 
 
 int
-get_free_replication_slots(PGconn *conn)
+get_free_replication_slot_count(PGconn *conn)
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
@@ -3657,6 +3740,47 @@ get_free_replication_slots(PGconn *conn)
 }
 
 
+int
+get_inactive_replication_slots(PGconn *conn, KeyValueList *list)
+{
+	PQExpBufferData query;
+	PGresult   *res = NULL;
+	int			i, inactive_slots = 0;
+
+	initPQExpBuffer(&query);
+
+	appendPQExpBuffer(&query,
+					  "   SELECT slot_name, slot_type "
+					  "     FROM pg_catalog.pg_replication_slots "
+					  "    WHERE active IS FALSE "
+					  " ORDER BY slot_name ");
+
+	res = PQexec(conn, query.data);
+	termPQExpBuffer(&query);
+
+	if (PQresultStatus(res) != PGRES_TUPLES_OK)
+	{
+		log_error(_("unable to execute replication slot query"));
+		log_detail("%s", PQerrorMessage(conn));
+		PQclear(res);
+		return -1;
+	}
+
+	inactive_slots = PQntuples(res);
+
+	for (i = 0; i < inactive_slots; i++)
+	{
+		key_value_list_set(list,
+						   PQgetvalue(res, i, 0),
+						   PQgetvalue(res, i, 1));
+	}
+
+	PQclear(res);
+	return inactive_slots;
+}
+
+
+
 /* ==================== */
 /* tablespace functions */
 /* ==================== */
@@ -4277,6 +4401,7 @@ get_last_wal_receive_location(PGconn *conn)
 /* BDR functions */
 /* ============= */
 
+
 static bool
 _is_bdr_db(PGconn *conn, PQExpBufferData *output, bool quiet)
 {
@@ -4287,7 +4412,9 @@ _is_bdr_db(PGconn *conn, PQExpBufferData *output, bool quiet)
 	initPQExpBuffer(&query);
 
 	appendPQExpBuffer(&query,
-					  "SELECT pg_catalog.count(*) FROM pg_catalog.pg_extension WHERE extname='bdr'");
+					  " SELECT (pg_catalog.regexp_matches(extversion, '^\\d+'))[1] AS major_version "
+					  "  FROM pg_catalog.pg_extension "
+					  " WHERE extname = 'bdr' ");
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
@@ -4295,14 +4422,18 @@ _is_bdr_db(PGconn *conn, PQExpBufferData *output, bool quiet)
 	if (PQresultStatus(res) != PGRES_TUPLES_OK || PQntuples(res) == 0)
 	{
 		is_bdr_db = false;
+		bdr_version_num = UNKNOWN_BDR_VERSION_NUM;
 	}
 	else
 	{
-		is_bdr_db = atoi(PQgetvalue(res, 0, 0)) == 1 ? true : false;
+		is_bdr_db = true;
+		bdr_version_num = atoi(PQgetvalue(res, 0, 0));
 	}
 
 	PQclear(res);
 
+	log_verbose(LOG_DEBUG, "BDR ext version number is %i", bdr_version_num);
+
 	if (is_bdr_db == false)
 	{
 		const char *warning = _("BDR extension is not available for this database");
@@ -4315,36 +4446,42 @@ _is_bdr_db(PGconn *conn, PQExpBufferData *output, bool quiet)
 		return is_bdr_db;
 	}
 
-	initPQExpBuffer(&query);
-
-	appendPQExpBuffer(&query,
-					  "SELECT bdr.bdr_is_active_in_db()");
-	res = PQexec(conn, query.data);
-	termPQExpBuffer(&query);
-
-	is_bdr_db = atobool(PQgetvalue(res, 0, 0));
-
-	if (is_bdr_db == false)
+	if (bdr_version_num < 3)
 	{
-		const char *warning = _("BDR extension available for this database, but the database is not configured for BDR");
+		initPQExpBuffer(&query);
 
-		if (output != NULL)
-			appendPQExpBuffer(output, "%s", warning);
-		else if (quiet == false)
-			log_warning("%s", warning);
+		appendPQExpBuffer(&query,
+						  "SELECT bdr.bdr_is_active_in_db()");
+		res = PQexec(conn, query.data);
+		termPQExpBuffer(&query);
+
+		is_bdr_db = atobool(PQgetvalue(res, 0, 0));
+
+		if (is_bdr_db == false)
+		{
+			const char *warning = _("BDR extension available for this database, but the database is not configured for BDR");
+
+			if (output != NULL)
+				appendPQExpBuffer(output, "%s", warning);
+			else if (quiet == false)
+				log_warning("%s", warning);
+		}
+
+		PQclear(res);
 	}
 
-	PQclear(res);
 
 	return is_bdr_db;
 }
 
+
 bool
 is_bdr_db(PGconn *conn, PQExpBufferData *output)
 {
 	return _is_bdr_db(conn, output, false);
 }
 
+
 bool
 is_bdr_db_quiet(PGconn *conn)
 {
@@ -4352,6 +4489,11 @@ is_bdr_db_quiet(PGconn *conn)
 }
 
 
+int
+get_bdr_version_num(void)
+{
+	return bdr_version_num;
+}
 
 bool
 is_active_bdr_node(PGconn *conn, const char *node_name)
@@ -4361,13 +4503,29 @@ is_active_bdr_node(PGconn *conn, const char *node_name)
 	bool		is_active_bdr_node = false;
 
 	initPQExpBuffer(&query);
-	appendPQExpBuffer(&query,
-					  "    SELECT COALESCE(s.active, TRUE) AS active"
-					  "      FROM bdr.bdr_nodes n "
-					  " LEFT JOIN pg_catalog.pg_replication_slots s "
-					  "        ON s.slot_name=bdr.bdr_format_slot_name(n.node_sysid, n.node_timeline, n.node_dboid, (SELECT oid FROM pg_catalog.pg_database WHERE datname = pg_catalog.current_database())) "
-					  "     WHERE n.node_name='%s' ",
-					  node_name);
+
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "    SELECT COALESCE(s.active, TRUE) AS active"
+						  "      FROM bdr.bdr_nodes n "
+						  " LEFT JOIN pg_catalog.pg_replication_slots s "
+						  "        ON s.slot_name=bdr.bdr_format_slot_name(n.node_sysid, n.node_timeline, n.node_dboid, (SELECT oid FROM pg_catalog.pg_database WHERE datname = pg_catalog.current_database())) "
+						  "     WHERE n.node_name='%s' ",
+						  node_name);
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "     SELECT COALESCE(s.active, FALSE) AS active"
+						  "       FROM bdr.node bn "
+						  " INNER JOIN pglogical.node pn "
+						  "         ON (pn.node_id = bn.pglogical_node_id) "
+						  "  LEFT JOIN pg_catalog.pg_replication_slots s "
+						  "         ON s.slot_name=bn.local_slot_name "
+						  "      WHERE pn.node_name='%s' ",
+						  node_name);
+	}
 
 	log_verbose(LOG_DEBUG, "is_active_bdr_node():\n  %s", query.data);
 
@@ -4421,6 +4579,64 @@ is_bdr_repmgr(PGconn *conn)
 }
 
 
+
+/*
+ * Get name of default BDR replication set.
+ *
+ * Caller must free provided value.
+ */
+char *
+get_default_bdr_replication_set(PGconn *conn)
+{
+	PQExpBufferData query;
+	PGresult   *res = NULL;
+	char	   *default_replication_set = NULL;
+	int			namelen;
+
+	if (bdr_version_num < 3)
+	{
+		/* For BDR2, we use a custom replication set */
+		namelen = strlen(BDR2_REPLICATION_SET_NAME);
+		default_replication_set = pg_malloc0(namelen + 1);
+		strncpy(default_replication_set, BDR2_REPLICATION_SET_NAME, namelen);
+
+		return default_replication_set;
+	}
+
+
+	initPQExpBuffer(&query);
+
+	appendPQExpBuffer(&query,
+					  "     SELECT rs.set_name "
+					  "       FROM pglogical.replication_set rs "
+					  " INNER JOIN bdr.node_group ng "
+					  "         ON ng.node_group_default_repset = rs.set_id ");
+
+	res = PQexec(conn, query.data);
+	termPQExpBuffer(&query);
+
+	if (PQresultStatus(res) != PGRES_TUPLES_OK || PQntuples(res) == 0)
+	{
+		log_warning(_("unable to retrieve default BDR replication set name"));
+
+		if (PQresultStatus(res) != PGRES_TUPLES_OK)
+			log_detail("%s", PQerrorMessage(conn));
+
+		PQclear(res);
+		return NULL;
+	}
+
+	namelen = strlen(PQgetvalue(res, 0, 0));
+	default_replication_set = pg_malloc0(namelen + 1);
+
+	strncpy(default_replication_set, PQgetvalue(res, 0, 0), namelen);
+
+	PQclear(res);
+
+	return default_replication_set;
+}
+
+
 bool
 is_table_in_bdr_replication_set(PGconn *conn, const char *tablename, const char *set)
 {
@@ -4430,12 +4646,28 @@ is_table_in_bdr_replication_set(PGconn *conn, const char *tablename, const char
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  "SELECT pg_catalog.count(*) "
-					  "  FROM pg_catalog.unnest(bdr.table_get_replication_sets('repmgr.%s')) AS repset "
-					  " WHERE repset='%s' ",
-					  tablename,
-					  set);
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "SELECT pg_catalog.count(*) "
+						  "  FROM pg_catalog.unnest(bdr.table_get_replication_sets('repmgr.%s')) AS repset "
+						  " WHERE repset='%s' ",
+						  tablename,
+						  set);
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "     SELECT pg_catalog.count(*) "
+						  "       FROM pglogical.replication_set s "
+						  " INNER JOIN pglogical.replication_set_table st "
+						  "         ON s.set_id = st.set_id "
+						  "      WHERE s.set_name = '%s' "
+						  "        AND st.set_reloid = 'repmgr.%s'::REGCLASS ",
+						  set,
+						  tablename);
+
+	}
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
@@ -4461,32 +4693,44 @@ add_table_to_bdr_replication_set(PGconn *conn, const char *tablename, const char
 {
 	PQExpBufferData query;
 	PGresult   *res = NULL;
+	bool		success = true;
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  "SELECT bdr.table_set_replication_sets('repmgr.%s', '{%s}')",
-					  tablename,
-					  set);
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "SELECT bdr.table_set_replication_sets('repmgr.%s', '{%s}')",
+						  tablename,
+						  set);
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  " SELECT bdr.replication_set_add_table( "
+						  "   relation := 'repmgr.%s', "
+						  "   set_name := '%s' "
+						  " ) ",
+						  tablename,
+						  set);
+	}
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
 
 	if (PQresultStatus(res) != PGRES_TUPLES_OK)
 	{
-		log_error(_("unable to add table \"repmgr.%s\" to replication set \"%s\":\n  %s"),
+		log_error(_("unable to add table \"repmgr.%s\" to replication set \"%s\""),
 				  tablename,
-				  set,
-				  PQerrorMessage(conn));
+				  set);
+		log_detail("%s", PQerrorMessage(conn));
 
-		if (res != NULL)
-			PQclear(res);
-
-		return false;
+		success = false;
 	}
+
 	PQclear(res);
 
-	return true;
+	return success;
 }
 
 
@@ -4499,8 +4743,16 @@ bdr_node_name_matches(PGconn *conn, const char *node_name, PQExpBufferData *bdr_
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  "SELECT bdr.bdr_get_local_node_name() AS node_name");
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "SELECT bdr.bdr_get_local_node_name() AS node_name");
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "SELECT node_name FROM bdr.local_node_info()");
+	}
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
@@ -4531,21 +4783,36 @@ get_bdr_node_replication_slot_status(PGconn *conn, const char *node_name)
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  " SELECT s.active "
-					  "   FROM pg_catalog.pg_replication_slots s "
-					  "  WHERE slot_name = "
-					  "    (SELECT bdr.bdr_format_slot_name(node_sysid, node_timeline, node_dboid, datoid) "
-					  "   FROM bdr.bdr_nodes "
-					  " WHERE node_name = '%s') ",
-					  node_name);
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  " SELECT s.active "
+						  "   FROM pg_catalog.pg_replication_slots s "
+						  "  WHERE slot_name = "
+						  "    (SELECT bdr.bdr_format_slot_name(node_sysid, node_timeline, node_dboid, datoid) "
+						  "   FROM bdr.bdr_nodes "
+						  " WHERE node_name = '%s') ",
+						  node_name);
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "     SELECT COALESCE(s.active, FALSE) AS active"
+						  "       FROM bdr.node bn "
+						  " INNER JOIN pglogical.node pn "
+						  "         ON (pn.node_id = bn.pglogical_node_id) "
+						  " INNER JOIN pg_catalog.pg_replication_slots s "
+						  "         ON s.slot_name=bn.local_slot_name "
+						  "      WHERE pn.node_name='%s' ",
+						  node_name);
+	}
 
 	log_verbose(LOG_DEBUG, "get_bdr_node_replication_slot_status():\n  %s", query.data);
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
 
-	if (PQresultStatus(res) != PGRES_TUPLES_OK)
+	if (PQresultStatus(res) != PGRES_TUPLES_OK || PQntuples(res) == 0)
 	{
 		status = SLOT_UNKNOWN;
 	}
@@ -4596,6 +4863,9 @@ get_bdr_other_node_name(PGconn *conn, int node_id, char *node_name)
 }
 
 
+/*
+ * For BDR 2.x only
+ */
 void
 add_extension_tables_to_bdr_replication_set(PGconn *conn)
 {
@@ -4617,7 +4887,7 @@ add_extension_tables_to_bdr_replication_set(PGconn *conn)
 
 	if (PQresultStatus(res) != PGRES_TUPLES_OK)
 	{
-		/* */
+		/* XXX log error */
 	}
 	else
 	{
@@ -4625,8 +4895,7 @@ add_extension_tables_to_bdr_replication_set(PGconn *conn)
 
 		for (i = 0; i < PQntuples(res); i++)
 		{
-			add_table_to_bdr_replication_set(
-											 conn,
+			add_table_to_bdr_replication_set(conn,
 											 PQgetvalue(res, i, 0),
 											 "repmgr");
 		}
@@ -4645,10 +4914,20 @@ get_all_bdr_node_records(PGconn *conn, BdrNodeInfoList *node_list)
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  "  SELECT " BDR_NODES_COLUMNS
-					  "    FROM bdr.bdr_nodes "
-					  "ORDER BY node_seq_id ");
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "  SELECT " BDR2_NODES_COLUMNS
+						  "    FROM bdr.bdr_nodes "
+						  "ORDER BY node_seq_id ");
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "     SELECT " BDR3_NODES_COLUMNS
+						  "       FROM bdr.node_summary ns "
+						  "   ORDER BY node_name");
+	}
 
 	log_verbose(LOG_DEBUG, "get_all_node_records():\n%s", query.data);
 
@@ -4669,11 +4948,22 @@ get_bdr_node_record_by_name(PGconn *conn, const char *node_name, t_bdr_node_info
 
 	initPQExpBuffer(&query);
 
-	appendPQExpBuffer(&query,
-					  "  SELECT " BDR_NODES_COLUMNS
-					  "    FROM bdr.bdr_nodes "
-					  "   WHERE node_name = '%s'",
-					  node_name);
+	if (bdr_version_num < 3)
+	{
+		appendPQExpBuffer(&query,
+						  "  SELECT " BDR2_NODES_COLUMNS
+						  "    FROM bdr.bdr_nodes "
+						  "   WHERE node_name = '%s'",
+						  node_name);
+	}
+	else
+	{
+		appendPQExpBuffer(&query,
+						  "     SELECT " BDR3_NODES_COLUMNS
+						  "       FROM bdr.node_summary ns "
+						  "      WHERE ns.node_name = '%s'",
+						  node_name);
+	}
 
 	log_verbose(LOG_DEBUG, "get_bdr_node_record_by_name():\n%s", query.data);
 
@@ -4743,16 +5033,12 @@ _populate_bdr_node_records(PGresult *res, BdrNodeInfoList *node_list)
 static void
 _populate_bdr_node_record(PGresult *res, t_bdr_node_info *node_info, int row)
 {
-	char		buf[MAXLEN] = "";
-
 	strncpy(node_info->node_sysid, PQgetvalue(res, row, 0), MAXLEN);
 	node_info->node_timeline = atoi(PQgetvalue(res, row, 1));
 	node_info->node_dboid = atoi(PQgetvalue(res, row, 2));
-	strncpy(buf, PQgetvalue(res, row, 3), MAXLEN);
-	node_info->node_status = buf[0];
-	strncpy(node_info->node_name, PQgetvalue(res, row, 4), MAXLEN);
-	strncpy(node_info->node_local_dsn, PQgetvalue(res, row, 5), MAXLEN);
-	strncpy(node_info->node_init_from_dsn, PQgetvalue(res, row, 6), MAXLEN);
+	strncpy(node_info->node_name, PQgetvalue(res, row, 3), MAXLEN);
+	strncpy(node_info->node_local_dsn, PQgetvalue(res, row, 4), MAXLEN);
+	strncpy(node_info->peer_state_name, PQgetvalue(res, row, 5), MAXLEN);
 }
 
 
@@ -4807,13 +5093,17 @@ bdr_node_has_repmgr_set(PGconn *conn, const char *node_name)
 	PGresult   *res = NULL;
 	bool		has_repmgr_set = false;
 
+	if (bdr_version_num >= 3)
+		return true;
+
 	initPQExpBuffer(&query);
 
 	appendPQExpBuffer(&query,
 					  " SELECT pg_catalog.count(*) "
 					  "   FROM pg_catalog.unnest(bdr.connection_get_replication_sets('%s') AS repset "
-					  "  WHERE repset = 'repmgr'",
-					  node_name);
+					  "  WHERE repset = '%s'",
+					  node_name,
+					  BDR2_REPLICATION_SET_NAME);
 
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
@@ -4840,26 +5130,40 @@ bdr_node_set_repmgr_set(PGconn *conn, const char *node_name)
 	PGresult   *res = NULL;
 	bool		success = true;
 
+	if (bdr_version_num >= 3)
+		return true;
+
 	initPQExpBuffer(&query);
 
+	/*
+	 * Here we extract a list of existing replication sets, add 'repmgr', and
+	 * set the replication sets to the new list.
+	 */
 	appendPQExpBuffer(&query,
 					  " SELECT bdr.connection_set_replication_sets( "
 					  "   ARRAY( "
 					  "     SELECT repset::TEXT "
 					  "       FROM pg_catalog.unnest(bdr.connection_get_replication_sets('%s')) AS repset "
 					  "         UNION "
-					  "     SELECT 'repmgr'::TEXT "
+					  "     SELECT '%s'::TEXT "
 					  "   ), "
 					  "   '%s' "
 					  " ) ",
 					  node_name,
+					  BDR2_REPLICATION_SET_NAME,
 					  node_name);
 
+	log_debug("bdr_node_set_repmgr_set():\n%s", query.data);
+
 	res = PQexec(conn, query.data);
 	termPQExpBuffer(&query);
 
+
 	if (PQresultStatus(res) != PGRES_TUPLES_OK)
 	{
+		log_debug("result status: %s", PQresStatus(PQresultStatus(res)));
+		log_error(_("unable to create replication set \"repmgr\""));
+		log_detail("%s", PQerrorMessage(conn));
 		success = false;
 	}
 

dbutils.h

@@ -29,7 +29,9 @@
 #include "voting.h"
 
 #define REPMGR_NODES_COLUMNS "n.node_id, n.type, n.upstream_node_id, n.node_name, n.conninfo, n.repluser, n.slot_name, n.location, n.priority, n.active, n.config_file, '' AS upstream_node_name "
-#define BDR_NODES_COLUMNS "node_sysid, node_timeline, node_dboid, node_status, node_name, node_local_dsn, node_init_from_dsn, node_read_only, node_seq_id"
+#define BDR2_NODES_COLUMNS "node_sysid, node_timeline, node_dboid, node_name, node_local_dsn, ''"
+#define BDR3_NODES_COLUMNS "ns.node_id, 0, 0, ns.node_name, ns.interface_connstr, ns.peer_state_name"
+
 
 #define ERRBUFF_SIZE 512
 
@@ -94,6 +96,14 @@ typedef enum
 	SLOT_ACTIVE
 } ReplSlotStatus;
 
