Update version string to 3.0.1

This is similar to pg_basebackup's '-c/--checkpoint=fast|spread'
option.

FAQ.md

@@ -90,6 +90,23 @@ General
 
   This option is only available when using the `--rsync-only` option.
 
+- How can I make the witness server use a particular port?
+
+  By default the witness server is configured to use port 5499; this
+  is intended to support running the witness server as  a separate
+  instance on a normal node server, rather than on its own dedicated server.
+
+  To specify a port for the witness server, supply the port number to
+  repmgr with the `-l/--local-port` command line option.
+
+- Do I need to include `shared_preload_libraries = 'repmgr_funcs'`
+  in `postgresql.conf` if I'm not using `repmgrd`?
+
+  No, the `repmgr_funcs` library is only needed when running `repmgrd`.
+  If you later decide to run `repmgrd`, you just need to add
+  `shared_preload_libraries = 'repmgr_funcs'` and restart PostgreSQL.
+
+
 `repmgrd`
 ---------
 
@@ -102,7 +119,7 @@ General
 
 - How can I prevent a node from ever being promoted to master?
 
-  In `rempgr.conf`, set its priority to a value of 0 or less.
+  In `repmgr.conf`, set its priority to a value of 0 or less.
 
 - Does `repmgrd` support delayed standbys?
 

HISTORY

@@ -1,4 +1,11 @@
-3.0
+3.0.1   2015-04-16
+        Prevent repmgrd from looping infinitely if node was not registered (Ian)
+        When promoting a standby, have repmgr (not repmgrd) handle metadata updates (Ian)
+        Re-use replication slot if it already exists (Ian)
+        Prevent a test SSH connection being made when not needed (Ian)
+        Correct monitoring table column names (Ian)
+
+3.0     2015-03-27
         Require PostgreSQL 9.3 or later (Ian)
         Use `pg_basebackup` by default (instead of `rsync`) to clone standby servers (Ian)
         Use `pg_ctl promote` to promote a standby to primary
@@ -11,6 +18,11 @@
         General usability and logging message improvements (Ian)
         Code consolidation and cleanup (Ian)
 
+2.0.3   2015-04-16
+        Add -S/--superuser option for witness database creation Ian)
+        Add -c/--fast-checkpoint option for cloning (Christoph)
+        Add option "--initdb-no-pwprompt" (Ian)
+
 2.0.2   2015-02-17
         Add "--checksum" in rsync when using "--force" (Jaime)
         Use createdb/createuser instead of psql (Jaime)

PACKAGES.md

@@ -98,7 +98,7 @@ packages installed::
 
   sudo apt-get install libxslt-dev libxml2-dev libpam-dev libedit-dev
 
-If your using Debian packages for PostgreSQL and are building repmgr with the
+If you're using Debian packages for PostgreSQL and are building repmgr with the
 USE_PGXS option you also need to install the corresponding development package::
 
   sudo apt-get install postgresql-server-dev-9.0

QUICKSTART.md

@@ -71,7 +71,10 @@ Standby setup
         [2015-03-03 18:18:23] [NOTICE] HINT: You can now start your postgresql server
         [2015-03-03 18:18:23] [NOTICE] for example : pg_ctl -D  /path/to/standby/data start
 
-  Note that at this point it does not matter if the `repmgr.conf` file is not found.
+  Note that the `repmgr.conf` file is not required when cloning a standby.
+  However we recommend providing a valid `repmgr.conf` if you wish to use
+  replication slots, or want `repmgr` to log the clone event to the
+  `repl_events` table.
 
   This will clone the PostgreSQL database files from the master, including its
   `postgresql.conf` and `pg_hba.conf` files, and additionally automatically create

RHEL/repmgrd.init

@@ -1,89 +1,114 @@
-#!/bin/bash
+#!/bin/sh
 #
-# repmgrd	Start up the repmgrd daemon
-#		repmrgd (replication manager daemon)
-#
-# chkconfig: -  75 16
-# description: repmgrd is the repliation manager daemon \
-# The repmgrd replication management and monitoring daemon for PostgreSQL.
-
-### BEGIN INIT INFO
-# Provides: repmgrd
-# Required-Start:  $local_fs $remote_fs $network $syslog postgresql
-# Required-Stop:     $local_fs $remote_fs $network $syslog postgresql
-# Should-Start:	     $syslog postgresql-9.3
-# Should-Stop: $syslog postgresql-9.3
-# Short-Description: start and stop repmrgd
-# Description: Enable repmgrd replication management and monitoring daemon for PostgreSQL
-#	       this is used to monitor a postgresql cluster.
-### END INIT INFO
+# chkconfig: - 75 16
+# description: Enable repmgrd replication management and monitoring daemon for PostgreSQL
+# processname: repmgrd
+# pidfile="/var/run/${NAME}.pid"
 