+typedef enum
+{
+	BACKUP_STATE_UNKNOWN = -1,
+	BACKUP_STATE_IN_BACKUP,
+	BACKUP_STATE_NO_BACKUP
+} BackupState;
+
+
 /*
  * Struct to store node information
  */
@@ -237,18 +247,14 @@ typedef struct s_bdr_node_info
 	char		node_sysid[MAXLEN];
 	uint32		node_timeline;
 	uint32		node_dboid;
-	char		node_status;
 	char		node_name[MAXLEN];
 	char		node_local_dsn[MAXLEN];
-	char		node_init_from_dsn[MAXLEN];
-	bool		read_only;
-	uint32		node_seq_id;
+	char		peer_state_name[MAXLEN];
 } t_bdr_node_info;
 
 #define T_BDR_NODE_INFO_INITIALIZER { \
 	"", InvalidOid, InvalidOid, \
-	'?', "", "", "", \
-    false, -1 \
+    "", "", "" \
 }
 
 
@@ -357,7 +363,7 @@ void		close_connection(PGconn **conn);
 
 /* conninfo manipulation functions */
 bool		get_conninfo_value(const char *conninfo, const char *keyword, char *output);
-
+bool		get_conninfo_default_value(const char *param, char *output, int maxlen);
 void		initialize_conninfo_params(t_conninfo_param_list *param_list, bool set_defaults);
 void		free_conninfo_params(t_conninfo_param_list *param_list);
 void		copy_conninfo_params(t_conninfo_param_list *dest_list, t_conninfo_param_list *source_list);
@@ -369,6 +375,7 @@ bool		parse_conninfo_string(const char *conninfo_str, t_conninfo_param_list *par
 char	   *param_list_to_string(t_conninfo_param_list *param_list);
 bool		has_passfile(void);
 
+
 /* transaction functions */
 bool		begin_transaction(PGconn *conn);
 bool		commit_transaction(PGconn *conn);
@@ -391,6 +398,7 @@ int			get_ready_archive_files(PGconn *conn, const char *data_directory);
 bool		identify_system(PGconn *repl_conn, t_system_identification *identification);
 bool		repmgrd_set_local_node_id(PGconn *conn, int local_node_id);
 int			repmgrd_get_local_node_id(PGconn *conn);
+BackupState	server_in_exclusive_backup_mode(PGconn *conn);
 
 /* extension functions */
 ExtensionStatus get_repmgr_extension_status(PGconn *conn);
@@ -454,7 +462,8 @@ void		create_slot_name(char *slot_name, int node_id);
 bool		create_replication_slot(PGconn *conn, char *slot_name, int server_version_num, PQExpBufferData *error_msg);
 bool		drop_replication_slot(PGconn *conn, char *slot_name);
 RecordStatus get_slot_record(PGconn *conn, char *slot_name, t_replication_slot *record);
-int			get_free_replication_slots(PGconn *conn);
+int			get_free_replication_slot_count(PGconn *conn);
+int			get_inactive_replication_slots(PGconn *conn, KeyValueList *list);
 
 /* tablespace functions */
 bool		get_tablespace_name_by_location(PGconn *conn, const char *location, char *name);
@@ -505,12 +514,14 @@ void		get_node_replication_stats(PGconn *conn, int server_version_num, t_node_in
 bool		is_downstream_node_attached(PGconn *conn, char *node_name);
 
 /* BDR functions */
+int			get_bdr_version_num(void);
 void		get_all_bdr_node_records(PGconn *conn, BdrNodeInfoList *node_list);
 RecordStatus get_bdr_node_record_by_name(PGconn *conn, const char *node_name, t_bdr_node_info *node_info);
 bool		is_bdr_db(PGconn *conn, PQExpBufferData *output);
 bool		is_bdr_db_quiet(PGconn *conn);
 bool		is_active_bdr_node(PGconn *conn, const char *node_name);
 bool		is_bdr_repmgr(PGconn *conn);
+char	   *get_default_bdr_replication_set(PGconn *conn);
 bool		is_table_in_bdr_replication_set(PGconn *conn, const char *tablename, const char *set);
 bool		add_table_to_bdr_replication_set(PGconn *conn, const char *tablename, const char *set);
 void		add_extension_tables_to_bdr_replication_set(PGconn *conn);

doc/appendix-packages.sgml

@@ -41,18 +41,19 @@
       <title>CentOS repositories</title>
 
       <para>
-        &repmgr; packages are available from the 2ndQuadrant repository, and also the PostgreSQL
-        community repository. The 2ndQuadrant repository is updated immediately after each
+        &repmgr; packages are available from the public 2ndQuadrant repository, and also the
+        PostgreSQL community repository. The 2ndQuadrant repository is updated immediately
+        after each
         &repmgr; release.
       </para>
 
       <table id="centos-2ndquadrant-repository">
-        <title>2ndQuadrant repository</title>
+        <title>2ndQuadrant public repository</title>
         <tgroup cols="2">
           <tbody>
             <row>
               <entry>Repository URL:</entry>
-              <entry><ulink url="http://packages.2ndquadrant.com/repmgr/">http://packages.2ndquadrant.com/repmgr/</ulink></entry>
+              <entry><ulink url="https://rpm.2ndquadrant.com/">https://rpm.2ndquadrant.com/</ulink></entry>
             </row>
             <row>
               <entry>Repository documentation:</entry>
@@ -363,4 +364,48 @@
    </sect2>
 
   </sect1>
+
+
+  <sect1 id="packages-packager-info" xreflabel="Information for packagers">
+    <title>Information for packagers</title>
+    <indexterm>
+      <primary>packages</primary>
+      <secondary>information for packagers</secondary>
+    </indexterm>
+    <para>
+      We recommend patching the following  parameters when
+      building the package as built-in default values for user convenience.
+      These values can nevertheless be overridden by the user, if desired.
+    </para>
+    <itemizedlist>
+      <listitem>
+        <para>
+          Configuration file location: the default configuration file location
+          can be hard-coded by patching <varname>package_conf_file</varname>
+          in <filename>configfile.c</filename>:
+          <programlisting>
+		/* packagers: if feasible, patch configuration file path into "package_conf_file" */
+		char		package_conf_file[MAXPGPATH] = "";</programlisting>
+        </para>
+        <para>
+          See also: <xref linkend="configuration-file">
+        </para>
+      </listitem>
+
+      <listitem>
+        <para>
+          PID file location: the default <application>repmgrd</application> PID file
+          location can be hard-coded by patching <varname>package_pid_file</varname>
+          in <filename>repmgrd.c</filename>:
+          <programlisting>
+		/* packagers: if feasible, patch PID file path into "package_pid_file" */
+		char		package_pid_file[MAXPGPATH] = "";</programlisting>
+        </para>
+        <para>
+          See also: <xref linkend="repmgrd-pid-file">
+        </para>
+      </listitem>
+
+    </itemizedlist>
+  </sect1>
 </appendix>

doc/appendix-release-notes.sgml

@@ -15,6 +15,310 @@
     See also: <xref linkend="upgrading-repmgr">
   </para>
 
+  <sect1 id="release-4.1.0">
+    <title>Release 4.1.0</title>
+    <para><emphasis>???? ??, 2018</emphasis></para>
+    <para>
+      &repmgr; 4.1.0 introduces some changes to <application>repmgrd</application>
+      behaviour and some additional configuration parameters.
+    </para>
+    <para>
+      This release can be installed as a simple package upgrade from repmgr 4.0 ~ 4.0.6.
+      The following post-upgrade steps must be carried out:
+
+       <itemizedlist>
+          <listitem>
+            <para>
+              <application>repmgrd</application> (if running) must be restarted.
+            </para>
+          </listitem>
+          <listitem>
+            <para>
+              Execute <command>ALTER EXTENSION repmgr UPDATE</command>
+              on the primary server in the database where &repmgr; is installed.
+            </para>
+          </listitem>
+       </itemizedlist>
+
+       A restart of the PostgreSQL server is <emphasis>not</emphasis> required
+       for this release.
+    </para>
+    <para>
+       See <xref linkend="upgrading-repmgr-extension"> for more details.
+    </para>
+
+    <para>
+      Configuration changes are backwards-compatible and no changes to
+      <filename>repmgr.conf</filename> are required. However users should
+      review the changes listed below.
+    </para>
+
+    <sect2>
+      <title>Configuration file changes</title>
+
+      <para>
+        <itemizedlist>
+
+          <listitem>
+            <para>
+              Default for <xref linkend="repmgr-conf-log-level"> is now <option>INFO</option>.
+              This produces additional informative log output, without creating excessive additional
+              log file volume, and matches the setting assumed for examples in the documentation.
+              (GitHub #470).
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <varname>recovery_min_apply_delay</varname> now accepts a minimum value
+              of <literal>zero</literal> (GitHub #448).
+            </para>
+          </listitem>
+
+        </itemizedlist>
+      </para>
+
+    </sect2>
+
+
+    <sect2>
+      <title>repmgr enhancements</title>
+      <para>
+        <itemizedlist>
+
+          <listitem>
+            <para>
+              <application>repmgr</application>: always exit with an error if an unrecognised
+              command line option is provided. This matches the behaviour of other PostgreSQL
+              utilities such as <application>psql</application>. (GitHub #464).
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <application>repmgr</application>: add <option>-q/--quiet</option> option to suppress non-error
+              output. (GitHub #468).
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-cluster-show">repmgr cluster show</link></command>,
+              <command><link linkend="repmgr-node-check">repmgr node check</link></command> and
+              <command><link linkend="repmgr-node-status">repmgr node status</link></command>
+              return non-zero exit code if node status issues detected. (GitHub #456).
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              Add <option>--csv</option> output option for
+              <command><link linkend="repmgr-cluster-event">repmgr cluster event</link></command>.
+              (GitHub #471).
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-witness-unregister">repmgr witness unregister</link></command>
+			  can be run on any node, by providing the ID of the witness node with <option>--node-id</option>.
+			  (GitHub #472).
+			</para>
+		  </listitem>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-standby-switchover">repmgr standby switchover</link></command>
+              will refuse to run if an exclusive backup is taking place on the current primary.
+			  (GitHub #476).
+			</para>
+		  </listitem>
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
+    <sect2>
+      <title>repmgrd enhancements</title>
+      <para>
+        <itemizedlist>
+
+          <listitem>
+            <para>
+              <application>repmgrd</application>: create a PID file by default
+              (GitHub #457). For details, see <xref linkend="repmgrd-pid-file">.
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <application>repmgrd</application>: daemonize process by default.
+			  In case, for whatever reason, the user does not wish to daemonize the
+			  process, provide <option>--daemonize=false</option>.
+			  (GitHub #458).
+            </para>
+          </listitem>
+
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
+
+    <sect2>
+      <title>Bug fixes</title>
+      <para>
+        <itemizedlist>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-standby-register">repmgr standby register --wait-sync</link></command>:
+              fix behaviour when no timeout provided.
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-cluster-cleanup">repmgr cluster cleanup</link></command>:
+			  add missing help options. (GitHub #461/#462).
+			</para>
+		  </listitem>
+
+          <listitem>
+            <para>
+			  Ensure witness node follows new primary after switchover. (GitHub #453).
+			</para>
+		  </listitem>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-node-check">repmgr node check</link></command> and
+              <command><link linkend="repmgr-node-status">repmgr node status</link></command>:
+			  fix witness node handling. (GitHub #451).
+			</para>
+		  </listitem>
+
+          <listitem>
+            <para>
+              When using <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>
+              with <option>--recovery-conf-only</option> and replication slots, ensure
+              <varname>primary_slot_name</varname> is set correctly. (GitHub #474).
+            </para>
+          </listitem>
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
+  </sect1>
+
+  <sect1 id="release-4.0.6">
+    <title>Release 4.0.6</title>
+    <para><emphasis>June 14, 2018</emphasis></para>
+    <para>
+	  &repmgr; 4.0.6 contains a number of bug fixes and usability enhancements.
+    </para>
+	<para>
+	  We recommend upgrading to this version as soon as possible.
+	  This release can be installed as a simple package upgrade from repmgr 4.0 ~ 4.0.5;
+      <application>repmgrd</application> (if running) should be restarted. See <xref linkend="upgrading-repmgr">
+      for more details.
+	</para>
+
+    <sect2>
+      <title>Usability enhancements</title>
+
+      <para>
+        <itemizedlist>
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-cluster-crosscheck">repmgr cluster crosscheck</link></command> and
+              <command><link linkend="repmgr-cluster-matrix">repmgr cluster matrix</link></command>:
+              return non-zero exit code if node connection issues detected (GitHub #447)
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
+			   Improve handling of external configuration file copying,  including consideration in
+			   <option>--dry-run</option> check
+			   (GitHub #443)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              When using <option>--dry-run</option>, force log level to <literal>INFO</literal>
+			  to ensure output will always be displayed
+			  (GitHub #441)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
+			   Improve documentation of <option>--recovery-conf-only</option> mode
+			   (GitHub #438)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
+			   Don't require presence of <varname>user</varname> parameter in conninfo string
+			   (GitHub #437)
+           </para>
+          </listitem>
+
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
+    <sect2>
+      <title>Bug fixes</title>
+      <para>
+
+        <itemizedlist>
+
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-witness-register">repmgr witness register</link></command>:
+              prevent registration of a witness server with the same name as an existing node
+           </para>
+          </listitem>
+
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-standby-follow">repmgr standby follow</link></command>:
+			   check node has actually connected to new primary before reporting success
+			   (GitHub #444)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-node-rejoin">repmgr node rejoin</link></command>:
+			   Fix bug when parsing <option>--config-files</option> parameter
+			   (GitHub #442)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              <application>repmgrd</application>: ensure local node is counted as quorum member
+			  (GitHub #439)
+           </para>
+          </listitem>
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
+  </sect1>
+
   <sect1 id="release-4.0.5">
     <title>Release 4.0.5</title>
     <para><emphasis>Wed May 2, 2018</emphasis></para>
@@ -24,6 +328,7 @@
       generation and (in <application>repmgrd</application>) handling of various
       corner-case situations, as well as a number of bug fixes.
     </para>
+
     <sect2>
       <title>Usability enhancements</title>
 
@@ -32,7 +337,7 @@
           <listitem>
             <para>
               Various documentation improvements, with particular emphasis on
-              the importance of setting appropriate <link linkend="configuration-service-commands">service commands</link>
+              the importance of setting appropriate <link linkend="configuration-file-service-commands">service commands</link>
               instead of relying on <application>pg_ctl</application>.
             </para>
           </listitem>

doc/appendix-signatures.sgml

@@ -33,34 +33,5 @@
 
  </sect1>
 
- <sect1 id="repmgr-rpm-key" xreflabel="repmgr rpm key">
-   <title>repmgr RPM signing key</title>
-   <para>
-     The signing key ID used for <application>repmgr</application> source code bundles is:
-     <ulink url="http://packages.2ndquadrant.com/repmgr/RPM-GPG-KEY-repmgr">
-       <literal>0x702D883A</literal></ulink>.
-   </para>
-
-   <para>
-     To download the <application>repmgr</application> source key to your computer:
-     <programlisting>
-       curl -s http://packages.2ndquadrant.com/repmgr/RPM-GPG-KEY-repmgr | gpg --import
-       gpg --fingerprint 0x702D883A
-     </programlisting>
-     then verify that the fingerprint is the expected value:
-     <programlisting>
-       AE4E 390E A58E 0037 6148  3F29 888D 018B 702D 883A</programlisting>
-   </para>
-
-   <para>
-     To check a repository RPM, use <application>rpmkeys</application> to load the
-      packaging signing key into the RPM database then use <literal>rpm -K</literal>, e.g.:
-     <programlisting>
-       sudo rpmkeys --import http://packages.2ndquadrant.com/repmgr/RPM-GPG-KEY-repmgr
-       rpm -K postgresql-bdr94-2ndquadrant-redhat-1.0-2.noarch.rpm
-     </programlisting>
-   </para>
-
- </sect1>
 
 </appendix>

doc/configuration-file-log-settings.sgml

@@ -0,0 +1,107 @@
+<sect1 id="configuration-file-log-settings" xreflabel="log settings">
+  <indexterm>
+    <primary>repmgr.conf</primary>
+    <secondary>log settings</secondary>
+  </indexterm>
+  <indexterm>
+    <primary>log settings</primary>
+    <secondary>configuration in repmgr.conf</secondary>
+  </indexterm>
+  <title>Log settings</title>
+
+  <para>
+    By default, &repmgr; and <application>repmgrd</application> write log output to
+    <literal>STDERR</literal>. An alternative log destination can be specified
+    (either a file or <literal>syslog</literal>).
+  </para>
+
+  <note>
+    <para>
+      The &repmgr; application itself will continue to write log output to <literal>STDERR</literal>
+      even if another log destination is configured, as otherwise any output resulting from a command
+      line operation will "disappear" into the log.
+    </para>
+    <para>
+      This behaviour can be overriden with the command line option <option>--log-to-file</option>,
+      which will redirect all logging output to the configured log destination. This is recommended
+      when &repmgr; is executed by another application, particularly <application>repmgrd</application>,
+      to enable log output generated by the &repmgr; application to be stored for later reference.
+    </para>
+  </note>
+
+  <variablelist>
+
+   <varlistentry id="repmgr-conf-log-level" xreflabel="log_level">
+    <term><varname>log_level</varname> (<type>string</type>)
+     <indexterm>
+      <primary><varname>log_level</varname> configuration file parameter</primary>
+     </indexterm>
+    </term>
+    <listitem>
+     <para>
+       One of <option>DEBUG</option>, <option>INFO</option>, <option>NOTICE</option>,
+       <option>WARNING</option>, <option>ERROR</option>, <option>ALERT</option>, <option>CRIT</option>
+       or <option>EMERG</option>.
+     </para>
+     <para>
+       Default is <option>INFO</option>.
+     </para>
+     <para>
+       Note that <option>DEBUG</option> will produce a substantial amount of log output
+       and should not be enabled in normal use.
+     </para>
+    </listitem>
+   </varlistentry>
+
+   <varlistentry id="repmgr-conf-log-facility" xreflabel="log_facility">
+    <term><varname>log_facility</varname> (<type>string</type>)
+     <indexterm>
+      <primary><varname>log_facility</varname> configuration file parameter</primary>
+     </indexterm>
+    </term>
+    <listitem>
+     <para>
+       Logging facility: possible values are <option>STDERR</option> (default), or for
+       syslog integration, one of <option>LOCAL0</option>, <option>LOCAL1</option>, <option>...</option>,
+       <option>LOCAL7</option>, <option>USER</option>.
+     </para>
+    </listitem>
+   </varlistentry>
+
+   <varlistentry id="repmgr-conf-log-file" xreflabel="log_file">
+    <term><varname>log_file</varname> (<type>string</type>)
+     <indexterm>
+      <primary><varname>log_file</varname> configuration file parameter</primary>
+     </indexterm>
+    </term>
+    <listitem>
+     <para>
+       If <xref linkend="repmgr-conf-log-facility"> is set to <option>STDERR</option>, log output
+       can be redirected to the specified file.
+     </para>
+     <para>
+       See <xref linkend="repmgrd-log-rotation"> for information on configuring log rotation.
+     </para>
+    </listitem>
+   </varlistentry>
+
+   <varlistentry id="repmgr-conf-log-status-interval" xreflabel="log_status_interval">
+    <term><varname>log_status_interval</varname> (<type>integer</type>)
+     <indexterm>
+      <primary><varname>log_status_interval</varname> configuration file parameter</primary>
+     </indexterm>
+    </term>
+    <listitem>
+     <para>
+       This setting causes <application>repmgrd</application> to emit a status log
+       line at the specified interval (in seconds, default <literal>300</literal>)
+       describing <application>repmgrd</application>'s current state, e.g.:
+     </para>
+     <programlisting>
+      [2018-07-12 00:47:32] [INFO] monitoring connection to upstream node "node1" (node ID: 1)</programlisting>
+    </listitem>
+   </varlistentry>
+
+  </variablelist>
+
+</sect1>

doc/configuration-file-required-settings.sgml

@@ -1,10 +1,10 @@
-<sect1 id="configuration-file-settings" xreflabel="configuration file settings">
+<sect1 id="configuration-file-settings" xreflabel="required configuration file settings">
   <indexterm>
     <primary>repmgr.conf</primary>
-    <secondary>basic settings</secondary>
+    <secondary>required settings</secondary>
   </indexterm>
 
- <title>Basic configuration file settings</title>
+ <title>Required configuration file settings</title>
  <para>
    Each <filename>repmgr.conf</filename> file must contain the following parameters:
  </para>

doc/configuration-file-service-commands.sgml

@@ -1,4 +1,4 @@
-<sect1 id="configuration-service-commands" xreflabel="service command settings">
+<sect1 id="configuration-file-service-commands" xreflabel="service command settings">
   <indexterm>
     <primary>repmgr.conf</primary>
     <secondary>service command settings</secondary>
@@ -50,10 +50,18 @@
 
   <note>
     <para>
-      It's also possible to specify a <varname>service_promote_command</varname>;
-      this overrides any value contained in the setting <varname>promote_command</varname>.
+      It's also possible to specify a <varname>service_promote_command</varname>.
       This is intended for systems which provide a package-level promote command,
-      such as Debian's <application>pg_ctlcluster</application>.
+      such as Debian's <application>pg_ctlcluster</application>, to promote the
+      PostgreSQL from standby to primary.
+    </para>
+    <para>
+      If your packaging system does not provide such a command, it can be left empty,
+      and &repmgr; will generate the appropriate `pg_ctl ... promote` command.
+    </para>
+    <para>
+      Do not confuse this with <varname>promote_command</varname>, which is used
+      by <application>repmgrd</application> to execute <xref linkend="repmgr-standby-promote">.
     </para>
   </note>
 

doc/configuration.sgml

@@ -2,16 +2,17 @@
   <title>repmgr configuration</title>
 
   &configuration-file;
-  &configuration-file-settings;
-  &configuration-service-commands;
+  &configuration-file-required-settings;
+  &configuration-file-log-settings;
+  &configuration-file-service-commands;
 
-  <sect1 id="configuration-permissions" xreflabel="User permissions">
+  <sect1 id="configuration-permissions" xreflabel="Database user permissions">
     <indexterm>
       <primary>configuration</primary>
-      <secondary>user permissions</secondary>
+      <secondary>database user permissions</secondary>
     </indexterm>
 
-    <title>repmgr user permissions</title>
+    <title>repmgr database user permissions</title>
     <para>
       &repmgr; will create an extension database containing objects
       for administering &repmgr; metadata. The user defined in the <varname>conninfo</varname>

doc/event-notifications.sgml

@@ -206,7 +206,7 @@
     <simpara><literal>repmgrd_failover_follow</literal></simpara>
    </listitem>
    <listitem>
-     <simpara><literal>repmgrd_failover_aborted</literal></simpara>
+    <simpara><literal>repmgrd_failover_aborted</literal></simpara>
    </listitem>
    <listitem>
     <simpara><literal>repmgrd_upstream_disconnect</literal></simpara>
@@ -217,9 +217,6 @@
    <listitem>
     <simpara><literal>repmgrd_promote_error</literal></simpara>
    </listitem>
-   <listitem>
-    <simpara><literal>repmgrd_failover_promote</literal></simpara>
-   </listitem>
    <listitem>
     <simpara><literal>bdr_failover</literal></simpara>
    </listitem>

doc/filelist.sgml

@@ -38,8 +38,9 @@
 <!ENTITY quickstart      SYSTEM "quickstart.sgml">
 <!ENTITY configuration      SYSTEM "configuration.sgml">
 <!ENTITY configuration-file      SYSTEM "configuration-file.sgml">
-<!ENTITY configuration-file-settings      SYSTEM "configuration-file-settings.sgml">
-<!ENTITY configuration-service-commands   SYSTEM "configuration-service-commands.sgml">
+<!ENTITY configuration-file-required-settings      SYSTEM "configuration-file-required-settings.sgml">
+<!ENTITY configuration-file-log-settings      SYSTEM "configuration-file-log-settings.sgml">
+<!ENTITY configuration-file-service-commands   SYSTEM "configuration-file-service-commands.sgml">
 <!ENTITY cloning-standbys  SYSTEM "cloning-standbys.sgml">
 <!ENTITY promoting-standby  SYSTEM "promoting-standby.sgml">
 <!ENTITY follow-new-primary  SYSTEM "follow-new-primary.sgml">

doc/install-packages.sgml

@@ -5,26 +5,27 @@
   system.
  </para>
 
- <sect2 id="installation-packages-redhat" xreflabel="Installing from packages on RHEL, Fedora and CentOS">
+ <sect2 id="installation-packages-redhat" xreflabel="Installing from packages on RHEL, CentOS and Fedora">
 
   <indexterm>
    <primary>installation</primary>
    <secondary>on Red Hat/CentOS/Fedora etc.</secondary>
   </indexterm>
 
-  <title>RedHat/Fedora/CentOS</title>
+  <title>RedHat/CentOS/Fedora</title>
   <para>
-   RPM packages for &repmgr; are available via Yum through
+	&repmgr; RPM packages for RedHat/CentOS variants and Fedora are available from the
+	<ulink url="https://2ndquadrant.com">2ndQuadrant</ulink>
+	<ulink url="https://rpm.2ndquadrant.com/">public RPM repository</ulink>; see following
+	section for details.
+  </para>
+  <para>
+   RPM packages for &repmgr; are also available via Yum through
    the PostgreSQL Global Development Group RPM repository
    (<ulink url="https://yum.postgresql.org/">http://yum.postgresql.org/</ulink>).
    Follow the instructions for your distribution (RedHat, CentOS,
-   Fedora, etc.) and architecture as detailed there.
-  </para>
-  <para>
-   <ulink url="https://2ndquadrant.com">2ndQuadrant</ulink> also provides its
-   own RPM packages which are made available
-   at the same time as each &repmgr; release, as it can take some days for
-   them to become available via the main PGDG repository. See following section for details:
+   Fedora, etc.) and architecture as detailed there. Note that it can take some days
+   for new &repmgr; packages to become available via the this repository.
   </para>
   <note>
     <para>
@@ -37,65 +38,74 @@
 
   <para>
     For more information on the package contents, including details of installation
-    paths and relevant <link linkend="configuration-service-commands">service commands</link>,
+    paths and relevant <link linkend="configuration-file-service-commands">service commands</link>,
     see the appendix section <xref linkend="packages-centos">.
   </para>
 
 
   <sect3 id="installation-packages-redhat-2ndq">
-    <title>2ndQuadrant repmgr yum repository</title>
+    <title>2ndQuadrant public RPM yum repository</title>
+
+	<note>
+	  <para>
+		<ulink url="https://2ndquadrant.com">2ndQuadrant</ulink> previously provided a dedicated
+        &repmgr; repository at
+        <ulink url="http://packages.2ndquadrant.com/repmgr/">http://packages.2ndquadrant.com/repmgr/</ulink>.
+		This repository will be deprecated in a future release as it is now replaced by
+		the <ulink url="https://rpm.2ndquadrant.com/">public RPM repository</ulink>
+		documented below.
+	  </para>
+	</note>
+
     <para>
-      Beginning with <ulink url="http://repmgr.org/release-notes-3.1.3.html">repmgr 3.1.3</ulink>,
+      Beginning with <ulink url="https://repmgr.org/docs/4.0/release-4.0.5.html">repmgr 4.0.5</ulink>,
       <ulink url="https://2ndquadrant.com/">2ndQuadrant</ulink> provides a dedicated <literal>yum</literal>
-      repository for &repmgr; releases. This repository complements the main
-      <ulink url="https://yum.postgresql.org/repopackages.php">PGDG community repository</ulink>,
-      but enables repmgr users to access the latest &repmgr; packages before they are
-      available via the PGDG repository, which can take several days to be updated following
-      a fresh  &repmgr; release.
-    </para>
+	  <ulink url="https://rpm.2ndquadrant.com/">public RPM repository</ulink> for 2ndQuadrant software,
+	  including &repmgr;. We recommend using this for all future &repmgr; releases.
+	</para>
+	<para>
+	  General instructions for using this repository can be found on its
+	  <ulink url="https://rpm.2ndquadrant.com/">homepage</ulink>. Specific instructions
+	  for installing &repmgr; follow below.
+	</para>
     <para>
       <emphasis>Installation</emphasis>
 
       <itemizedlist>
-        <listitem>
-          <para>
-            Import the repository public key (optional but recommended):
-            <programlisting>
-              rpm --import http://packages.2ndquadrant.com/repmgr/RPM-GPG-KEY-repmgr</programlisting>
-          </para>
-        </listitem>
+		<listitem>
+		  <para>
+			Locate the repository RPM for your PostgreSQL version from the list at:
+			<ulink url="https://rpm.2ndquadrant.com/">https://rpm.2ndquadrant.com/</ulink>
+		  </para>
+		</listitem>
 
         <listitem>
           <para>
-            Install the repository RPM for your distribution (this enables the 2ndQuadrant
-            repository as a source of repmgr packages):
-            <itemizedlist>
-              <listitem>
-                <simpara>
-                  <emphasis>Fedora:</emphasis>
-                  <ulink url="http://packages.2ndquadrant.com/repmgr/yum-repo-rpms/repmgr-fedora-1.0-1.noarch.rpm">http://packages.2ndquadrant.com/repmgr/yum-repo-rpms/repmgr-fedora-1.0-1.noarch.rpm</ulink>
-                </simpara>
-              </listitem>
-              <listitem>
-                <simpara>
-                  <emphasis>RHEL, CentOS etc:</emphasis>
-                  <ulink url="http://packages.2ndquadrant.com/repmgr/yum-repo-rpms/repmgr-rhel-1.0-1.noarch.rpm">http://packages.2ndquadrant.com/repmgr/yum-repo-rpms/repmgr-rhel-1.0-1.noarch.rpm</ulink>
-                </simpara>
-              </listitem>
-            </itemizedlist>
-          </para>
-          <para>
-            e.g.:
-            <programlisting>
-              $ yum install http://packages.2ndquadrant.com/repmgr/yum-repo-rpms/repmgr-rhel-1.0-1.noarch.rpm</programlisting>
-          </para>
-        </listitem>
+            Install the repository RPM for your distribution and PostgreSQL version
+			(this enables the 2ndQuadrant repository as a source of &repmgr; packages).
+		  </para>
+		  <para>
+			For example, for PostgreSQL 10 on CentOS, execute:
+			<programlisting>
+sudo yum install https://rpm.2ndquadrant.com/site/content/2ndquadrant-repo-10-1-1.el7.noarch.rpm
+			</programlisting>
+		  </para>
+		  <para>
+			Verify that the repository is installed with:
+			<programlisting>
+sudo yum repolist</programlisting>
+			The output should contain two entries like this:
+			<programlisting>
+2ndquadrant-repo-10/7/x86_64         2ndQuadrant packages for PG10 for rhel 7 - x86_64           1
+2ndquadrant-repo-10-debug/7/x86_64   2ndQuadrant packages for PG10 for rhel 7 - x86_64 - Debug   1</programlisting>
+		  </para>
+		</listitem>
 
         <listitem>
           <para>
-            Install the repmgr version appropriate for your PostgreSQL version (e.g. <literal>repmgr96</literal>), e.g.:
+            Install the &repmgr version appropriate for your PostgreSQL version (e.g. <literal>repmgr10</literal>):
             <programlisting>
-              $ yum install repmgr96</programlisting>
+$ yum install repmgr10</programlisting>
           </para>
         </listitem>
       </itemizedlist>
@@ -105,13 +115,13 @@
       <emphasis>Compatibility with PGDG Repositories</emphasis>
     </para>
     <para>
-        The 2ndQuadrant &repmgr; yum repository uses exactly the same package definitions as the
-        main PGDG repository and is effectively a selective mirror for &repmgr; packages only.
+      The 2ndQuadrant &repmgr; yum repository packages use the same definitions and file system layout as the
+      main PGDG repository.
     </para>
     <para>
-        Normally yum should prioritize the repository with the most recent &repmgr; version.
-        Once the PGDG repository has been updated, it doesn't matter which repository
-        the packages are installed from.
+      Normally <application>yum</application> will prioritize the repository with the most recent &repmgr; version.
+      Once the PGDG repository has been updated, it doesn't matter which repository
+      the packages are installed from.
     </para>
     <para>
       To ensure the 2ndQuadrant repository is always prioritised, install <literal>yum-plugin-priorities</literal>
@@ -125,30 +135,23 @@
       To install a specific package version, execute <command>yum --showduplicates list</command>
       for the package in question:
       <programlisting>
-        [root@localhost ~]# yum --showduplicates list repmgr96
+        [root@localhost ~]# yum --showduplicates list repmgr10
         Loaded plugins: fastestmirror
         Loading mirror speeds from cached hostfile
          * base: ftp.iij.ad.jp
          * extras: ftp.iij.ad.jp
          * updates: ftp.iij.ad.jp
         Available Packages
-        repmgr96.x86_64               3.2-1.el6                    2ndquadrant-repmgr
-        repmgr96.x86_64               3.2.1-1.el6                  2ndquadrant-repmgr
-        repmgr96.x86_64               3.3-1.el6                    2ndquadrant-repmgr
-        repmgr96.x86_64               3.3.1-1.el6                  2ndquadrant-repmgr
-        repmgr96.x86_64               3.3.2-1.el6                  2ndquadrant-repmgr
-        repmgr96.x86_64               3.3.2-1.rhel6                pgdg96
-        repmgr96.x86_64               4.0.0-1.el6                  2ndquadrant-repmgr
-        repmgr96.x86_64               4.0.0-1.rhel6                pgdg96</programlisting>
+		repmgr10.x86_64                       4.0.3-1.rhel7                        pgdg10
+		repmgr10.x86_64                       4.0.4-1.rhel7                        pgdg10
+		repmgr10.x86_64                       4.0.5-1.el7                          2ndquadrant-repo-10</programlisting>
       then append the appropriate version number to the package name with a hyphen, e.g.:
       <programlisting>
-        [root@localhost ~]# yum install repmgr96-3.3.2-1.el6</programlisting>
+        [root@localhost ~]# yum install repmgr10-4.0.3-1.rhel7</programlisting>
     </para>
   </sect3>
  </sect2>
 
-
-
  <sect2 id="installation-packages-debian" xreflabel="Installing from packages on Debian or Ubuntu">
 
   <indexterm>
@@ -164,10 +167,83 @@
   </para>
   <para>
     For more information on the package contents, including details of installation
-    paths and relevant <link linkend="configuration-service-commands">service commands</link>,
+    paths and relevant <link linkend="configuration-file-service-commands">service commands</link>,
     see the appendix section <xref linkend="packages-debian-ubuntu">.
   </para>
 
+  <sect3 id="installation-packages-debian-ubuntu-2ndq">
+    <title>2ndQuadrant public apt repository for Debian/Ubuntu</title>
+
+    <para>
+      Beginning with <ulink url="https://repmgr.org/docs/4.0/release-4.0.5.html">repmgr 4.0.5</ulink>,
+      <ulink url="https://2ndquadrant.com/">2ndQuadrant</ulink> provides a
+	  <ulink url="https://apt.2ndquadrant.com/">public apt repository</ulink> for 2ndQuadrant software,
+	  including &repmgr;.
+	</para>
+	<para>
+	  General instructions for using this repository can be found on its
+	  <ulink url="https://apt.2ndquadrant.com/">homepage</ulink>. Specific instructions
+	  for installing &repmgr; follow below.
+	</para>
+
+    <para>
+      <emphasis>Installation</emphasis>
+
+      <itemizedlist>
+
+		<listitem>
+		  <para>
+			If not already present, install the  <application>apt-transport-https</application> package:
+			<programlisting>
+sudo apt-get install apt-transport-https</programlisting>
+		  </para>
+		</listitem>
+
+		<listitem>
+		  <para>
+			Create <filename>/etc/apt/sources.list.d/2ndquadrant.list</filename> as follows:
+			<programlisting>
+sudo sh -c 'echo "deb https://apt.2ndquadrant.com/ $(lsb_release -cs)-2ndquadrant main" > /etc/apt/sources.list.d/2ndquadrant.list'</programlisting>
+		  </para>
+		</listitem>
+
+		<listitem>
+		  <para>
+			Install the 2ndQuadrant <ulink url="https://apt.2ndquadrant.com/site/keys/9904CD4BD6BAF0C3.asc">repository key</ulink>:
+			<programlisting>
+sudo apt-get install curl ca-certificates
+curl https://apt.2ndquadrant.com/site/keys/9904CD4BD6BAF0C3.asc | sudo apt-key add -</programlisting>
+		  </para>
+		</listitem>
+
+		<listitem>
+		  <para>
+			Update the package list
+			<programlisting>
+ sudo apt-get update</programlisting>
+		  </para>
+		</listitem>
+
+		<listitem>
+		  <para>
+            Install the &repmgr version appropriate for your PostgreSQL version (e.g. <literal>repmgr10</literal>):
+            <programlisting>
+$ apt-get install postgresql-10-repmgr</programlisting>
+		  </para>
+          <note>
+            <para>
+            For packages for PostgreSQL 9.6 and earlier, the package name includes
+            a period between major and minor version numbers, e.g.
+            <literal>postgresql-9.6-repmgr</literal>.
+            </para>
+          </note>
+		</listitem>
+
+	  </itemizedlist>
+
+	</para>
+
+  </sect3>
  </sect2>
 
 </sect1>

doc/install-source.sgml

@@ -80,7 +80,7 @@
    </para>
 
    <para>
-    There are also tags for each &repmgr; release, e.g. <filename>REL4_0_STABLE</filename>.
+    There are also tags for each &repmgr; release, e.g. <filename>4.0.5</filename>.
    </para>
 
    <para>

doc/quickstart.sgml

@@ -234,7 +234,7 @@
   <para>
    <filename>repmgr.conf</filename> should not be stored inside the PostgreSQL data directory,
    as it could be overwritten when setting up or reinitialising the PostgreSQL
-   server. See sections on <xref linkend="configuration-file"> and <xref linkend="configuration-file-settings">
+   server. See sections <xref linkend="configuration"> and <xref linkend="configuration-file">
    for further details about <filename>repmgr.conf</filename>.
   </para>
   <tip>

doc/repmgr-cluster-crosscheck.sgml

@@ -38,5 +38,34 @@
       and therefore determine the state of outbound connections from that node.
     </para>
   </refsect1>
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr cluster crosscheck</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            The check completed successfully and all nodes are reachable.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_NODE_STATUS (25)</option></term>
+        <listitem>
+          <para>
+            One or more nodes could not be reached.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
 </refentry>
 

doc/repmgr-cluster-event.sgml

@@ -49,6 +49,22 @@
     </para>
   </refsect1>
 
+  <refsect1>
+    <title>Output format</title>
+    <para>
+      <itemizedlist spacing="compact" mark="bullet">
+
+        <listitem>
+          <simpara>
+            <literal>--csv</literal>: generate output in CSV format. Note that the <literal>Details</literal>
+            column will currently not be emitted in CSV format.
+          </simpara>
+        </listitem>
+
+      </itemizedlist>
+    </para>
+  </refsect1>
+
   <refsect1>
     <title>Example</title>
     <para>

doc/repmgr-cluster-matrix.sgml

@@ -97,5 +97,35 @@
     useful result.
   </para>
   </refsect1>
+
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr cluster matrix</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            The check completed successfully and all nodes are reachable.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_NODE_STATUS (25)</option></term>
+        <listitem>
+          <para>
+            One or more nodes could not be reached.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
 </refentry>
 

doc/repmgr-cluster-show.sgml

@@ -113,4 +113,40 @@
     </para>
   </refsect1>
 
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr cluster show</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            No issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_NODE_STATUS (25)</option></term>
+        <listitem>
+          <para>
+            One or more issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
+  <refsect1>
+    <title>See also</title>
+    <para>
+     <xref linkend="repmgr-node-status">, <xref linkend="repmgr-node-check">
+    </para>
+  </refsect1>
+
 </refentry>

doc/repmgr-node-check.sgml

@@ -61,7 +61,9 @@
 
      <listitem>
       <simpara>
-        <literal>--archive-ready</literal>: checks for WAL files which have not yet been archived
+        <literal>--archive-ready</literal>: checks for WAL files which have not yet been archived,
+        and returns <literal>WARNING</literal> or <literal>CRITICAL</literal> if the number
+        exceeds <varname>archive_ready_warning</varname> or <varname>archive_ready_critical</varname> respectively.
       </simpara>
      </listitem>
 