 # Source function library.
-. /etc/init.d/functions
+INITD=/etc/rc.d/init.d
+. $INITD/functions
 
-# Source networking configuration.
+# Get function listing for cross-distribution logic.
+TYPESET=`typeset -f|grep "declare"`
+
+# Get network config.
 . /etc/sysconfig/network
 
-prog=repmgrd
-REPMGRD_ENABLED=yes
+DESC="PostgreSQL replication management and monitoring daemon"
+NAME=repmgrd
+
+REPMGRD_ENABLED=no
 REPMGRD_OPTS=
 REPMGRD_USER=postgres
-DAEMONIZE="-d"
+REPMGRD_BIN=/usr/pgsql-9.3/bin/repmgrd
+REPMGRD_PIDFILE=/var/run/repmgrd.pid
+REPMGRD_LOCK=/var/lock/subsys/${NAME}
+REPMGRD_LOG=/var/lib/pgsql/9.3/data/pg_log/repmgrd.log
 
-# pull in sysconfig settings
-[ -f /etc/sysconfig/repmgrd ] && . /etc/sysconfig/repmgrd
+# Read configuration variable file if it is present
+[ -r /etc/sysconfig/$NAME ] && . /etc/sysconfig/$NAME
 
-LOCKFILE=/var/lock/subsys/$prog
-RETVAL=0
+# For SELinux we need to use 'runuser' not 'su'
+if [ -x /sbin/runuser ]
+then
+    SU=runuser
+else
+    SU=su
+fi
+
+test -x $REPMGRD_BIN || exit 0
 
 case "$REPMGRD_ENABLED" in
     [Yy]*)
-	#nothing to do here	
+	break
 	;;
     *)
-	exit 2
+	exit 0
 	;;
 esac
 
 
-if [ -z "$REPMGRD_OPTS" ]
+if [ -z "${REPMGRD_OPTS}" ]
 then
-    echo "Not starting $prog, REPMGRD_OPTS not set in /etc/sysconfig/$prog"
-    exit 2
+    echo "Not starting ${NAME}, REPMGRD_OPTS not set in /etc/sysconfig/${NAME}"
+    exit 0
 fi
 
-start() {
-	[ "$EUID" != "0" ] && exit 4
-        [ "$NETWORKING" = "no" ] && exit 1
+start()
+{
+    REPMGRD_START=$"Starting ${NAME} service: "
 
-        # Start daemons.
-        echo -n $"Starting $prog: "
-        daemon --user $REPMGRD_USER $prog $DAEMONIZE $REPMGRD_OPTS
-        RETVAL=$?
+    # Make sure startup-time log file is valid
+    if [ ! -e "${REPMGRD_LOG}" -a ! -h "${REPMGRD_LOG}" ]
+    then
+        touch "${REPMGRD_LOG}" || exit 1
+        chown ${REPMGRD_USER}:postgres "${REPMGRD_LOG}"
+        chmod go-rwx "${REPMGRD_LOG}"
+        [ -x /sbin/restorecon ] && /sbin/restorecon "${REPMGRD_LOG}"
+    fi
+
+    echo -n "${REPMGRD_START}"
+    $SU -l $REPMGRD_USER -c "${REPMGRD_BIN} ${REPMGRD_OPTS} -p ${REPMGRD_PIDFILE} &" >> "${REPMGRD_LOG}" 2>&1 < /dev/null
+    sleep 2
+    pid=`head -n 1 "${REPMGRD_PIDFILE}" 2>/dev/null`
+    if [ "x${pid}" != "x" ]
+    then
+        success "${REPMGRD_START}"
+        touch "${REPMGRD_LOCK}"
+        echo $pid > "${REPMGRD_PIDFILE}"
         echo
-        [ $RETVAL -eq 0 ] && touch $LOCKFILE
-        return $RETVAL
+    else
+        failure "${REPMGRD_START}"
+        echo
+        script_result=1
+    fi
 }
 