@@ -77,11 +79,110 @@
       </simpara>
      </listitem>
 
+     <listitem>
+      <simpara>
+        <literal>--missing-slots</literal>: checks there are no missing replication slots
+      </simpara>
+     </listitem>
+
     </itemizedlist>
   </para>
-  <para>
-   Individual checks can also be output in a Nagios-compatible format by additionally
-   providing the option <literal>--nagios</literal>.
-  </para>
   </refsect1>
+
+  <refsect1>
+    <title>Output format</title>
+    <para>
+      <itemizedlist spacing="compact" mark="bullet">
+
+        <listitem>
+          <simpara>
+            <literal>--csv</literal>: generate output in CSV format (not available
+            for individual checks)
+          </simpara>
+        </listitem>
+
+        <listitem>
+          <simpara>
+            <literal>--nagios</literal>: generate output in a Nagios-compatible format
+          </simpara>
+        </listitem>
+      </itemizedlist>
+    </para>
+  </refsect1>
+
+  <refsect1>
+    <title>Exit codes</title>
+
+    <para>
+      When executing <command>repmgr node check</command> with one of the individual
+      checks listed above, &repmgr; will emit one of the following Nagios-style exit codes
+      (even if <literal>--nagios</literal> is not supplied):
+
+      <itemizedlist spacing="compact" mark="bullet">
+
+        <listitem>
+          <simpara>
+            <literal>0</literal>: OK
+          </simpara>
+        </listitem>
+
+        <listitem>
+          <simpara>
+            <literal>1</literal>: WARNING
+          </simpara>
+        </listitem>
+
+        <listitem>
+          <simpara>
+            <literal>2</literal>: ERROR
+          </simpara>
+        </listitem>
+
+        <listitem>
+          <simpara>
+            <literal>3</literal>: UNKNOWN
+          </simpara>
+        </listitem>
+
+      </itemizedlist>
+    </para>
+
+
+
+    <para>
+      Following exit codes can be emitted by <command>repmgr status check</command>
+      if no individual check was specified.
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            No issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_NODE_STATUS (25)</option></term>
+        <listitem>
+          <para>
+            One or more issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
+
+
+  <refsect1>
+    <title>See also</title>
+    <para>
+     <xref linkend="repmgr-node-status">, <xref linkend="repmgr-cluster-show">
+    </para>
+  </refsect1>
+
 </refentry>

doc/repmgr-node-rejoin.sgml

@@ -115,7 +115,24 @@
 
     </variablelist>
   </refsect1>
+ <refsect1>
+    <title>Configuration file settings</title>
 
+    <para>
+      <itemizedlist spacing="compact" mark="bullet">
+       <listitem>
+         <simpara>
+           <literal>node_rejoin_timeout</literal>:
+		   the maximum length of time (in seconds) to wait for
+		   the node to reconnect to the replication cluster (defaults to
+		   the value set in <literal>standby_reconnect_timeout</literal>,
+		   60 seconds).
+		 </simpara>
+	   </listitem>
+	  </itemizedlist>
+	</para>
+
+ </refsect1>
   <refsect1>
     <title>Event notifications</title>
     <para>

doc/repmgr-node-status.sgml

@@ -24,7 +24,7 @@
     <title>Example</title>
     <para>
     <programlisting>
-        $ repmgr -f /etc/repmgr.comf node status
+        $ repmgr -f /etc/repmgr.conf node status
         Node "node1":
             PostgreSQL version: 10beta1
             Total data size: 30 MB
@@ -38,10 +38,54 @@
     </para>
   </refsect1>
 
+  <refsect1>
+    <title>Output format</title>
+    <para>
+      <itemizedlist spacing="compact" mark="bullet">
+
+        <listitem>
+          <simpara>
+            <literal>--csv</literal>: generate output in CSV format
+          </simpara>
+        </listitem>
+      </itemizedlist>
+    </para>
+  </refsect1>
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr node status</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            No issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_NODE_STATUS (25)</option></term>
+        <listitem>
+          <para>
+            One or more issues were detected.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
+
   <refsect1>
     <title>See also</title>
     <para>
-      See <xref linkend="repmgr-node-check"> to diagnose issues.
+      See <xref linkend="repmgr-node-check"> to diagnose issues and <xref linkend="repmgr-cluster-show">
+      for an overview of all nodes in the cluster.
     </para>
   </refsect1>
 </refentry>

doc/repmgr-primary-register.sgml

@@ -17,7 +17,7 @@
     <title>Description</title>
     <para>
       <command>repmgr primary register</command> registers a primary node in a
-      streaming replication cluster, and configures it for use with repmgr, including
+      streaming replication cluster, and configures it for use with &repmgr;, including
       installing the &repmgr; extension. This command needs to be executed before any
       standby nodes are registered.
     </para>

doc/repmgr-standby-clone.sgml

@@ -124,7 +124,7 @@
      <para>
        We recommend using <ulink url="https://www.pgbarman.org/">Barman</ulink> to manage
        WAL file archiving. For more details on combining &repmgr; and <application>Barman</application>,
-       in particular using <varname>restore_command</varname> to configure Barman as a backu source of
+       in particular using <varname>restore_command</varname> to configure Barman as a backup source of
        WAL files, see <xref linkend="cloning-from-barman">.
      </para>
    </note>
@@ -177,12 +177,13 @@
    <title>Using a standby cloned by another method</title>
    <para>
      &repmgr; supports standbys cloned by another method (e.g. using <application>barman</application>'s
-     <command>barman recover</command> command).
+     <command><ulink url="http://docs.pgbarman.org/release/2.4/#recover">barman recover</ulink></command> command).
    </para>
    <para>
      To integrate the standby as a &repmgr; node, ensure the <filename>repmgr.conf</filename>
-     file is created for the node, then execute the command
-     <command>repmgr standby clone --recovery-conf-only</command>.
+     file is created for the node, and that it has been registered using
+     <command><link linkend="repmgr-standby-register">repmgr standby register</link></command>.
+     Then execute the command <command>repmgr standby clone --recovery-conf-only</command>.
      This will create the <filename>recovery.conf</filename> file needed to attach
      the node to its upstream, and will also create a replication slot on the
      upstream node if required.
@@ -212,6 +213,15 @@
 
     <variablelist>
 
+      <varlistentry>
+        <term><option>-d, --dbname=CONNINFO</option></term>
+        <listitem>
+          <para>
+            Connection string of the upstream node to use for cloning.
+          </para>
+        </listitem>
+      </varlistentry>
+
       <varlistentry>
         <term><option>--dry-run</option></term>
         <listitem>

doc/repmgr-standby-follow.sgml

@@ -26,10 +26,18 @@
       running. It can only be used to attach an active standby to the current primary node
    (and not to another standby).
     </para>
-    <para>
-      To re-add an inactive node to the replication cluster, see
-      <xref linkend="repmgr-node-rejoin">
-    </para>
+	<tip>
+      <para>
+		To re-add an inactive node to the replication cluster, use
+		<xref linkend="repmgr-node-rejoin">.
+      </para>
+	</tip>
+
+	<para>
+	  <command>repmgr standby follow</command> will wait up to
+	  <varname>standby_follow_timeout</varname> seconds (default: <literal>30</literal>)
+	  to verify the standby has actually connected to the new primary.
+	</para>
 
   </refsect1>
 
@@ -92,7 +100,7 @@
       A <literal>standby_follow</literal> <link linkend="event-notifications">event notification</link> will be generated.
     </para>
     <para>
-      If provided, &repmgr; will subsitute the placeholders <literal>%p</literal> with the node ID of the primary
+      If provided, &repmgr; will substitute the placeholders <literal>%p</literal> with the node ID of the primary
       being followed, <literal>%c</literal> with its <literal>conninfo</literal> string, and
       <literal>%a</literal> with its node name.
     </para>

doc/repmgr-standby-promote.sgml

@@ -32,6 +32,7 @@
       check the promotion every <varname>promote_check_interval</varname> seconds (default: 1 second).
       Both values can be defined in <filename>repmgr.conf</filename>.
     </para>
+
   </refsect1>
 
   <refsect1>

doc/repmgr-standby-register.sgml

@@ -173,7 +173,7 @@
     </para>
 
     <para>
-      If provided, &repmgr; will subsitute the placeholders <literal>%p</literal> with the node ID of the
+      If provided, &repmgr; will substitute the placeholders <literal>%p</literal> with the node ID of the
       primary node, <literal>%c</literal> with its <literal>conninfo</literal> string, and
       <literal>%a</literal> with its node name.
     </para>

doc/repmgr-standby-switchover.sgml

@@ -12,6 +12,7 @@
     <refpurpose>promote a standby to primary and demote the existing primary to a standby</refpurpose>
   </refnamediv>
 
+
   <refsect1>
     <title>Description</title>
 
@@ -39,6 +40,17 @@
       For more details on performing a switchover, including preparation and configuration,
       see section <xref linkend="performing-switchover">.
     </para>
+
+    <note>
+      <para>
+        <application>repmgrd</application> should not be active on any nodes while a switchover is being
+        executed. This restriction may be lifted in a later version.
+      </para>
+      <para>
+        &repmgr; will not perform the switchover if an exclusive backup is running on the current primary.
+      </para>
+    </note>
+
   </refsect1>
 
   <refsect1>
@@ -154,8 +166,8 @@
        <listitem>
          <simpara>
            <literal>standby_reconnect_timeout</literal>:
-           Number of seconds to attempt to reconnect to the demoted primary
-           once it has been restarted.
+           number of seconds to attempt to wait for the demoted primary
+           to reconnect to the promoted primary (default: 60 seconds)
          </simpara>
        </listitem>
 
@@ -171,10 +183,12 @@
       Execute with the <literal>--dry-run</literal> option to test the switchover as far as
       possible without actually changing the status of either node.
     </para>
-    <para>
-      <application>repmgrd</application> should not be active on any nodes while a switchover is being
-      executed. This restriction may be lifted in a later version.
-    </para>
+    <important>
+      <para>
+        <application>repmgrd</application> must be shut down on all nodes while a switchover is being
+        executed. This restriction will be removed in a future &repmgr; version.
+      </para>
+    </important>
     <para>
       External database connections, e.g. from an application, should not be permitted while
       the switchover is taking place. In particular, active transactions on the primary
@@ -199,7 +213,7 @@
   <refsect1>
     <title>Exit codes</title>
     <para>
-      Following exit codes can be emitted by <literal>repmgr standby switchover</literal>:
+      Following exit codes can be emitted by <command>repmgr standby switchover</command>:
     </para>
     <variablelist>
 
@@ -227,7 +241,7 @@
           <para>
             The switchover was executed but a problem was encountered.
             Typically this means the former primary could not be reattached
-            as a standby.
+            as a standby. Check preceding log messages for more information.
           </para>
         </listitem>
       </varlistentry>

doc/repmgr-witness-unregister.sgml

@@ -20,7 +20,10 @@
     </para>
     <para>
       The node does not have to be running to be unregistered, however if this is the
-      case then connection information for the primary server must be provided.
+      case then either provide connection information for the primary server, or
+	  execute <command>repmgr witness unregister</command> on a running node and
+	  provide the parameter <option>--node-id</option> with the node ID of the
+	  witness server.
     </para>
     <para>
       Execute with the <literal>--dry-run</literal> option to check what would happen
@@ -36,17 +39,17 @@
     INFO: connecting to witness node "node3" (ID: 3)
     INFO: unregistering witness node 3
     INFO: witness unregistration complete
-    DETAIL: witness node with id 3 (conninfo: host=node3 dbname=repmgr user=repmgr port=5499) successfully unregistered</programlisting>
+    DETAIL: witness node with UD 3 successfully unregistered</programlisting>
     </para>
     <para>
       Unregistering a non-running witness node:
       <programlisting>
         $ repmgr -f /etc/repmgr.conf witness unregister -h node1 -p 5501  -F
-        INFO: connecting to witness node "node3" (ID: 3)
-        NOTICE: unable to connect to witness node "node3" (ID: 3), removing node record on cluster primary only
+        INFO: connecting to node "node3" (ID: 3)
+        NOTICE: unable to connect to node "node3" (ID: 3), removing node record on cluster primary only
         INFO: unregistering witness node 3
         INFO: witness unregistration complete
-        DETAIL: witness node with id 3 (conninfo: host=node3 dbname=repmgr user=repmgr port=5499) successfully unregistered</programlisting>
+        DETAIL: witness node with id ID 3 successfully unregistered</programlisting>
     </para>
   </refsect1>
 
@@ -62,6 +65,32 @@
     </para>
   </refsect1>
 
+  <refsect1>
+
+    <title>Options</title>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>--dry-run</option></term>
+        <listitem>
+          <para>
+            Check prerequisites but don't actually unregister the witness.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+       <term><option>--node-id</option></term>
+        <listitem>
+          <para>
+            Unregister witness server with the specified node ID.
+          </para>
+        </listitem>
+      </varlistentry>
+
+    </variablelist>
+  </refsect1>
+
 
   <refsect1>
     <title>Event notifications</title>

doc/repmgrd-bdr.sgml

@@ -99,15 +99,16 @@
       replication cluster. The database must be the BDR-enabled database.
     </para>
     <para>
-      If defined, the evenr <application>event_notifications</application> parameter
-      will restrict execution of <varname>event_notification_command</varname>
+      If defined, the <varname>event_notifications</varname> parameter will restrict
+      execution of the script defined in  <varname>event_notification_command</varname>
       to the specified event(s).
     </para>
     <note>
       <simpara>
         <varname>event_notification_command</varname> is the script which does the actual "heavy lifting"
         of reconfiguring the proxy server/ connection pooler. It is fully
-        user-definable; a reference implementation is documented below.
+        user-definable; see section <xref linkend="bdr-event-notification-command"> for a reference
+        implementation.
       </simpara>
     </note>
 
@@ -169,8 +170,8 @@
     </para>
   </sect1>
 
-  <sect1 id="bdr-event-notification-command" xreflabel="BDR failover event notification command">
-    <title>Defining the "event_notification_command"</title>
+  <sect1 id="bdr-event-notification-command" xreflabel="Defining the BDR failover &quot;event_notification command&quot;">
+    <title>Defining the BDR failover "event_notification_command"</title>
     <para>
       Key to "failover" execution is the <literal>event_notification_command</literal>,
       which is a user-definable script specified in <filename>repmpgr.conf</filename>

doc/repmgrd-configuration.sgml

@@ -24,7 +24,7 @@
 
     <para>
       To use <application>repmgrd</application>, its associated function library <emphasis>must</emphasis> be
-      included in <filename>postgresql.conf</filename> with:
+      included via <filename>postgresql.conf</filename> with:
 
       <programlisting>
         shared_preload_libraries = 'repmgr'</programlisting>
@@ -34,6 +34,25 @@
       the <ulink url="https://www.postgresql.org/docs/current/static/runtime-config-client.html#GUC-SHARED-PRELOAD-LIBRARIES">PostgreSQL documentation</ulink>.
     </para>
 
+    <para>
+      To apply configuration file changes to a running <application>repmgrd</application>
+      daemon, execute the operating system's r<application>repmgrd</application> service reload command
+      (see <xref linkend="appendix-packages"> for examples),
+      or for instances  which were manually started, execute <command>kill -HUP</command>, e.g.
+      <command>kill -HUP `cat /tmp/repmgrd.pid`</command>.
+    </para>
+    <note>
+      <para>
+        Check the <application>repmgrd</application> log to see what changes were
+        applied, or if any issues were encountered when reloading the configuration.
+      </para>
+    </note>
+    <para>
+      Note that only a subset of configuration file parameters can be changed on a
+      running <application>repmgrd</application> daemon.
+    </para>
+
+
     <sect2 id="repmgrd-automatic-failover-configuration">
       <title>automatic failover configuration</title>
       <para>
@@ -112,7 +131,7 @@
         particularly on <application>systemd</application>-based systems.
       </para>
       <para>
-        For more details, see <xref linkend="configuration-service-commands">.
+        For more details, see <xref linkend="configuration-file-service-commands">.
       </para>
     </sect2>
 
@@ -159,16 +178,62 @@
     <para>
       <application>repmgrd</application> can be started manually like this:
       <programlisting>
-        repmgrd -f /etc/repmgr.conf --pid-file /tmp/repmgrd.pid --daemonize</programlisting>
+        repmgrd -f /etc/repmgr.conf --pid-file /tmp/repmgrd.pid</programlisting>
       and stopped with <command>kill `cat /tmp/repmgrd.pid`</command>. Adjust paths as appropriate.
     </para>
-    <para>
-      To apply configuration file changes to a running <application>repmgrd</application>
-      daemon, execute the operating system's service reload command (for manually started
-      instances, execute <command>kill -HUP `cat /tmp/repmgrd.pid`</command>).
-      Note that only a subset of configuration file parameters can be changed on a
-      running <application>repmgrd</application> daemon.
-    </para>
+
+    <sect2 id="repmgrd-pid-file" xreflabel="repmgrd's PID file">
+      <indexterm>
+        <primary>repmgrd</primary>
+        <secondary>PID file</secondary>
+      </indexterm>
+      <indexterm>
+        <primary>PID file</primary>
+        <secondary>repmgrd</secondary>
+      </indexterm>
+      <title>repmgrd's PID file</title>
+      <para>
+        <application>repmgrd</application> will generate a PID file by default.
+      </para>
+      <note>
+        <simpara>
+          This is a behaviour change from previous versions (earlier than 4.1), where
+          the PID file had to be explicitly specified with the command line
+          parameter <option> --pid-file</option>.
+        </simpara>
+      </note>
+      <para>
+        The PID file can be specified in <filename>repmgr.conf</filename> with the configuration
+        parameter <varname>repmgrd_pid_file</varname>.
+      </para>
+      <para>
+        It can also be specified on the command line (as in previous versions) with
+        the command line parameter <option>--pid-file</option>. Note this will override
+        any value set in <filename>repmgr.conf</filename> with <varname>repmgrd_pid_file</varname>.
+        <option>--pid-file</option> may be deprecated in future releases.
+      </para>
+      <para>
+        If a PID file location was specified by the package maintainer, <application>repmgrd</application>
+        will use that. This only applies if &repmgr; was installed from a package and the package
+        maintainer has specified the PID file location.
+      </para>
+      <para>
+        If none of the above apply, <application>repmgrd</application> will create a PID file
+        in the operating system's temporary directory (das etermined by the environment variable
+        <varname>TMPDIR</varname>, or if that is not set, will use <filename>/tmp</filename>).
+      </para>
+      <para>
+        To prevent a PID file being generated at all, provide the command line option
+        <option>--no-pid-file</option>.
+      </para>
+      <para>
+        To see which PID file <application>repmgrd</application> would use, execute <application>repmgrd</application>
+        with the option <option>--show-pid-file</option>. <application>repmgrd</application>
+        will not start if this option is provided. Note that the value shown is the
+        file  <application>repmgrd</application> would use next time it starts, and is
+        not necessarily the PID file currently in use.
+      </para>
+    </sect2>
 
     <sect2 id="repmgrd-configuration-debian-ubuntu">
       <indexterm>

doc/switchover.sgml

@@ -57,7 +57,7 @@
 
    <para>
     As mentioned in the previous section, success of the switchover operation depends on
-    &repmgr;  being able to shut down the current primary server quickly and cleanly.
+    &repmgr; being able to shut down the current primary server quickly and cleanly.
    </para>
 
    <para>
@@ -104,7 +104,7 @@
        server.
      </para>
      <para>
-       For more details, see <xref linkend="configuration-service-commands">.
+       For more details, see <xref linkend="configuration-file-service-commands">.
      </para>
    </important>
 
@@ -121,15 +121,21 @@
     </simpara>
    </note>
 
-
    <para>
-    Check that access from applications is minimalized or preferably blocked
-    completely, so applications are not unexpectedly interrupted.
+     Check that access from applications is minimalized or preferably blocked
+     completely, so applications are not unexpectedly interrupted.
    </para>
 
+   <note>
+     <para>
+       If an exclusive backup is running on the current primary, &repmgr; will not perform the
+       switchover.
+     </para>
+   </note>
+
    <para>
-    Check there is no significant replication lag on standbys attached to the
-    current primary.
+     Check there is no significant replication lag on standbys attached to the
+     current primary.
    </para>
 
    <para>
@@ -140,10 +146,13 @@
     manually with <command>repmgr node check --archive-ready</command>.
    </para>
 
-   <para>
-    Ensure that <application>repmgrd</application> is *not* running anywhere to prevent it unintentionally
-    promoting a node.
-   </para>
+   <note>
+     <para>
+       Ensure that <application>repmgrd</application> is *not* running anywhere to prevent it unintentionally
+       promoting a node. This restriction will be removed in a future &repmgr; version.
+     </para>
+   </note>
+
 
    <para>
     Finally, consider executing <command>repmgr standby switchover</command> with the

doc/upgrading-repmgr.sgml

@@ -29,8 +29,18 @@
       </listitem>
       <listitem>
         <simpara>
-          In the database where the &repmgr; extension is installed, execute
-          <command>ALTER EXTENSION repmgr UPDATE</command>.
+          <application>repmgrd</application> (if running) must be restarted.
+        </simpara>
+      </listitem>
+      <listitem>
+        <simpara>
+          For major releases, e.g. from <literal>4.0.x</literal> to <literal>4.1</literal>,
+          execute <command>ALTER EXTENSION repmgr UPDATE</command>
+          on the primary node in the database where the &repmgr; extension is installed.
+        </simpara>
+        <simpara>
+          This will update the extension metadata and, if necessary, apply
+          changes to the &repmgr; extension objects.
         </simpara>
       </listitem>
     </orderedlist>
@@ -41,10 +51,6 @@
     release as they may contain upgrade instructions particular to individual versions.
   </para>
 