-stop() {
-	[ "$EUID" != "0" ] && exit 4
-        echo -n $"Shutting down $prog: "
-        killproc $prog
-        RETVAL=$?
-        echo
-        [ $RETVAL -eq 0 ] && rm -f $LOCKFILE
-        return $RETVAL
-}
-status() {
-        if [ -f "$LOCKFILE" ]; then
-            echo "$prog is running"
+stop()
+{
+    echo -n $"Stopping ${NAME} service: "
+    if [ -e "${REPMGRD_LOCK}" ]
+    then
+        killproc ${NAME}
+        ret=$? 
+        if [ $ret -eq 0 ]
+        then
+            echo_success
+            rm -f "${REPMGRD_PIDFILE}"
+            rm -f "${REPMGRD_LOCK}"
         else
-	    RETVAL=3
-            echo "$prog is stopped"
+            echo_failure
+            script_result=1
         fi
-        return $RETVAL
+    else
+        # not running; per LSB standards this is "ok"   
+        echo_success
+    fi
+    echo
 }
 
+
 # See how we were called.
 case "$1" in
   start)
@@ -93,22 +118,16 @@ case "$1" in
         stop
         ;;
   status)
-        status $prog
+        status -p $REPMGRD_PIDFILE $NAME
+        script_result=$?
         ;;
-  restart|force-reload)
+  restart)
         stop
-        start
-        ;;
-  try-restart|condrestart)
-        if status $prog > /dev/null; then
-            stop
-            start
-        fi
-        ;;
-  reload)
-        exit 3
+	start
         ;;
   *)
-        echo $"Usage: $0 {start|stop|status|restart|try-restart|force-reload}"
+        echo $"Usage: $0 {start|stop|status|restart}"
         exit 2
 esac
+
+exit $script_result

RHEL/repmgrd.sysconfig

@@ -1,4 +1,21 @@
-#default sysconfig file for repmrgd
-#custom overrides can be placed here
+# default settings for repmgrd. This file is source by /bin/sh from
+# /etc/init.d/repmgrd
 
-REPMGRD_OPTS="-f /etc/repmgr/repmgr.conf"
+# disable repmgrd by default so it won't get started upon installation
+# valid values: yes/no
+REPMGRD_ENABLED=no
+
+# Options for repmgrd (required)
+#REPMGRD_OPTS="--verbose -d -f /var/lib/pgsql/repmgr/repmgr.conf"
+
+# User to run repmgrd as
+#REPMGRD_USER=postgres
+
+# repmgrd binary
+#REPMGRD_BIN=/usr/bin/repmgr
+
+# pid file
+#REPMGRD_PIDFILE=/var/lib/pgsql/repmgr/repmgrd.pid
+
+# log file
+#REPMGRD_LOG=/var/lib/pgsql/repmgr/repmgrd.log

TODO

@@ -5,6 +5,8 @@ Known issues in repmgr
   the database server using the ``pg_ctl`` command may accidentally
   terminate after their associated ssh session ends.
 
+* PGPASSFILE may not be passed to pg_basebackup
+
 Planned feature improvements
 ============================
 

dbutils.c

@@ -197,7 +197,7 @@ is_pgup(PGconn *conn, int timeout)
 
 
 /*
- * Return the id of the active master node, or -1 if no
+ * Return the id of the active master node, or NODE_NOT_FOUND if no
  * record available.
  *
  * This reports the value stored in the database only and
@@ -224,12 +224,12 @@ get_master_node_id(PGconn *conn, char *cluster)
 	{
 		log_err(_("get_master_node_id(): query failed\n%s\n"),
 				PQerrorMessage(conn));
-		retval = -1;
+		retval = NODE_NOT_FOUND;
 	}
 	else if (PQntuples(res) == 0)
 	{
 		log_warning(_("get_master_node_id(): no active primary found\n"));
-		retval = -1;
+		retval = NODE_NOT_FOUND;
 	}
 	else
 	{
@@ -511,7 +511,7 @@ get_master_connection(PGconn *standby_conn, char *cluster,
 
 	if(master_id != NULL)
 	{
-		*master_id = -1;
+		*master_id = NODE_NOT_FOUND;
 	}
 
 	/* find all nodes belonging to this cluster */
@@ -728,6 +728,49 @@ create_replication_slot(PGconn *conn, char *slot_name)
 	char		sqlquery[QUERY_STR_LEN];
 	PGresult   *res;
 