-  <para>
-    If the <application>repmgrd</application> daemon is in use, we recommend stopping it
-    before upgrading &repmgr;.
-  </para>
   <para>
     Note that it may be necessary to restart the PostgreSQL server if the upgrade contains
     changes to the shared object file used by <application>repmgrd</application>; check the

doc/version.sgml

@@ -1 +1 @@
-<!ENTITY repmgrversion "4.0.5">
+<!ENTITY repmgrversion "4.1.0">

errcode.h

@@ -46,5 +46,6 @@
 #define ERR_SWITCHOVER_INCOMPLETE 22
 #define ERR_FOLLOW_FAIL 23
 #define ERR_REJOIN_FAIL 24
+#define ERR_NODE_STATUS 25
 
 #endif							/* _ERRCODE_H_ */

log.c

@@ -42,7 +42,7 @@ _stderr_log_with_level(const char *level_name, int level, const char *fmt, va_li
 __attribute__((format(PG_PRINTF_ATTRIBUTE, 3, 0)));
 
 int			log_type = REPMGR_STDERR;
-int			log_level = LOG_NOTICE;
+int			log_level = LOG_INFO;
 int			last_log_level = LOG_INFO;
 int			verbose_logging = false;
 int			terse_logging = false;
@@ -70,7 +70,7 @@ _stderr_log_with_level(const char *level_name, int level, const char *fmt, va_li
 
 	/*
 	 * Store the requested level so that if there's a subsequent log_hint() or
-	 * log_detail(), we can suppress that if appropriate.
+	 * log_detail(), we can suppress that if --terse was specified,
 	 */
 	last_log_level = level;
 
@@ -329,6 +329,21 @@ logger_set_terse(void)
 }
 
 
+void
+logger_set_level(int new_log_level)
+{
+	log_level = new_log_level;
+}
+
+
+void
+logger_set_min_level(int min_log_level)
+{
+	if (min_log_level > log_level)
+		log_level = min_log_level;
+}
+
+
 int
 detect_log_level(const char *level)
 {

log.h

@@ -128,6 +128,8 @@ bool		logger_shutdown(void);
 
 void		logger_set_verbose(void);
 void		logger_set_terse(void);
+void		logger_set_min_level(int min_log_level);
+void		logger_set_level(int new_log_level);
 
 void
 log_detail(const char *fmt,...)

repmgr--4.0--4.1.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "CREATE EXTENSION repmgr" to load this file. \quit

repmgr--4.1.sql

@@ -0,0 +1,167 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "CREATE EXTENSION repmgr" to load this file. \quit
+
+CREATE TABLE repmgr.nodes (
+  node_id          INTEGER     PRIMARY KEY,
+  upstream_node_id INTEGER     NULL REFERENCES nodes (node_id) DEFERRABLE,
+  active           BOOLEAN     NOT NULL DEFAULT TRUE,
+  node_name        TEXT        NOT NULL,
+  type             TEXT        NOT NULL CHECK (type IN('primary','standby','witness','bdr')),
+  location         TEXT        NOT NULL DEFAULT 'default',
+  priority         INT         NOT NULL DEFAULT 100,
+  conninfo         TEXT        NOT NULL,
+  repluser         VARCHAR(63) NOT NULL,
+  slot_name        TEXT        NULL,
+  config_file      TEXT        NOT NULL
+);
+
+CREATE TABLE repmgr.events (
+  node_id          INTEGER NOT NULL,
+  event            TEXT NOT NULL,
+  successful       BOOLEAN NOT NULL DEFAULT TRUE,
+  event_timestamp  TIMESTAMP WITH TIME ZONE NOT NULL DEFAULT CURRENT_TIMESTAMP,
+  details          TEXT NULL
+);
+
+DO $repmgr$
+DECLARE
+  DECLARE server_version_num INT;
+BEGIN
+  SELECT setting
+    FROM pg_catalog.pg_settings
+   WHERE name = 'server_version_num'
+    INTO server_version_num;
+  IF server_version_num >= 90400 THEN
+    EXECUTE $repmgr_func$
+CREATE TABLE repmgr.monitoring_history (
+  primary_node_id                INTEGER NOT NULL,
+  standby_node_id                INTEGER NOT NULL,
+  last_monitor_time              TIMESTAMP WITH TIME ZONE NOT NULL,
+  last_apply_time                TIMESTAMP WITH TIME ZONE,
+  last_wal_primary_location      PG_LSN NOT NULL,
+  last_wal_standby_location      PG_LSN,
+  replication_lag                BIGINT NOT NULL,
+  apply_lag                      BIGINT NOT NULL
+)
+    $repmgr_func$;
+  ELSE
+    EXECUTE $repmgr_func$
+CREATE TABLE repmgr.monitoring_history (
+  primary_node_id                INTEGER NOT NULL,
+  standby_node_id                INTEGER NOT NULL,
+  last_monitor_time              TIMESTAMP WITH TIME ZONE NOT NULL,
+  last_apply_time                TIMESTAMP WITH TIME ZONE,
+  last_wal_primary_location      TEXT NOT NULL,
+  last_wal_standby_location      TEXT,
+  replication_lag                BIGINT NOT NULL,
+  apply_lag                      BIGINT NOT NULL
+)
+    $repmgr_func$;
+  END IF;
+END$repmgr$;
+
+
+
+CREATE INDEX idx_monitoring_history_time
+          ON repmgr.monitoring_history (last_monitor_time, standby_node_id);
+
+CREATE VIEW repmgr.show_nodes AS
+   SELECT n.node_id,
+          n.node_name,
+          n.active,
+          n.upstream_node_id,
+          un.node_name AS upstream_node_name,
+          n.type,
+          n.priority,
+          n.conninfo
+     FROM repmgr.nodes n
+LEFT JOIN repmgr.nodes un
+       ON un.node_id = n.upstream_node_id;
+
+
+/* XXX update upgrade scripts! */
+CREATE TABLE repmgr.voting_term (
+  term INT NOT NULL
+);
+
+CREATE UNIQUE INDEX voting_term_restrict
+ON repmgr.voting_term ((TRUE));
+
+CREATE RULE voting_term_delete AS
+   ON DELETE TO repmgr.voting_term
+   DO INSTEAD NOTHING;
+
+
+/* ================= */
+/* repmgrd functions */
+/* ================= */
+
+/* monitoring functions */
+
+CREATE FUNCTION set_local_node_id(INT)
+  RETURNS VOID
+  AS 'MODULE_PATHNAME', 'set_local_node_id'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION get_local_node_id()
+  RETURNS INT
+  AS 'MODULE_PATHNAME', 'get_local_node_id'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION standby_set_last_updated()
+  RETURNS TIMESTAMP WITH TIME ZONE
+  AS 'MODULE_PATHNAME', 'standby_set_last_updated'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION standby_get_last_updated()
+  RETURNS TIMESTAMP WITH TIME ZONE
+  AS 'MODULE_PATHNAME', 'standby_get_last_updated'
+  LANGUAGE C STRICT;
+
+/* failover functions */
+
+CREATE FUNCTION notify_follow_primary(INT)
+  RETURNS VOID
+  AS 'MODULE_PATHNAME', 'notify_follow_primary'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION get_new_primary()
+  RETURNS INT
+  AS 'MODULE_PATHNAME', 'get_new_primary'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION reset_voting_status()
+  RETURNS VOID
+  AS 'MODULE_PATHNAME', 'reset_voting_status'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION am_bdr_failover_handler(INT)
+  RETURNS BOOL
+  AS 'MODULE_PATHNAME', 'am_bdr_failover_handler'
+  LANGUAGE C STRICT;
+
+CREATE FUNCTION unset_bdr_failover_handler()
+  RETURNS VOID
+  AS 'MODULE_PATHNAME', 'unset_bdr_failover_handler'
+  LANGUAGE C STRICT;
+
+
+CREATE VIEW repmgr.replication_status AS
+  SELECT m.primary_node_id, m.standby_node_id, n.node_name AS standby_name,
+ 	     n.type AS node_type, n.active, last_monitor_time,
+         CASE WHEN n.type='standby' THEN m.last_wal_primary_location ELSE NULL END AS last_wal_primary_location,
+         m.last_wal_standby_location,
+         CASE WHEN n.type='standby' THEN pg_catalog.pg_size_pretty(m.replication_lag) ELSE NULL END AS replication_lag,
+         CASE WHEN n.type='standby' THEN
+           CASE WHEN replication_lag > 0 THEN age(now(), m.last_apply_time) ELSE '0'::INTERVAL END
+           ELSE NULL
+         END AS replication_time_lag,
+         CASE WHEN n.type='standby' THEN pg_catalog.pg_size_pretty(m.apply_lag) ELSE NULL END AS apply_lag,
+         AGE(NOW(), CASE WHEN pg_catalog.pg_is_in_recovery() THEN repmgr.standby_get_last_updated() ELSE m.last_monitor_time END) AS communication_time_lag
+    FROM repmgr.monitoring_history m
+    JOIN repmgr.nodes n ON m.standby_node_id = n.node_id
+   WHERE (m.standby_node_id, m.last_monitor_time) IN (
+	          SELECT m1.standby_node_id, MAX(m1.last_monitor_time)
+			    FROM repmgr.monitoring_history m1 GROUP BY 1
+         );
+

repmgr-action-bdr.c

@@ -83,9 +83,10 @@ do_bdr_register(void)
 		exit(ERR_BAD_CONFIG);
 	}
 
-	if (bdr_nodes.node_count > 2)
+	/* BDR 2 implementation is for 2 nodes only */
+	if (get_bdr_version_num() < 3 && bdr_nodes.node_count > 2)
 	{
-		log_error(_("repmgr can only support BDR clusters with 2 nodes"));
+		log_error(_("repmgr can only support BDR 2.x clusters with 2 nodes"));
 		log_detail(_("this BDR cluster has %i nodes"), bdr_nodes.node_count);
 		PQfinish(conn);
 		pfree(dbname);
@@ -176,6 +177,7 @@ do_bdr_register(void)
 
 	if (bdr_node_has_repmgr_set(conn, config_file_options.node_name) == false)
 	{
+		log_debug("bdr_node_has_repmgr_set() = false");
 		bdr_node_set_repmgr_set(conn, config_file_options.node_name);
 	}
 
@@ -201,6 +203,7 @@ do_bdr_register(void)
 			if (bdr_nodes.node_count == 0)
 			{
 				log_error(_("unable to retrieve any BDR node records"));
+				log_detail("%s", PQerrorMessage(conn));
 				PQfinish(conn);
 				exit(ERR_BAD_CONFIG);
 			}
@@ -252,7 +255,35 @@ do_bdr_register(void)
 	}
 
 	/* Add the repmgr extension tables to a replication set */
-	add_extension_tables_to_bdr_replication_set(conn);
+
+	if (get_bdr_version_num() < 3)
+	{
+		add_extension_tables_to_bdr_replication_set(conn);
+	}
+	else
+	{
+		/* this is the only table we need to replicate */
+		char *replication_set = get_default_bdr_replication_set(conn);
+
+		/*
+		 * this probably won't happen, but we need to be sure we're using
+		 * the replication set metadata correctly...
+		 */
+		if (conn == NULL)
+		{
+			log_error(_("unable to retrieve default BDR replication set"));
+			log_hint(_("see preceding messages"));
+			log_debug("check query in get_default_bdr_replication_set()");
+			exit(ERR_BAD_CONFIG);
+		}
+
+		if (is_table_in_bdr_replication_set(conn, "nodes", replication_set) == false)
+		{
+			add_table_to_bdr_replication_set(conn, "nodes", replication_set);
+		}
+
+		pfree(replication_set);
+	}
 
 	initPQExpBuffer(&event_details);
 

repmgr-action-cluster.c

@@ -83,6 +83,7 @@ do_cluster_show(void)
 	int			i = 0;
 	ItemList	warnings = {NULL, NULL};
 	bool		success = false;
+	bool		error_found = false;
 
 	/* Connect to local database to obtain cluster connection data */
 	log_verbose(LOG_INFO, _("connecting to database"));
@@ -218,6 +219,7 @@ do_cluster_show(void)
 						else
 						{
 							appendPQExpBuffer(&details, "- failed");
+							error_found = true;
 						}
 					}
 				}
@@ -281,6 +283,7 @@ do_cluster_show(void)
 						else
 						{
 							appendPQExpBuffer(&details, "- failed");
+							error_found = true;
 						}
 					}
 				}
@@ -292,17 +295,27 @@ do_cluster_show(void)
 					if (cell->node_info->node_status == NODE_STATUS_UP)
 					{
 						if (cell->node_info->active == true)
+						{
 							appendPQExpBuffer(&details, "* running");
+						}
 						else
+						{
 							appendPQExpBuffer(&details, "! running");
+							error_found = true;
+						}
 					}
 					/* node is unreachable */
 					else
 					{
 						if (cell->node_info->active == true)
+						{
 							appendPQExpBuffer(&details, "? unreachable");
+						}
 						else
+						{
 							appendPQExpBuffer(&details, "- failed");
+							error_found = true;
+						}
 					}
 				}
 				break;
@@ -310,6 +323,7 @@ do_cluster_show(void)
 				{
 					/* this should never happen */
 					appendPQExpBuffer(&details, "? unknown node type");
+						error_found = true;
 				}
 				break;
 		}
@@ -414,7 +428,6 @@ do_cluster_show(void)
 	PQfinish(conn);
 
 	/* emit any warnings */
-
 	if (warnings.head != NULL && runtime_options.terse == false && runtime_options.output_mode != OM_CSV)
 	{
 		ItemListCell *cell = NULL;
@@ -425,6 +438,20 @@ do_cluster_show(void)
 			printf(_("  - %s\n"), cell->string);
 		}
 	}
+
+	/*
+	 * If warnings were noted, even if they're not displayed (e.g. in --csv node),
+	 * that means something's not right so we need to emit a non-zero exit code.
+	 */
+	if (warnings.head != NULL)
+	{
+		error_found = true;
+	}
+
+	if (error_found == true)
+	{
+		exit(ERR_NODE_STATUS);
+	}
 }
 
 
@@ -436,6 +463,7 @@ do_cluster_show(void)
  *   --all
  *   --node-[id|name]
  *   --event
+ *   --csv
  */
 
 void
@@ -480,8 +508,12 @@ do_cluster_event(void)
 	strncpy(headers_event[EV_TIMESTAMP].title, _("Timestamp"), MAXLEN);
 	strncpy(headers_event[EV_DETAILS].title, _("Details"), MAXLEN);
 
-	/* if --terse provided, simply omit the "Details" column */
-	if (runtime_options.terse == true)
+	/*
+	 * If --terse or --csv provided, simply omit the "Details" column.
+	 * In --csv mode we'd need to quote/escape the contents "Details" column,
+	 * which is doable but which will remain a TODO for now.
+	 */
+	if (runtime_options.terse == true || runtime_options.output_mode == OM_CSV)
 		column_count --;
 
 	for (i = 0; i < column_count; i++)
@@ -504,47 +536,64 @@ do_cluster_event(void)
 
 	}
 
-	for (i = 0; i < column_count; i++)
+	if (runtime_options.output_mode == OM_TEXT)
 	{
-		if (i == 0)
-			printf(" ");
-		else
-			printf(" | ");
+		for (i = 0; i < column_count; i++)
+		{
+			if (i == 0)
+				printf(" ");
+			else
+				printf(" | ");
 
-		printf("%-*s",
-			   headers_event[i].max_length,
-			   headers_event[i].title);
+			printf("%-*s",
+				   headers_event[i].max_length,
+				   headers_event[i].title);
+		}
+		printf("\n");
+		printf("-");
+		for (i = 0; i < column_count; i++)
+		{
+			int			j;
+
+			for (j = 0; j < headers_event[i].max_length; j++)
+				printf("-");
+
+			if (i < (column_count - 1))
+				printf("-+-");
+			else
+				printf("-");
+		}
+
+		printf("\n");
 	}
-	printf("\n");
-	printf("-");
-	for (i = 0; i < column_count; i++)
-	{
-		int			j;
-
-		for (j = 0; j < headers_event[i].max_length; j++)
-			printf("-");
-
-		if (i < (column_count - 1))
-			printf("-+-");
-		else
-			printf("-");
-	}
-
-	printf("\n");
 
 	for (i = 0; i < PQntuples(res); i++)
 	{
 		int			j;
 
-		printf(" ");
-		for (j = 0; j < column_count; j++)
+		if (runtime_options.output_mode == OM_CSV)
 		{
-			printf("%-*s",
-				   headers_event[j].max_length,
-				   PQgetvalue(res, i, j));
+			for (j = 0; j < column_count; j++)
+			{
+				printf("%s", PQgetvalue(res, i, j));
+				if ((j + 1) < column_count)
+				{
+					printf(",");
+				}
+			}
+		}
+		else
+		{
+			printf(" ");
+			for (j = 0; j < column_count; j++)
+			{
+				printf("%-*s",
+					   headers_event[j].max_length,
+					   PQgetvalue(res, i, j));
 
-			if (j < (column_count - 1))
-				printf(" | ");
+				if (j < (column_count - 1))
+					printf(" | ");
+			}
 		}
 
 		printf("\n");
@@ -554,7 +603,8 @@ do_cluster_event(void)
 
 	PQfinish(conn);
 
-	puts("");
+	if (runtime_options.output_mode == OM_TEXT)
+		puts("");
 }
 
 
@@ -569,6 +619,8 @@ do_cluster_crosscheck(void)
 
 	t_node_status_cube **cube;
 
+	bool		error_found = false;
+
 	n = build_cluster_crosscheck(&cube, &name_length);
 	if (runtime_options.output_mode == OM_CSV)
 	{
@@ -648,9 +700,11 @@ do_cluster_crosscheck(void)
 				{
 					case -2:
 						c = '?';
+						error_found = true;
 						break;
 					case -1:
 						c = 'x';
+						error_found = true;
 						break;
 					case 0:
 						c = '*';
@@ -689,6 +743,11 @@ do_cluster_crosscheck(void)
 
 		free(cube);
 	}
+
+	if (error_found == true)
+	{
+		exit(ERR_NODE_STATUS);
+	}
 }
 
 
@@ -704,6 +763,8 @@ do_cluster_matrix()
 
 	t_node_matrix_rec **matrix_rec_list;
 
+	bool		error_found = false;
+
 	n = build_cluster_matrix(&matrix_rec_list, &name_length);
 
 	if (runtime_options.output_mode == OM_CSV)
@@ -742,9 +803,11 @@ do_cluster_matrix()
 				{
 					case -2:
 						c = '?';
+						error_found = true;
 						break;
 					case -1:
 						c = 'x';
+						error_found = true;
 						break;
 					case 0:
 						c = '*';
@@ -770,6 +833,11 @@ do_cluster_matrix()
 	}
 
 	free(matrix_rec_list);
+
+	if (error_found == true)
+	{
+		exit(ERR_NODE_STATUS);
+	}
 }
 
 
@@ -1329,6 +1397,7 @@ do_cluster_help(void)
 	printf(_("    %s [OPTIONS] cluster matrix\n"), progname());
 	printf(_("    %s [OPTIONS] cluster crosscheck\n"), progname());
 	printf(_("    %s [OPTIONS] cluster event\n"), progname());
+	printf(_("    %s [OPTIONS] cluster cleanup\n"), progname());
 	puts("");
 
 	printf(_("CLUSTER SHOW\n"));
@@ -1368,6 +1437,7 @@ do_cluster_help(void)
 	printf(_("    --event                   filter specific event\n"));
 	printf(_("    --node-id                 restrict entries to node with this ID\n"));
 	printf(_("    --node-name               restrict entries to node with this name\n"));
+	printf(_("    --csv                     emit output as CSV\n"));
 	puts("");
 
 	printf(_("CLUSTER CLEANUP\n"));

repmgr-action-standby.c

@@ -87,7 +87,7 @@ static void initialise_direct_clone(t_node_info *node_record);
 static int	run_basebackup(t_node_info *node_record);
 static int	run_file_backup(t_node_info *node_record);
 
-static void copy_configuration_files(void);
+static void copy_configuration_files(bool delete_after_copy);
 
 static void drop_replication_slot_if_exists(PGconn *conn, int node_id, char *slot_name);
 
@@ -498,7 +498,33 @@ do_standby_clone(void)
 
 			termPQExpBuffer(&msg);
 