+	/*
+	 * Check whether slot exists already; if it exists and is active, that
+	 * means another active standby is using it, which creates an error situation;
+	 * if not we can reuse it as-is
+	 */
+
+	sqlquery_snprintf(sqlquery,
+					  "SELECT active, slot_type "
+                      "  FROM pg_replication_slots "
+					  " WHERE slot_name = '%s' ",
+					  slot_name);
+
+	res = PQexec(conn, sqlquery);
+	if (!res || PQresultStatus(res) != PGRES_TUPLES_OK)
+	{
+		log_err(_("unable to query pg_replication_slots: %s\n"),
+				PQerrorMessage(conn));
+		PQclear(res);
+		return false;
+	}
+
+	if(PQntuples(res))
+	{
+		if(strcmp(PQgetvalue(res, 0, 1), "physical") != 0)
+		{
+			log_err(_("Slot '%s' exists and is not a physical slot\n"),
+					slot_name);
+			PQclear(res);
+		}
+		if(strcmp(PQgetvalue(res, 0, 0), "f") == 0)
+		{
+			PQclear(res);
+			log_debug(_("Replication slot '%s' exists but is inactive; reusing\n"),
+						slot_name);
+
+			return true;
+		}
+		PQclear(res);
+		log_err(_("Slot '%s' already exists as an active slot\n"),
+				slot_name);
+		return false;
+	}
+
 	sqlquery_snprintf(sqlquery,
 					  "SELECT * FROM pg_create_physical_replication_slot('%s')",
 					  slot_name);
@@ -1051,7 +1094,12 @@ create_event_record(PGconn *conn, t_configuration_options *options, int node_id,
 	bool		success = true;
 	struct tm	ts;
 
-	if(conn != NULL)
+	/* Only attempt to write a record if a connection handle was provided/
+	   Also check that the repmgr schema has been properly intialised - if
+	   not it means no configuration file was provided, which can happen with
+	   e.g. `repmgr standby clone`, and we won't know which schema to write to.
+	 */
+	if(conn != NULL && strcmp(repmgr_schema, DEFAULT_REPMGR_SCHEMA_PREFIX) != 0)
 	{
 		int n_node_id = htonl(node_id);
 		char *t_successful = successful ? "TRUE" : "FALSE";

repmgr.c

@@ -77,6 +77,7 @@ static bool write_recovery_file_line(FILE *recovery_file, char *recovery_file_pa
 static void check_master_standby_version_match(PGconn *conn, PGconn *master_conn);
 static int	check_server_version(PGconn *conn, char *server_type, bool exit_on_error, char *server_version_string);
 static bool check_upstream_config(PGconn *conn, int server_version_num, bool exit_on_error);
+static bool update_node_record_set_master(PGconn *conn, int this_node_id);
 
 static char *make_pg_path(char *file);
 
@@ -139,11 +140,11 @@ main(int argc, char **argv)
 		{"verbose", no_argument, NULL, 'v'},
 		{"pg_bindir", required_argument, NULL, 'b'},
 		{"rsync-only", no_argument, NULL, 'r'},
+		{"fast-checkpoint", no_argument, NULL, 'c'},
 		{"initdb-no-pwprompt", no_argument, NULL, 1},
 		{"check-upstream-config", no_argument, NULL, 2},
 		{"recovery-min-apply-delay", required_argument, NULL, 3},
-		{"fast-checkpoint", no_argument, NULL, 4},
-		{"ignore-external-config-files", no_argument, NULL, 5},
+		{"ignore-external-config-files", no_argument, NULL, 4},
 		{NULL, 0, NULL, 0}
 	};
 
@@ -174,7 +175,7 @@ main(int argc, char **argv)
 	/* Prevent getopt_long() from printing an error message */
 	opterr = 0;
 