-			/* TODO: check all files are readable */
+			/*
+			 * Here we'll attempt an initial test copy of the detected external
+			 * files, to detect any issues before we run the base backup.
+			 *
+			 * Note this will exit with an error, unless -F/--force supplied.
+			 *
+			 * TODO: put the files in a temporary directory and move to their final
+			 * destination once the database has been cloned.
+			 */
+
+			if (runtime_options.copy_external_config_files_destination == CONFIG_FILE_SAMEPATH)
+			{
+				/*
+				 * Files will be placed in the same path as on the source server;
+				 * don't delete after copying.
+				 */
+				copy_configuration_files(false);
+
+			}
+			else
+			{
+				/*
+				 * Files will be placed in the data directory - delete after copying.
+				 * They'll be copied again later; see TODO above.
+				 */
+				copy_configuration_files(true);
+			}
 		}
 
 
@@ -597,7 +623,12 @@ do_standby_clone(void)
 	 */
 	if (runtime_options.copy_external_config_files == true && config_files.entries > 0)
 	{
-		copy_configuration_files();
+		/*
+		 * If "--copy-external-config-files=samepath" was used, the files will already
+		 * have been copied.
+		 */
+		if (runtime_options.copy_external_config_files_destination == CONFIG_FILE_PGDATA)
+			copy_configuration_files(false);
 	}
 
 	/* Write the recovery.conf file */
@@ -938,7 +969,6 @@ _do_create_recovery_conf(void)
 			log_detail("%s", PQerrorMessage(source_conn));
 		}
 
-
 		exit(ERR_BAD_CONFIG);
 	}
 
@@ -955,7 +985,10 @@ _do_create_recovery_conf(void)
 		{
 			log_detail("%s", PQerrorMessage(source_conn));
 		}
-
+		else
+		{
+			log_hint(_("standby must be registered before a new recovery.conf file can be created"));
+		}
 
 		exit(ERR_BAD_CONFIG);
 	}
@@ -1021,6 +1054,7 @@ _do_create_recovery_conf(void)
 								  local_node_record.slot_name,
 								  upstream_node_record.node_name,
 								  upstream_node_id);
+
 				if (runtime_options.force == false && runtime_options.dry_run == false)
 				{
 					log_error("%s", msg.data);
@@ -1052,7 +1086,7 @@ _do_create_recovery_conf(void)
 				initPQExpBuffer(&msg);
 
 				appendPQExpBuffer(&msg,
-								  _("insufficient free replicaiton slots on upstream node \"%s\" (ID: %i)"),
+								  _("insufficient free replication slots on upstream node \"%s\" (ID: %i)"),
 								  upstream_node_record.node_name,
 								  upstream_node_id);
 
@@ -1108,14 +1142,14 @@ _do_create_recovery_conf(void)
 	if (runtime_options.dry_run == true)
 	{
 		char		recovery_conf_contents[MAXLEN] = "";
-		create_recovery_file(&upstream_node_record, &recovery_conninfo, recovery_conf_contents, false);
+		create_recovery_file(&local_node_record, &recovery_conninfo, recovery_conf_contents, false);
 
 		log_info(_("would create \"recovery.conf\" file in \"%s\""), local_data_directory);
 		log_detail(_("\n%s"), recovery_conf_contents);
 	}
 	else
 	{
-		if (!create_recovery_file(&upstream_node_record, &recovery_conninfo, local_data_directory, true))
+		if (!create_recovery_file(&local_node_record, &recovery_conninfo, local_data_directory, true))
 		{
 			log_error(_("unable to create \"recovery.conf\""));
 		}
@@ -1675,11 +1709,16 @@ do_standby_register(void)
 
 	termPQExpBuffer(&details);
 
-	/* if --wait-sync option set, wait for the records to synchronise */
+	/*
+	 * If --wait-sync option set, wait for the records to synchronise
+	 * (unless 0 seconds provided, which disables it, which is the same as
+	 *  not providing the option). The default value is -1, which means
+	 * no timeout.
+	 */
 
 	if (PQstatus(conn) == CONNECTION_OK &&
 		runtime_options.wait_register_sync == true &&
-		runtime_options.wait_register_sync_seconds > 0)
+		runtime_options.wait_register_sync_seconds != 0)
 	{
 		bool		sync_ok = false;
 		int			timer = 0;
@@ -1703,7 +1742,11 @@ do_standby_register(void)
 		{
 			bool		records_match = true;
 
-			if (runtime_options.wait_register_sync_seconds && runtime_options.wait_register_sync_seconds == timer)
+			/*
+			 * If timeout set to a positive value, check if we've reached it and
+			 * exit the loop
+			 */
+			if (runtime_options.wait_register_sync_seconds > 0 && runtime_options.wait_register_sync_seconds == timer)
 				break;
 
 			node_record_status = get_node_record(conn,
@@ -2126,7 +2169,13 @@ do_standby_follow(void)
 
 	log_verbose(LOG_DEBUG, "do_standby_follow()");
 
-	local_conn = establish_db_connection(config_file_options.conninfo, true);
+	local_conn = establish_db_connection(config_file_options.conninfo, false);
+
+	if (PQstatus(local_conn) != CONNECTION_OK)
+	{
+		log_hint(_("use \"repmgr node rejoin\" to re-add an inactive node to the replication cluster"));
+		exit(ERR_DB_CONN);
+	}
 
 	log_verbose(LOG_INFO, _("connected to local node"));
 
@@ -2218,7 +2267,7 @@ do_standby_follow(void)
 
 	if (config_file_options.use_replication_slots)
 	{
-		int free_slots = get_free_replication_slots(primary_conn);
+		int free_slots = get_free_replication_slot_count(primary_conn);
 		if (free_slots < 0)
 		{
 			log_error(_("unable to determine number of free replication slots on the primary"));
@@ -2313,6 +2362,74 @@ do_standby_follow(void)
 										 &follow_output,
 										 &follow_error_code);
 
+	/* unable to restart the standby */
+	if (success == false)
+	{
+		create_event_notification_extended(
+			primary_conn,
+			&config_file_options,
+			config_file_options.node_id,
+			"standby_follow",
+			success,
+			follow_output.data,
+			&event_info);
+
+		PQfinish(primary_conn);
+
+		log_notice(_("STANDBY FOLLOW failed"));
+		if (strlen( follow_output.data ))
+			log_detail("%s", follow_output.data);
+
+		termPQExpBuffer(&follow_output);
+		exit(follow_error_code);
+	}
+
+	termPQExpBuffer(&follow_output);
+
+	initPQExpBuffer(&follow_output);
+
+	/*
+	 * Wait up to "standby_follow_timeout" seconds for standby to connect to
+	 * upstream.
+	 * For 9.6 and later, we could check pg_stat_wal_receiver on the local node.
+	 */
+
+	/* assume success, necessary if standby_follow_timeout is zero */
+	success = true;
+
+	for (timer = 0; timer < config_file_options.standby_follow_timeout; timer++)
+	{
+		success = is_downstream_node_attached(primary_conn, config_file_options.node_name);
+		if (success == true)
+			break;
+
+		log_verbose(LOG_DEBUG, "sleeping %i of max %i seconds waiting for standby to attach to primary",
+					timer + 1,
+					config_file_options.standby_follow_timeout);
+		sleep(1);
+	}
+
+	if (success == true)
+	{
+		log_notice(_("STANDBY FOLLOW successful"));
+		appendPQExpBuffer(&follow_output,
+						  "standby attached to upstream node \"%s\" (node ID: %i)",
+						  primary_node_record.node_name,
+						  primary_node_id);
+	}
+	else
+	{
+		log_error(_("STANDBY FOLLOW failed"));
+		appendPQExpBuffer(&follow_output,
+						  "standby did not attach to upstream node \"%s\" (node ID: %i) after %i seconds",
+						  primary_node_record.node_name,
+						  primary_node_id,
+						  config_file_options.standby_follow_timeout);
+
+	}
+
+	log_detail("%s", follow_output.data);
+
 	create_event_notification_extended(
 		primary_conn,
 		&config_file_options,
@@ -2324,20 +2441,11 @@ do_standby_follow(void)
 
 	PQfinish(primary_conn);
 
-	if (success == false)
-	{
-		log_notice(_("STANDBY FOLLOW failed"));
-		log_detail("%s", follow_output.data);
-
-		termPQExpBuffer(&follow_output);
-		exit(follow_error_code);
-	}
-
-	log_notice(_("STANDBY FOLLOW successful"));
-	log_detail("%s", follow_output.data);
-
 	termPQExpBuffer(&follow_output);
 
+	if (success == false)
+		exit(ERR_FOLLOW_FAIL);
+
 	return;
 }
 
@@ -2803,6 +2911,25 @@ do_standby_switchover(void)
 		exit(ERR_DB_QUERY);
 	}
 
+	/*
+	 * Check that there's no exclusive backups running on the primary.
+	 * We don't want to end up damaging the backup and also leaving the server in an
+	 * state where there's control data saying it's in backup mode but there's no
+	 * backup_label in PGDATA.
+	 * If the DBA wants to do the switchover anyway, he should first stop the
+	 * backup that's running.
+	 */
+	if (server_in_exclusive_backup_mode(remote_conn) != BACKUP_STATE_NO_BACKUP)
+	{
+		log_error(_("unable to perform a switchover while primary server is in exclusive backup mode"));
+		log_hint(_("stop backup before attempting the switchover"));
+
+		PQfinish(local_conn);
+		PQfinish(remote_conn);
+
+		exit(ERR_SWITCHOVER_FAIL);
+	}
+
 	/*
 	 * Check this standby is attached to the demotion candidate
 	 * TODO:
@@ -3335,8 +3462,6 @@ do_standby_switchover(void)
 				}
 			}
 
-
-
 			/*
 			 * check there are sufficient free walsenders - obviously there's potential
 			 * for a later race condition if some walsenders come into use before the
@@ -3760,7 +3885,6 @@ do_standby_switchover(void)
 	 * If --siblings-follow specified, attempt to make them follow the new
 	 * primary
 	 */
-
 	if (runtime_options.siblings_follow == true && sibling_nodes.node_count > 0)
 	{
 		int			failed_follow_count = 0;
@@ -3787,8 +3911,17 @@ do_standby_switchover(void)
 			initPQExpBuffer(&remote_command_str);
 			make_remote_repmgr_path(&remote_command_str, &sibling_node_record);
 
-			appendPQExpBuffer(&remote_command_str,
-							  "standby follow 2>/dev/null && echo \"1\" || echo \"0\"");
+			if (sibling_node_record.type == WITNESS)
+			{
+				appendPQExpBuffer(&remote_command_str,
+								  "witness register -d \\'%s\\' --force 2>/dev/null && echo \"1\" || echo \"0\"",
+								  local_node_record.conninfo);
+			}
+			else
+			{
+				appendPQExpBuffer(&remote_command_str,
+								  "standby follow 2>/dev/null && echo \"1\" || echo \"0\"");
+			}
 			get_conninfo_value(cell->node_info->conninfo, "host", host);
 			log_debug("executing:\n  %s", remote_command_str.data);
 
@@ -3803,8 +3936,16 @@ do_standby_switchover(void)
 
 			if (success == false || command_output.data[0] == '0')
 			{
-				log_warning(_("STANDBY FOLLOW failed on node \"%s\""),
-							cell->node_info->node_name);
+				if (sibling_node_record.type == WITNESS)
+				{
+					log_warning(_("WITNESS REGISTER failed on node \"%s\""),
+								cell->node_info->node_name);
+				}
+				else
+				{
+					log_warning(_("STANDBY FOLLOW failed on node \"%s\""),
+								cell->node_info->node_name);
+				}
 				failed_follow_count++;
 			}
 
@@ -3909,6 +4050,8 @@ check_source_server()
 	PGconn	   *privileged_conn = NULL;
 
 	char		cluster_size[MAXLEN];
+	char	   *connstr = NULL;
+
 	t_node_info node_record = T_NODE_INFO_INITIALIZER;
 	RecordStatus record_status = RECORD_NOT_FOUND;
 	ExtensionStatus extension_status = REPMGR_UNKNOWN;
@@ -3917,8 +4060,11 @@ check_source_server()
 	log_verbose(LOG_DEBUG, "check_source_server()");
 	log_info(_("connecting to source node"));
 
-	source_conn = establish_db_connection_by_params(&source_conninfo, false);
+	connstr = param_list_to_string(&source_conninfo);
+	log_detail(_("connection string is: %s"), connstr);
+	pfree(connstr);
 
+	source_conn = establish_db_connection_by_params(&source_conninfo, false);
 	/*
 	 * Unless in barman mode, exit with an error;
 	 * establish_db_connection_by_params() will have already logged an error
@@ -4073,13 +4219,25 @@ check_source_server()
 		if (record_status == RECORD_FOUND)
 		{
 			t_conninfo_param_list upstream_conninfo = T_CONNINFO_PARAM_LIST_INITIALIZER;
+			char	   *upstream_conninfo_user;
 
 			initialize_conninfo_params(&upstream_conninfo, false);
 			parse_conninfo_string(node_record.conninfo, &upstream_conninfo, NULL, false);
 
 			strncpy(recovery_conninfo_str, node_record.conninfo, MAXLEN);
 			strncpy(upstream_repluser, node_record.repluser, NAMEDATALEN);
-			strncpy(upstream_user, param_get(&upstream_conninfo, "user"), NAMEDATALEN);
+
+			upstream_conninfo_user = param_get(&upstream_conninfo, "user");
+			if (upstream_conninfo_user != NULL)
+			{
+				strncpy(upstream_user, upstream_conninfo_user, NAMEDATALEN);
+			}
+			else
+			{
+				get_conninfo_default_value("user", upstream_user, NAMEDATALEN);
+			}
+
+			log_verbose(LOG_DEBUG, "upstream_user is \"%s\"", upstream_user);
 
 			upstream_conninfo_found = true;
 		}
@@ -4632,7 +4790,7 @@ initialise_direct_clone(t_node_info *node_record)
 		}
 		else
 		{
-			TablespaceListCell *cell = false;
+			TablespaceListCell *cell;
 			KeyValueList not_found = {NULL, NULL};
 			int			total = 0,
 						matched = 0;
@@ -5690,7 +5848,7 @@ get_barman_property(char *dst, char *name, char *local_repmgr_directory)
 
 
 static void
-copy_configuration_files(void)
+copy_configuration_files(bool delete_after_copy)
 {
 	int			i,
 				r;
@@ -5735,13 +5893,35 @@ copy_configuration_files(void)
 		r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
 							  file->filepath, dest_path.data, false, source_server_version_num);
 
-		termPQExpBuffer(&dest_path);
+		/*
+		 * TODO: collate errors into list
+		 */
 
 		if (WEXITSTATUS(r))
 		{
 			log_error(_("standby clone: unable to copy config file \"%s\""),
 					  file->filename);
+			log_hint(_("see preceding messages for details"));
+
+			if (runtime_options.force == false)
+				exit(ERR_BAD_RSYNC);
 		}
+
+		/*
+		 * This is to check we can actually copy the files before running the
+		 * main clone operation
+		 */
+		if (delete_after_copy == true)
+		{
+			/* this is very unlikely to happen, but log in case it does */
+			if (unlink(dest_path.data) < 0 && errno != ENOENT)
+			{
+				log_warning(_("unable to delete %s"), dest_path.data);
+				log_detail("%s", strerror(errno));
+			}
+		}
+
+		termPQExpBuffer(&dest_path);
 	}
 
 	return;
@@ -6353,6 +6533,7 @@ do_standby_help(void)
 	puts("");
 	printf(_("  \"standby clone\" clones a standby from the primary or an upstream node.\n"));
 	puts("");
+	printf(_("  -d, --dbname=conninfo               conninfo of the upstream node to use for cloning.\n"));
 	printf(_("  -c, --fast-checkpoint               force fast checkpoint\n"));
 	printf(_("  --copy-external-config-files[={samepath|pgdata}]\n" \
 			 "                                      copy configuration files located outside the \n" \

repmgr-action-witness.c

@@ -137,7 +137,7 @@ do_witness_register(void)
 	}
 
 	/*
-	 * TODO:sanity check witness node is not part of main cluster; we could
+	 * TODO: sanity check witness node is not part of main cluster; we could
 	 * add a random application_name to the respective connections,
 	 * and do a simple check of pg_stat_activity
 	 */
@@ -193,8 +193,26 @@ do_witness_register(void)
 		}
 	}
 
+	/*
+	 * Check that an active node with the same node_name doesn't exist already
+	 */
 
-	// XXX check other node with same name does not exist
+	record_status = get_node_record_by_name(primary_conn,
+											config_file_options.node_name,
+											&node_record);
+
+
+	if (record_status == RECORD_FOUND)
+	{
+		if (node_record.active == true && node_record.node_id != config_file_options.node_id)
+		{
+			log_error(_("node %i exists already with node_name \"%s\""),
+					  node_record.node_id,
+					  config_file_options.node_name);
+			PQfinish(primary_conn);
+			exit(ERR_BAD_CONFIG);
+		}
+	}
 
 	/*
 	 * if repmgr.nodes contains entries, delete if -F/--force provided,
@@ -225,6 +243,7 @@ do_witness_register(void)
 		PQfinish(witness_conn);
 		exit(SUCCESS);
 	}
+
 	/* create record on primary */
 
 	/*
@@ -291,55 +310,59 @@ do_witness_register(void)
 void
 do_witness_unregister(void)
 {
-	PGconn	   *witness_conn = NULL;
+	PGconn	   *local_conn = NULL;
 	PGconn	   *primary_conn = NULL;
 	t_node_info node_record = T_NODE_INFO_INITIALIZER;
 	RecordStatus record_status = RECORD_NOT_FOUND;
 	bool		node_record_deleted = false;
-	bool		witness_available = true;
+	bool		local_node_available = true;
+	int			witness_node_id = UNKNOWN_NODE_ID;
 
-	log_info(_("connecting to witness node \"%s\" (ID: %i)"),
+	if (runtime_options.node_id != UNKNOWN_NODE_ID)
+	{
+		/* user has specified the witness node id */
+		witness_node_id = runtime_options.node_id;
+	}
+	else
+	{
+		/* assume witness node is local node */
+		witness_node_id = config_file_options.node_id;
+	}
+
+	log_info(_("connecting to node \"%s\" (ID: %i)"),
 			 config_file_options.node_name,
 			 config_file_options.node_id);
 
-	witness_conn = establish_db_connection_quiet(config_file_options.conninfo);
+	local_conn = establish_db_connection_quiet(config_file_options.conninfo);
 
-	if (PQstatus(witness_conn) != CONNECTION_OK)
+	if (PQstatus(local_conn) != CONNECTION_OK)
 	{
 		if (!runtime_options.force)
 		{
-			log_error(_("unable to connect to witness node \"%s\" (ID: %i)"),
+			log_error(_("unable to connect to node \"%s\" (ID: %i)"),
 					  config_file_options.node_name,
 					  config_file_options.node_id);
-			log_detail("%s", PQerrorMessage(witness_conn));
-			log_hint(_("provide -F/--force to remove the witness record if the server is not running"));
+			log_detail("%s", PQerrorMessage(local_conn));
 			exit(ERR_BAD_CONFIG);
 		}
 
 		log_notice(_("unable to connect to witness node \"%s\" (ID: %i), removing node record on cluster primary only"),
 				   config_file_options.node_name,
 				   config_file_options.node_id);
-		witness_available = false;
+		local_node_available = false;
 	}
 
-	if (witness_available == true)
+	if (local_node_available == true)
 	{
-		primary_conn = get_primary_connection_quiet(witness_conn, NULL, NULL);
+		primary_conn = get_primary_connection_quiet(local_conn, NULL, NULL);
 	}
 	else
 	{
 		/*
-		 * Extract the repmgr user and database names from the conninfo string
-		 * provided in repmgr.conf
+		 * Assume user has provided connection details for the primary server
 		 */
-		get_conninfo_value(config_file_options.conninfo, "user", repmgr_user);
-		get_conninfo_value(config_file_options.conninfo, "dbname", repmgr_db);
-
-		param_set_ine(&source_conninfo, "user", repmgr_user);
-		param_set_ine(&source_conninfo, "dbname", repmgr_db);
 
 		primary_conn = establish_db_connection_by_params(&source_conninfo, false);
-
 	}
 
 	if (PQstatus(primary_conn) != CONNECTION_OK)
@@ -347,26 +370,26 @@ do_witness_unregister(void)
 		log_error(_("unable to connect to primary"));
 		log_detail("%s", PQerrorMessage(primary_conn));
 
-		if (witness_available == true)
+		if (local_node_available == true)
 		{
-			PQfinish(witness_conn);
+			PQfinish(local_conn);
 		}
-		else
+		else if (runtime_options.connection_param_provided == false)
 		{
-			log_hint(_("provide connection details to primary server"));
+			log_hint(_("provide connection details for the primary server"));
 		}
 		exit(ERR_BAD_CONFIG);
 	}
 
 	/* Check node exists and is really a witness */
-	record_status = get_node_record(primary_conn, config_file_options.node_id, &node_record);
+	record_status = get_node_record(primary_conn, witness_node_id, &node_record);
 
 	if (record_status != RECORD_FOUND)
 	{
-		log_error(_("no record found for node %i"), config_file_options.node_id);
+		log_error(_("no record found for node %i"), witness_node_id);
 
-		if (witness_available == true)
-			PQfinish(witness_conn);
+		if (local_node_available == true)
+			PQfinish(local_conn);
 		PQfinish(primary_conn);
 
 		exit(ERR_BAD_CONFIG);
@@ -374,11 +397,17 @@ do_witness_unregister(void)
 
 	if (node_record.type != WITNESS)
 	{
+		/*
+		 * The node (either explicitly provided with --node-id, or the local node)
+		 * is not a witness.
+		 *
+		 * TODO: scan node list and print hint about identity of known witness servers.
+		 */
 		log_error(_("node %i is not a witness node"), config_file_options.node_id);
 		log_detail(_("node %i is a %s node"), config_file_options.node_id, get_node_type_string(node_record.type));
 
-		if (witness_available == true)
-			PQfinish(witness_conn);
+		if (local_node_available == true)
+			PQfinish(local_conn);
 		PQfinish(primary_conn);
 
 		exit(ERR_BAD_CONFIG);
@@ -387,49 +416,43 @@ do_witness_unregister(void)
 	if (runtime_options.dry_run == true)
 	{
 		log_info(_("prerequisites for unregistering the witness node are met"));
-		if (witness_available == true)
-			PQfinish(witness_conn);
+		if (local_node_available == true)
+			PQfinish(local_conn);
 		PQfinish(primary_conn);
 
 		exit(SUCCESS);
 	}
 
-	log_info(_("unregistering witness node %i"), config_file_options.node_id);
+	log_info(_("unregistering witness node %i"), witness_node_id);
 	node_record_deleted = delete_node_record(primary_conn,
-										     config_file_options.node_id);
+										     witness_node_id);
 
 	if (node_record_deleted == false)
 	{
 		PQfinish(primary_conn);
-		PQfinish(witness_conn);
-		exit(ERR_BAD_CONFIG);
-	}
 
-	/* sync records from primary */
-	if (witness_available == true && witness_copy_node_records(primary_conn, witness_conn) == false)
-	{
-		log_error(_("unable to copy repmgr node records from primary"));
-		PQfinish(primary_conn);
-		PQfinish(witness_conn);
+		if (local_node_available == true)
+			PQfinish(local_conn);
+		PQfinish(local_conn);
 		exit(ERR_BAD_CONFIG);
 	}
 
 	/* Log the event */
 	create_event_record(primary_conn,
 						&config_file_options,
-						config_file_options.node_id,
+						witness_node_id,
 						"witness_unregister",
 						true,
 						NULL);
 
 	PQfinish(primary_conn);
 
-	if (witness_available == true)
-		PQfinish(witness_conn);
+	if (local_node_available == true)
+		PQfinish(local_conn);
 
 	log_info(_("witness unregistration complete"));
-	log_detail(_("witness node with id %i (conninfo: %s) successfully unregistered"),
-			    config_file_options.node_id, config_file_options.conninfo);
+	log_detail(_("witness node with ID %i successfully unregistered"),
+			    witness_node_id);
 
 	return;
 }
@@ -449,16 +472,19 @@ void do_witness_help(void)
 	puts("");
 	printf(_("  Requires provision of connection information for the primary\n"));
 	puts("");
-	printf(_("  --dry-run                           check prerequisites but don't make any changes\n"));
-	printf(_("  -F, --force                         overwrite an existing node record\n"));
+	printf(_("  --dry-run                check prerequisites but don't make any changes\n"));
+	printf(_("  -F, --force              overwrite an existing node record\n"));
 	puts("");
 
 	printf(_("WITNESS UNREGISTER\n"));
 	puts("");
 	printf(_("  \"witness register\" unregisters a witness node.\n"));
 	puts("");
-	printf(_("  --dry-run                           check prerequisites but don't make any changes\n"));
-	printf(_("  -F, --force                         unregister when witness node not running\n"));
+	printf(_("  --dry-run                check prerequisites but don't make any changes\n"));
+	printf(_("  -F, --force              unregister when witness node not running\n"));
+	printf(_("  --node-id                node ID of the witness node (provide if executing on\n"));
+	printf(_("                             another node)\n"));
+
 	puts("");
 
 	return;

repmgr-client-global.h

@@ -47,6 +47,7 @@ typedef struct
 	/* logging options */
 	char		log_level[MAXLEN];	/* overrides setting in repmgr.conf */
 	bool		log_to_file;
+	bool		quiet;
 	bool		terse;
 	bool		verbose;
 
@@ -106,6 +107,7 @@ typedef struct
 	bool		replication_lag;
 	bool		role;
 	bool		slots;
+	bool		missing_slots;
 	bool		has_passfile;
 	bool		replication_connection;
 
@@ -137,7 +139,7 @@ typedef struct
 		/* general configuration options */	\
 		"", false, false, "", false, false,	\
 		/* logging options */ \
-		"", false, false, false, \
+		"", false, false, false, false,	\
 		/* output options */ \
 		false, false, false,  \
 		/* database connection options */ \
@@ -152,13 +154,13 @@ typedef struct
 		/* "standby clone"/"standby follow" options */ \
 		NO_UPSTREAM_NODE, \
 		/* "standby register" options */ \
-		false, 0, DEFAULT_WAIT_START,   \
+		false, -1, DEFAULT_WAIT_START,   \
 		/* "standby switchover" options */ \
 		false, false, "", false,		   \
 		/* "node status" options */ \
 		false, \
 		/* "node check" options */ \
-		false, false, false, false, false, false, false, \
+		false, false, false, false, false, false, false, false,	\
 		/* "node join" options */ \
 		"", \
 		/* "node service" options */ \

repmgr-client.c

@@ -98,7 +98,7 @@ main(int argc, char **argv)
 {
 	t_conninfo_param_list default_conninfo = T_CONNINFO_PARAM_LIST_INITIALIZER;
 
-	int			optindex;
+	int			optindex = 0;
 	int			c;
 
 	char	   *repmgr_command = NULL;
@@ -108,6 +108,7 @@ main(int argc, char **argv)
 	char	   *dummy_action = "";
 
 	bool		help_option = false;
+	bool		option_error_found = false;
 
 	set_progname(argv[0]);
 
@@ -178,7 +179,10 @@ main(int argc, char **argv)
 		strncpy(runtime_options.username, pw->pw_name, MAXLEN);
 	}
 
-	while ((c = getopt_long(argc, argv, "?Vb:f:FwWd:h:p:U:R:S:D:ck:L:tvC:", long_options,
+	/* Make getopt emitting errors */
+	opterr = 1;
+
+	while ((c = getopt_long(argc, argv, "?Vb:f:FwWd:h:p:U:R:S:D:ck:L:qtvC:", long_options,
 							&optindex)) != -1)
 	{
 		/*
@@ -196,13 +200,7 @@ main(int argc, char **argv)
 			case OPT_HELP:		/* --help */
 				help_option = true;
 				break;
-			case '?':
-				/* Actual help option given */
-				if (strcmp(argv[optind - 1], "-?") == 0)
-				{
-					help_option = true;
-				}
-				break;
+
 			case 'V':
 
 				/*
@@ -473,6 +471,10 @@ main(int argc, char **argv)
 				runtime_options.slots = true;
 				break;
 
+			case OPT_MISSING_SLOTS:
+				runtime_options.missing_slots = true;
+				break;
+
 			case OPT_HAS_PASSFILE:
 				runtime_options.has_passfile = true;
 				break;
@@ -572,6 +574,12 @@ main(int argc, char **argv)
 				logger_output_mode = OM_DAEMON;
 				break;
 
+
+				/* --quiet */
+			case 'q':
+				runtime_options.quiet = true;
+				break;
+
 				/* --terse */
 			case 't':
 				runtime_options.terse = true;
@@ -627,14 +635,29 @@ main(int argc, char **argv)
 								 _("--recovery-min-apply-delay is now a configuration file parameter, \"recovery_min_apply_delay\""));
 				break;
 
+			case ':':   /* missing option argument */
+				option_error_found = true;
+				break;
+			case '?':
+				/* Actual help option given? */
+				if (strcmp(argv[optind - 1], "-?") == 0)
+				{
+					help_option = true;
+					break;
+				}
+				/* otherwise fall through to default */
+			default:    /* invalid option */
+				option_error_found = true;
+				break;
 		}
 	}
 
+
 	/*
 	 * If -d/--dbname appears to be a conninfo string, validate by attempting
 	 * to parse it (and if successful, store the parsed parameters)
 	 */
-	if (runtime_options.dbname)
+	if (runtime_options.dbname[0])
 	{
 		if (strncmp(runtime_options.dbname, "postgresql://", 13) == 0 ||
 			strncmp(runtime_options.dbname, "postgres://", 11) == 0 ||
@@ -730,9 +753,10 @@ main(int argc, char **argv)
 	if (cli_errors.head != NULL)
 	{
 		free_conninfo_params(&source_conninfo);
-		exit_with_cli_errors(&cli_errors);
+		exit_with_cli_errors(&cli_errors, NULL);
 	}
 
+
 	/*----------
 	 * Determine the node type and action; following are valid:
 	 *
@@ -979,9 +1003,30 @@ main(int argc, char **argv)
 	if (cli_errors.head != NULL)
 	{
 		free_conninfo_params(&source_conninfo);
-		exit_with_cli_errors(&cli_errors);
+
+		exit_with_cli_errors(&cli_errors, valid_repmgr_command_found == true ? repmgr_command : NULL);
 	}
 
+	/* no errors detected by repmgr, but getopt might have */
+	if (option_error_found == true)
+	{
+		if (valid_repmgr_command_found == true)
+		{
+			printf(_("Try \"%s --help\" or \"%s %s --help\" for more information.\n"),
+				   progname(),
+				   progname(),
+				   repmgr_command);
+		}
+		else
+		{
+			printf(_("Try \"repmgr --help\" for more information.\n"));
+		}
+
+		free_conninfo_params(&source_conninfo);
+		exit(ERR_BAD_CONFIG);
+	}
+
+
 	/*
 	 * Print any warnings about inappropriate command line options, unless
 	 * -t/--terse set
@@ -1010,7 +1055,6 @@ main(int argc, char **argv)
 		runtime_options.output_mode = OM_OPTFORMAT;
 	}
 
-
 	/*
 	 * Check for configuration file items which can be overriden by runtime
 	 * options
@@ -1068,6 +1112,28 @@ main(int argc, char **argv)
 	if (runtime_options.terse)
 		logger_set_terse();
 
+	/*
+	 * If --dry-run specified, ensure log_level is at least LOG_INFO, regardless
+	 * of what's in the configuration file or -L/--log-level paremeter, otherwise
+	 * some or output might not be displayed.
+	 */
+	if (runtime_options.dry_run == true)
+	{
+		logger_set_min_level(LOG_INFO);
+	}
+
+	/*
+	 * If -q/--quiet supplied, suppress any non-ERROR log output.
+	 * This overrides everything else; we'll leave it up to the user to deal with the
+	 * consequences of e.g. running --dry-run together with -q/--quiet.
+	 */
+	if (runtime_options.quiet == true)
+	{
+		logger_set_level(LOG_ERROR);
+	}
+
+
+
 	/*
 	 * Node configuration information is not needed for all actions, with
 	 * STANDBY CLONE being the main exception.
@@ -1453,6 +1519,7 @@ check_cli_parameters(const int action)
 		{
 			case PRIMARY_UNREGISTER:
 			case STANDBY_UNREGISTER:
+			case WITNESS_UNREGISTER:
 			case CLUSTER_EVENT:
 			case CLUSTER_MATRIX:
 			case CLUSTER_CROSSCHECK:
@@ -1493,6 +1560,7 @@ check_cli_parameters(const int action)
 			case STANDBY_CLONE:
 			case STANDBY_REGISTER:
 			case STANDBY_FOLLOW:
+			case BDR_REGISTER:
 				break;
 			default:
 				item_list_append_format(&cli_warnings,
@@ -1835,7 +1903,7 @@ do_help(void)
 	printf(_("    %s [OPTIONS] standby {register|unregister|clone|promote|follow|switchover}\n"), progname());
 	printf(_("    %s [OPTIONS] bdr     {register|unregister}\n"), progname());
 	printf(_("    %s [OPTIONS] node    {status|check|rejoin|service}\n"), progname());
-	printf(_("    %s [OPTIONS] cluster {show|event|matrix|crosscheck}\n"), progname());
+	printf(_("    %s [OPTIONS] cluster {show|event|matrix|crosscheck|cleanup}\n"), progname());
 	printf(_("    %s [OPTIONS] witness {register|unregister}\n"), progname());
 
 	puts("");
@@ -1884,6 +1952,7 @@ do_help(void)
 	printf(_("  --dry-run                           show what would happen for action, but don't execute it\n"));
 	printf(_("  -L, --log-level                     set log level (overrides configuration file; default: NOTICE)\n"));
 	printf(_("  --log-to-file                       log to file (or logging facility) defined in repmgr.conf\n"));
+	printf(_("  -q, --quiet                         suppress all log output apart from errors\n"));
 	printf(_("  -t, --terse                         don't display detail, hints and other non-critical output\n"));
 	printf(_("  -v, --verbose                       display additional log output (useful for debugging)\n"));
 

repmgr-client.h

@@ -87,6 +87,7 @@
 #define OPT_REMOTE_NODE_ID                 1038
 #define OPT_RECOVERY_CONF_ONLY             1039
 #define OPT_NO_WAIT                        1040
+#define OPT_MISSING_SLOTS                  1041
 
 /* deprecated since 3.3 */
 #define OPT_DATA_DIR						999
@@ -125,6 +126,7 @@ static struct option long_options[] =
 /* logging options */
 	{"log-level", required_argument, NULL, 'L'},
 	{"log-to-file", no_argument, NULL, OPT_LOG_TO_FILE},
+	{"quiet",  no_argument, NULL, 'q'},
 	{"terse", no_argument, NULL, 't'},
 	{"verbose", no_argument, NULL, 'v'},
 
@@ -164,6 +166,7 @@ static struct option long_options[] =
 	{"replication-lag", no_argument, NULL, OPT_REPLICATION_LAG},
 	{"role", no_argument, NULL, OPT_ROLE},
 	{"slots", no_argument, NULL, OPT_SLOTS},
+	{"missing-slots", no_argument, NULL, OPT_MISSING_SLOTS},
 	{"has-passfile", no_argument, NULL, OPT_HAS_PASSFILE},
 	{"replication-connection", no_argument, NULL, OPT_REPL_CONN},
 

repmgr.conf.sample

@@ -98,7 +98,7 @@
 #log_facility=STDERR		 # Logging facility: possible values are STDERR, or for
 				 # syslog integration, one of LOCAL0, LOCAL1, ..., LOCAL7, USER
 
-#log_file=''			 # stderr can be redirected to an arbitrary file:
+#log_file=''			 # STDERR can be redirected to an arbitrary file
 #log_status_interval=300	 # interval (in seconds) for repmgrd to log a status message
 
 
@@ -207,16 +207,40 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 
 
 #------------------------------------------------------------------------------
-# Standby follow settings
+# "standby follow" settings
 #------------------------------------------------------------------------------
 
 # These settings apply when instructing a standby to follow the new primary
 # ("repmgr standby follow").
 
-#primary_follow_timeout=60		# The length of time (in seconds) to wait
+#primary_follow_timeout=60		# The max length of time (in seconds) to wait
 					# for the new primary to become available
+#standby_follow_timeout=15		# The max length of time (in seconds) to wait
+					# for the standby to connect to the primary
 
 
+#------------------------------------------------------------------------------
+# "standby switchover" settings
+#------------------------------------------------------------------------------
+
+# These settings apply when switching roles between a primary and a standby
+# ("repmgr standby switchover").
+
+#standby_reconnect_timeout=60		# The max length of time (in seconds) to wait
+					# for the demoted standby to reconnect to the promoted
+					# primary (note: this value should be equal to or greater
+					# than that set for "node_rejoin_timeout")
+
+#------------------------------------------------------------------------------
+# "node rejoin" settings
+#------------------------------------------------------------------------------
+
+# These settings apply when reintegrating a node into a replication cluster
+# with "repmgrd_node_rejoin"
+
+#node_rejoin_timeout=60		# The maximum length of time (in seconds) to wait for
+					# the node to reconnect to the replication cluster
+
 #------------------------------------------------------------------------------
 # Barman options
 #------------------------------------------------------------------------------
@@ -234,6 +258,11 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 # These settings are only applied when repmgrd is running. Values shown
 # are defaults.
 
+#repmgrd_pid_file=			# Path of PID file to use for repmgrd; if not set, a PID file will
+                                        # be generated in a temporary directory specified by the environment
+                                        # variable $TMPDIR, or if not set, in "/tmp". This value can be overridden
+                                        # by the command line option "-p/--pid-file"; the command line option
+                                        # "--no-pid-file" will force PID file creation to be skipped.
 #failover=manual			# one of 'automatic', 'manual'.
 					# determines what action to take in the event of upstream failure
 					#
@@ -243,7 +272,7 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 					#    manual attention to reattach it to replication
 					# (does not apply to BDR mode)
 
-#priority=100				# indicate a preferred priorty for promoting nodes;
+#priority=100				# indicate a preferred priority for promoting nodes;
 					# a value of zero prevents the node being promoted to primary
 					# (default: 100)
 
@@ -251,11 +280,11 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 					# primary (or other upstream node)
 #reconnect_interval=10			# Interval between attempts  to reconnect to an unreachable
 					# primary (or other upstream node)
-#promote_command=			# command to execute when promoting a new primary; use something like:
+#promote_command=			# command repmgrd executes when promoting a new primary; use something like:
 					#
 					#     repmgr standby promote -f /etc/repmgr.conf
 					#
-#follow_command=			# command to execute when instructing a standby to follow a new primary;
+#follow_command=			# command repmgrd executes when instructing a standby to follow a new primary;
 					# use something like:
 					#
 					#     repmgr standby follow -f /etc/repmgr.conf -W --upstream-node-id=%n
@@ -263,8 +292,9 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 #primary_notification_timeout=60	# Interval (in seconds) which repmgrd on a standby
 					# will wait for a notification from the new primary,
 					# before falling back to degraded monitoring
-#standby_reconnect_timeout=60	# Interval (in seconds) which repmgrd on a standby will wait
-					# to reconnect to the local node after executing "follow_command"
+#repmgrd_standby_startup_timeout=60	# Interval (in seconds) which repmgrd on a standby will wait
+					# for the the local node to restart and become ready to accept connections after
+					# executing "follow_command" (defaults to the value set in "standby_reconnect_timeout")
 
 #monitoring_history=no                  # Whether to write monitoring data to the "montoring_history" table
 #monitor_interval_secs=2                # Interval (in seconds) at which to write monitoring data
@@ -308,11 +338,11 @@ ssh_options='-q -o ConnectTimeout=10'	# Options to append to "ssh"
 #service_stop_command = ''
 #service_restart_command = ''
 #service_reload_command = ''
-#service_promote_command = ''		# Note: this overrides any value contained in the setting
-					# "promote_command". This is intended for systems which
-					# provide a package-level promote command, such as Debian's
-					# "pg_ctlcluster"
-
+#service_promote_command = ''		# This parameter is intended for systems which provide a
+					# package-level promote command, such as Debian's
+					# "pg_ctlcluster". *IMPORTANT*: it is *not* a substitute
+					# for "promote_command"; do not use "repmgr standby promote"
+					# (or a script which executes "repmgr standby promote") here.
 
 #------------------------------------------------------------------------------
 # Status check thresholds

repmgr.control

@@ -1,6 +1,6 @@
 # repmgr extension
 comment = 'Replication manager for PostgreSQL'
-default_version = '4.0'
+default_version = '4.1'
 module_pathname = '$libdir/repmgr'
 relocatable = false
 schema = repmgr

repmgr.h

@@ -49,6 +49,8 @@
 #define REPLICATION_TYPE_BDR	  2
 
 #define UNKNOWN_SERVER_VERSION_NUM -1
+#define UNKNOWN_BDR_VERSION_NUM -1
+
 #define UNKNOWN_TIMELINE_ID -1
 #define UNKNOWN_SYSTEM_IDENTIFIER 0
 
@@ -58,6 +60,8 @@
 
 #define VOTING_TERM_NOT_SET -1
 
+#define BDR2_REPLICATION_SET_NAME "repmgr"
+
 /*
  * various default values - ensure repmgr.conf.sample is update
  * if any of these are changed
@@ -70,6 +74,7 @@
 #define DEFAULT_ASYNC_QUERY_TIMEOUT          60  /* seconds */
 #define DEFAULT_PRIMARY_NOTIFICATION_TIMEOUT 60  /* seconds */
 #define DEFAULT_PRIMARY_FOLLOW_TIMEOUT       60  /* seconds */
+#define DEFAULT_STANDBY_FOLLOW_TIMEOUT       30  /* seconds */
 #define DEFAULT_BDR_RECOVERY_TIMEOUT         30  /* seconds */
 #define DEFAULT_ARCHIVE_READY_WARNING        16  /* WAL files */
 #define DEFAULT_ARCHIVE_READY_CRITICAL       128 /* WAL files */
@@ -80,6 +85,7 @@
 #define DEFAULT_PROMOTE_CHECK_TIMEOUT        60  /* seconds */
 #define DEFAULT_PROMOTE_CHECK_INTERVAL       1   /* seconds */
 #define DEFAULT_STANDBY_RECONNECT_TIMEOUT    60  /* seconds */
+#define DEFAULT_NODE_REJOIN_TIMEOUT          60  /* seconds */
 
 #ifndef RECOVERY_COMMAND_FILE
 #define RECOVERY_COMMAND_FILE "recovery.conf"

repmgr_version.h.in

@@ -1,3 +1,2 @@
 #define REPMGR_VERSION_DATE ""
-#define REPMGR_VERSION "4.0.5"
-
+#define REPMGR_VERSION "4.1.0"

repmgrd-physical.c

@@ -58,7 +58,7 @@ static FailoverState failover_state = FAILOVER_STATE_UNKNOWN;
 
 static int	primary_node_id = UNKNOWN_NODE_ID;
 static t_node_info upstream_node_info = T_NODE_INFO_INITIALIZER;
-static NodeInfoList standby_nodes = T_NODE_INFO_LIST_INITIALIZER;
+static NodeInfoList sibling_nodes = T_NODE_INFO_LIST_INITIALIZER;
 
 
 static ElectionResult do_election(void);
@@ -162,8 +162,8 @@ do_physical_node_check(void)
 	if (config_file_options.failover == FAILOVER_AUTOMATIC)
 	{
 		/*
-		 * check that promote/follow commands are defined, otherwise repmgrd
-		 * won't be able to perform any useful action
+		 * Check that "promote_command" and "follow_command" are defined, otherwise repmgrd
+		 * won't be able to perform any useful action in a failover situation.
 		 */
 
 		bool		required_param_missing = false;
@@ -175,14 +175,24 @@ do_physical_node_check(void)
 			if (config_file_options.service_promote_command[0] != '\0')
 			{
 				/*
-				 * if repmgrd executes "service_promote_command" directly,
-				 * repmgr metadata won't get updated
+				 * "service_promote_command" is *not* a substitute for "promote_command";
+				 * it is intended for use in those systems (e.g. Debian) where there's a service
+				 * level promote command (e.g. pg_ctlcluster).
+				 *
+				 * "promote_command" should either execute "repmgr standby promote" directly, or
+				 * a script which executes "repmgr standby promote". This is essential, as the
+				 * repmgr metadata is updated by "repmgr standby promote".
+				 *
+				 * "service_promote_command", if set, will be executed by "repmgr standby promote",
+				 * but never by repmgrd.
+				 *
 				 */
 				log_hint(_("\"service_promote_command\" is set, but can only be executed by \"repmgr standby promote\""));
 			}
 
 			required_param_missing = true;
 		}
+
 		if (config_file_options.follow_command[0] == '\0')
 		{
 			log_error(_("\"follow_command\" must be defined in the configuration file"));
@@ -816,6 +826,29 @@ monitor_streaming_standby(void)
 		{
 			int			degraded_monitoring_elapsed = calculate_elapsed(degraded_monitoring_start);
 
+			if (config_file_options.degraded_monitoring_timeout > 0
+				&& degraded_monitoring_elapsed > config_file_options.degraded_monitoring_timeout)
+			{
+				initPQExpBuffer(&event_details);
+
+				appendPQExpBuffer(&event_details,
+								  _("degraded monitoring timeout (%i seconds) exceeded, terminating"),
+								  degraded_monitoring_elapsed);
+
+				log_notice("%s", event_details.data);
+
+				create_event_notification(NULL,
+										  &config_file_options,
+										  config_file_options.node_id,
+										  "repmgrd_shutdown",
+										  true,
+										  event_details.data);
+
+				termPQExpBuffer(&event_details);
+				terminate(ERR_MONITORING_TIMEOUT);
+			}
+
+
 			log_debug("monitoring node %i in degraded state for %i seconds",
 					  upstream_node_info.node_id,
 					  degraded_monitoring_elapsed);
@@ -918,8 +951,8 @@ monitor_streaming_standby(void)
 						get_active_sibling_node_records(local_conn,
 														local_node_info.node_id,
 														former_upstream_node_id,
-														&standby_nodes);
-						notify_followers(&standby_nodes, local_node_info.node_id);
+														&sibling_nodes);
+						notify_followers(&sibling_nodes, local_node_info.node_id);
 
 						/* this will restart monitoring in primary mode */
 						monitoring_state = MS_NORMAL;
@@ -958,12 +991,12 @@ monitor_streaming_standby(void)
 					get_active_sibling_node_records(local_conn,
 													local_node_info.node_id,
 													local_node_info.upstream_node_id,
-													&standby_nodes);
+													&sibling_nodes);
 
-					if (standby_nodes.node_count > 0)
+					if (sibling_nodes.node_count > 0)
 					{
-						log_debug("scanning %i node records to detect new primary...", standby_nodes.node_count);
-						for (cell = standby_nodes.head; cell; cell = cell->next)
+						log_debug("scanning %i node records to detect new primary...", sibling_nodes.node_count);
+						for (cell = sibling_nodes.head; cell; cell = cell->next)
 						{
 							/* skip local node check, we did that above */
 							if (cell->node_info->node_id == local_node_info.node_id)
@@ -993,7 +1026,7 @@ monitor_streaming_standby(void)
 							follow_new_primary(follow_node_id);
 						}
 					}
-					clear_node_info_list(&standby_nodes);
+					clear_node_info_list(&sibling_nodes);
 				}
 			}
 		}
@@ -1395,12 +1428,12 @@ monitor_streaming_witness(void)
 				get_active_sibling_node_records(local_conn,
 												local_node_info.node_id,
 												local_node_info.upstream_node_id,
-												&standby_nodes);
+												&sibling_nodes);
 
-				if (standby_nodes.node_count > 0)
+				if (sibling_nodes.node_count > 0)
 				{
-					log_debug("scanning %i node records to detect new primary...", standby_nodes.node_count);
-					for (cell = standby_nodes.head; cell; cell = cell->next)
+					log_debug("scanning %i node records to detect new primary...", sibling_nodes.node_count);
+					for (cell = sibling_nodes.head; cell; cell = cell->next)
 					{
 						/* skip local node check, we did that above */
 						if (cell->node_info->node_id == local_node_info.node_id)
@@ -1430,7 +1463,7 @@ monitor_streaming_witness(void)
 						witness_follow_new_primary(follow_node_id);
 					}
 				}
-				clear_node_info_list(&standby_nodes);
+				clear_node_info_list(&sibling_nodes);
 			}
 		}
 loop:
@@ -1516,8 +1549,15 @@ loop:
 static bool
 do_primary_failover(void)
 {
+	ElectionResult election_result;
+
+	/*
+	 * Double-check status of the local connection
+	 */
+	check_connection(&local_node_info, &local_conn);
+
 	/* attempt to initiate voting process */
-	ElectionResult election_result = do_election();
+	election_result = do_election();
 
 	/* TODO add pre-event notification here */
 	failover_state = FAILOVER_STATE_UNKNOWN;
@@ -1531,7 +1571,7 @@ do_primary_failover(void)
 	}
 	else if (election_result == ELECTION_WON)
 	{
-		if (standby_nodes.node_count > 0)
+		if (sibling_nodes.node_count > 0)
 		{
 			log_notice("this node is the winner, will now promote itself and inform other nodes");
 		}
@@ -1576,7 +1616,7 @@ do_primary_failover(void)
 				get_active_sibling_node_records(local_conn,
 												local_node_info.node_id,
 												upstream_node_info.node_id,
-												&standby_nodes);
+												&sibling_nodes);
 
 			}
 			else if (config_file_options.failover == FAILOVER_MANUAL)
@@ -1638,10 +1678,10 @@ do_primary_failover(void)
 	{
 		case FAILOVER_STATE_PROMOTED:
 			/* notify former siblings that they should now follow this node */
-			notify_followers(&standby_nodes, local_node_info.node_id);
+			notify_followers(&sibling_nodes, local_node_info.node_id);
 
 			/* we no longer care about our former siblings */
-			clear_node_info_list(&standby_nodes);
+			clear_node_info_list(&sibling_nodes);
 
 			/* pass control back down to start_monitoring() */
 			log_info(_("switching to primary monitoring mode"));
@@ -1655,10 +1695,10 @@ do_primary_failover(void)
 			 * notify siblings that they should resume following the original
 			 * primary
 			 */
-			notify_followers(&standby_nodes, upstream_node_info.node_id);
+			notify_followers(&sibling_nodes, upstream_node_info.node_id);
 
 			/* we no longer care about our former siblings */
-			clear_node_info_list(&standby_nodes);
+			clear_node_info_list(&sibling_nodes);
 
 			/* pass control back down to start_monitoring() */
 			log_info(_("resuming standby monitoring mode"));
@@ -1918,7 +1958,7 @@ do_upstream_standby_failover(void)
 	 * completes, so poll for a while until we get a connection.
 	 */
 
-	for (i = 0; i < config_file_options.standby_reconnect_timeout; i++)
+	for (i = 0; i < config_file_options.repmgrd_standby_startup_timeout; i++)
 	{
 		local_conn = establish_db_connection(local_node_info.conninfo, false);
 
@@ -1927,7 +1967,7 @@ do_upstream_standby_failover(void)
 
 		log_debug("sleeping 1 second; %i of %i attempts to reconnect to local node",
 				  i + 1,
-				  config_file_options.standby_reconnect_timeout);
+				  config_file_options.repmgrd_standby_startup_timeout);
 		sleep(1);
 	}
 
@@ -2033,10 +2073,10 @@ promote_self(void)
 		return FAILOVER_STATE_PROMOTION_FAILED;
 	}
 
-	/* the presence of either of this command has been established already */
+	/* the presence of this command has been established already */
 	promote_command = config_file_options.promote_command;
 
-	log_debug("promote command is:\n  \"%s\"",
+	log_info(_("promote_command is:\n  \"%s\""),
 			  promote_command);
 
 	if (log_type == REPMGR_STDERR && *config_file_options.log_file)
@@ -2368,7 +2408,7 @@ follow_new_primary(int new_primary_id)
 	 * completes, so poll for a while until we get a connection.
 	 */
 
-	for (i = 0; i < config_file_options.standby_reconnect_timeout; i++)
+	for (i = 0; i < config_file_options.repmgrd_standby_startup_timeout; i++)
 	{
 		local_conn = establish_db_connection(local_node_info.conninfo, false);
 
@@ -2377,7 +2417,7 @@ follow_new_primary(int new_primary_id)
 
 		log_debug("sleeping 1 second; %i of %i attempts to reconnect to local node",
 				  i + 1,
-				  config_file_options.standby_reconnect_timeout);
+				  config_file_options.repmgrd_standby_startup_timeout);
 		sleep(1);
 	}
 
@@ -2543,6 +2583,7 @@ do_election(void)
 
 	/* we're visible */
 	int			visible_nodes = 1;
+	int			total_nodes = 0;
 
 	NodeInfoListCell *cell = NULL;
 
@@ -2593,14 +2634,16 @@ do_election(void)
 	get_active_sibling_node_records(local_conn,
 									local_node_info.node_id,
 									upstream_node_info.node_id,
-									&standby_nodes);
+									&sibling_nodes);
+
+	total_nodes = sibling_nodes.node_count + 1;
 
 	log_debug("do_election(): primary location is %s", upstream_node_info.location);
 
 	local_node_info.last_wal_receive_lsn = InvalidXLogRecPtr;
 
 	/* fast path if no other standbys (or witness) exists - normally win by default */
-	if (standby_nodes.node_count == 0)
+	if (sibling_nodes.node_count == 0)
 	{
 		if (strncmp(upstream_node_info.location, local_node_info.location, MAXLEN) == 0)
 		{
@@ -2628,7 +2671,7 @@ do_election(void)
 	}
 	else
 	{
-		/* standby nodes found - check if we're in the primary location befor checking theirs */
+		/* standby nodes found - check if we're in the primary location before checking theirs */
 		if (strncmp(upstream_node_info.location, local_node_info.location, MAXLEN) == 0)
 		{
 			primary_location_seen = true;
@@ -2643,7 +2686,7 @@ do_election(void)
 	/* pointer to "winning" node, initially self */
 	candidate_node = &local_node_info;
 
-	for (cell = standby_nodes.head; cell; cell = cell->next)
+	for (cell = sibling_nodes.head; cell; cell = cell->next)
 	{
 		/* assume the worst case */
 		cell->node_info->node_status = NODE_STATUS_UNKNOWN;
@@ -2698,7 +2741,7 @@ do_election(void)
 			candidate_node = cell->node_info;
 		}
 		/* LSN is same - tiebreak on priority, then node_id */
-		else if(cell->node_info->last_wal_receive_lsn == candidate_node->last_wal_receive_lsn)
+		else if (cell->node_info->last_wal_receive_lsn == candidate_node->last_wal_receive_lsn)
 		{
 			log_verbose(LOG_DEBUG, "node %i has same LSN as current candidate %i",
 						cell->node_info->node_id,
@@ -2750,9 +2793,9 @@ do_election(void)
 
 	log_debug("visible nodes: %i; total nodes: %i",
 			  visible_nodes,
-			  standby_nodes.node_count);
+			  total_nodes);
 
-	if (visible_nodes <= (standby_nodes.node_count / 2.0))
+	if (visible_nodes <= (total_nodes / 2.0))
 	{
 		log_notice(_("unable to reach a qualified majority of nodes"));
 		log_detail(_("node will enter degraded monitoring state waiting for reconnect"));

repmgrd.c

@@ -35,8 +35,10 @@
 
 static char *config_file = NULL;
 static bool verbose = false;
-static char *pid_file = NULL;
-static bool daemonize = false;
+static char pid_file[MAXPGPATH];
+static bool daemonize = true;
+static bool show_pid_file = false;
+static bool no_pid_file = false;
 
 t_configuration_options config_file_options = T_CONFIGURATION_OPTIONS_INITIALIZER;
 
@@ -99,8 +101,10 @@ main(int argc, char **argv)
 		{"config-file", required_argument, NULL, 'f'},
 
 /* daemon options */
-		{"daemonize", no_argument, NULL, 'd'},
+		{"daemonize", optional_argument, NULL, 'd'},
 		{"pid-file", required_argument, NULL, 'p'},
+		{"show-pid-file", no_argument, NULL, 's'},
+		{"no-pid-file", no_argument, NULL, OPT_NO_PID_FILE},
 
 /* logging options */
 		{"log-level", required_argument, NULL, 'L'},
@@ -113,8 +117,6 @@ main(int argc, char **argv)
 
 	set_progname(argv[0]);
 
-	srand(time(NULL));
-
 	/* Disallow running as root */
 	if (geteuid() == 0)
 	{
@@ -128,6 +130,10 @@ main(int argc, char **argv)
 		exit(1);
 	}
 
+	srand(time(NULL));
+
+	memset(pid_file, 0, MAXPGPATH);
+
 	while ((c = getopt_long(argc, argv, "?Vf:L:vdp:m", long_options, &optindex)) != -1)
 	{
 		switch (c)
@@ -169,11 +175,22 @@ main(int argc, char **argv)
 				/* daemon options */
 
 			case 'd':
-				daemonize = true;
+				if (optarg != NULL)
+				{
+					daemonize = parse_bool(optarg, "-d/--daemonize", &cli_errors);
+				}
 				break;
 
 			case 'p':
-				pid_file = optarg;
+				strncpy(pid_file, optarg, MAXPGPATH);
+				break;
+
+			case 's':
+				show_pid_file = true;
+				break;
+
+			case OPT_NO_PID_FILE:
+				no_pid_file = true;
 				break;
 
 				/* logging options */
@@ -220,7 +237,7 @@ main(int argc, char **argv)
 	/* Exit here already if errors in command line options found */
 	if (cli_errors.head != NULL)
 	{
-		exit_with_cli_errors(&cli_errors);
+		exit_with_cli_errors(&cli_errors, NULL);
 	}
 
 	startup_event_logged = false;
@@ -239,6 +256,58 @@ main(int argc, char **argv)
 	 */
 	load_config(config_file, verbose, false, &config_file_options, argv[0]);
 
+	/* Determine pid file location, unless --no-pid-file supplied */
+
+	if (no_pid_file == false)
+	{
+		if (config_file_options.repmgrd_pid_file[0] != '\0')
+		{
+			if (pid_file[0] != '\0')
+			{
+				log_warning(_("\"repmgrd_pid_file\" will be overridden by --pid-file"));
+			}
+			else
+			{
+				strncpy(pid_file, config_file_options.repmgrd_pid_file, MAXPGPATH);
+			}
+		}
+
+		/* no pid file provided - determine location */
+		if (pid_file[0] == '\0')
+		{
+			/* packagers: if feasible, patch PID file path into "package_pid_file" */
+			char		package_pid_file[MAXPGPATH] = "";
+
+			if (package_pid_file[0] != '\0')
+			{
+				maxpath_snprintf(pid_file, "%s", package_pid_file);
+			}
+			else
+			{
+				const char *tmpdir = getenv("TMPDIR");
+
+				if (!tmpdir)
+					tmpdir = "/tmp";
+
+				maxpath_snprintf(pid_file, "%s/repmgrd.pid", tmpdir);
+			}
+		}
+	}
+	else
+	{
+		/* --no-pid-file supplied - overwrite any value provided with --pid-file ... */
+		memset(pid_file, 0, MAXPGPATH);
+	}
+
+
+	/* If --show-pid-file supplied, output the location (if set) and exit */
+
+	if (show_pid_file == true)
+	{
+		printf("%s\n", pid_file);
+		exit(SUCCESS);
+	}
+
 
 	/* Some configuration file items can be overriden by command line options */
 
@@ -414,7 +483,7 @@ main(int argc, char **argv)
 		daemonize_process();
 	}
 
-	if (pid_file != NULL)
+	if (pid_file[0] != '\0')
 	{
 		check_and_create_pid_file(pid_file);
 	}
@@ -669,6 +738,8 @@ show_help(void)
 {
 	printf(_("%s: replication management daemon for PostgreSQL\n"), progname());
 	puts("");
+	printf(_("%s monitors a cluster of servers and optionally performs failover.\n"), progname());
+	puts("");
 
 	printf(_("Usage:\n"));
 	printf(_("  %s [OPTIONS]\n"), progname());
@@ -688,12 +759,14 @@ show_help(void)
 
 	puts("");
 
-	printf(_("General configuration options:\n"));
-	printf(_("  -d, --daemonize           detach process from foreground\n"));
-	printf(_("  -p, --pid-file=PATH       write a PID file\n"));
+	printf(_("Daemon configuration options:\n"));
+	printf(_("  -d, --daemonize[=true/false]\n"));
+	printf(_("                            detach process from foreground (default: true)\n"));
+	printf(_("  -p, --pid-file=PATH       use the specified PID file\n"));
+	printf(_("  -s, --show-pid-file       show PID file which would be used by the current configuration\n"));
+	printf(_("  --no-pid-file             don't write a PID file\n"));
 	puts("");
 
-	printf(_("%s monitors a cluster of servers and optionally performs failover.\n"), progname());
 }
 
 
@@ -802,7 +875,7 @@ terminate(int retval)
 {
 	logger_shutdown();
 
-	if (pid_file)
+	if (pid_file[0] != '\0')
 	{
 		unlink(pid_file);
 	}

repmgrd.h

@@ -10,6 +10,8 @@
 #include <time.h>
 #include "portability/instr_time.h"
 
+#define OPT_NO_PID_FILE                  1000
+
 extern volatile sig_atomic_t got_SIGHUP;
 extern MonitoringState monitoring_state;
 extern instr_time degraded_monitoring_start;
@@ -26,4 +28,6 @@ const char *print_monitoring_state(MonitoringState monitoring_state);
 
 void		update_registration(PGconn *conn);
 void		terminate(int retval);
+
+
 #endif							/* _REPMGRD_H_ */