-	while ((c = getopt_long(argc, argv, "d:h:p:U:S:D:l:f:R:w:k:FWIvr:b:", long_options,
+	while ((c = getopt_long(argc, argv, "d:h:p:U:S:D:l:f:R:w:k:FWIvb:r:c", long_options,
 							&optindex)) != -1)
 	{
 		switch (c)
@@ -239,6 +240,9 @@ main(int argc, char **argv)
 			case 'r':
 				runtime_options.rsync_only = true;
 				break;
+			case 'c':
+				runtime_options.fast_checkpoint = true;
+				break;
 			case 1:
 				runtime_options.initdb_no_pwprompt = true;
 				break;
@@ -267,9 +271,6 @@ main(int argc, char **argv)
 				strncpy(runtime_options.recovery_min_apply_delay, optarg, MAXLEN);
 				break;
 			case 4:
-				runtime_options.fast_checkpoint = true;
-				break;
-			case 5:
 				runtime_options.ignore_external_config_files = true;
 				break;
 			default:
@@ -485,7 +486,7 @@ main(int argc, char **argv)
 	 */
 	if (config_file_required)
 	{
-		if (options.node == -1)
+		if (options.node == NODE_NOT_FOUND)
 		{
 			if(config_file_parsed == true)
 			{
@@ -925,7 +926,7 @@ do_standby_clone(void)
 	int			i;
 	bool		pg_start_backup_executed = false;
 	bool		target_directory_provided = false;
-	bool		config_file_copy_required = false;
+	bool		external_config_file_copy_required = false;
 
 	char		master_data_directory[MAXFILENAME];
 	char		local_data_directory[MAXFILENAME];
@@ -1101,7 +1102,7 @@ do_standby_clone(void)
 			if(strcmp(PQgetvalue(res, i, 2), "f") == 0)
 			{
 				config_file_outside_pgdata = true;
-				config_file_copy_required = true;
+				external_config_file_copy_required = true;
 				strncpy(master_config_file, PQgetvalue(res, i, 1), MAXFILENAME);
 			}
 		}
@@ -1110,7 +1111,7 @@ do_standby_clone(void)
 			if(strcmp(PQgetvalue(res, i, 2), "f") == 0)
 			{
 				hba_file_outside_pgdata  = true;
-				config_file_copy_required = true;
+				external_config_file_copy_required = true;
 				strncpy(master_hba_file, PQgetvalue(res, i, 1), MAXFILENAME);
 			}
 		}
@@ -1119,13 +1120,14 @@ do_standby_clone(void)
 			if(strcmp(PQgetvalue(res, i, 2), "f") == 0)
 			{
 				ident_file_outside_pgdata = true;
-				config_file_copy_required = true;
+				external_config_file_copy_required = true;
 				strncpy(master_ident_file, PQgetvalue(res, i, 1), MAXFILENAME);
 			}
 		}
 		else
 			log_warning(_("unknown parameter: %s\n"), PQgetvalue(res, i, 0));
 	}
+
 	PQclear(res);
 
 	/*
@@ -1316,12 +1318,6 @@ do_standby_clone(void)
 		}
 
 		PQclear(res);
-
-		/*
-		 * With rsync we'll need to explicitly copy configuration files in any
-		 * case
-		 */
-		config_file_copy_required = true;
 	}
 	else
 	{
@@ -1335,14 +1331,14 @@ do_standby_clone(void)
 	}
 
 	/*
-	 * If configuration files were not in the data directory, we need to copy
-	 * them via SSH
+	 * If configuration files were not inside the data directory, we;ll need to
+	 * copy them via SSH (unless `--ignore-external-config-files` was provided)
 	 *
 	 * TODO: add option to place these files in the same location on the
 	 * standby server as on the primary?
 	 */
 
-	if(config_file_copy_required == true)
+	if(external_config_file_copy_required && !runtime_options.ignore_external_config_files)
 	{
 		log_notice(_("copying configuration files from master\n"));
 		r = test_ssh_connection(runtime_options.host, runtime_options.remote_user);
@@ -1354,102 +1350,81 @@ do_standby_clone(void)
 			goto stop_backup;
 		}
 
-		if(strlen(master_config_file))
+		if(config_file_outside_pgdata)
 		{
-			if(runtime_options.ignore_external_config_files && config_file_outside_pgdata)
-			{
-				log_notice(_("standby clone: not copying master config file '%s'\n"), master_config_file);
-			}
-			else
-			{
-				log_info(_("standby clone: master config file '%s'\n"), master_config_file);
-				r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
-									  master_config_file, local_config_file, false, server_version_num);
-				if (r != 0)
-				{
-					log_warning(_("standby clone: failed copying master config file '%s'\n"),
-								master_config_file);
-					retval = ERR_BAD_SSH;
-					goto stop_backup;
-				}
-			}
-		}
-
-		if(strlen(master_hba_file))
-		{
-			if(runtime_options.ignore_external_config_files && hba_file_outside_pgdata)
-			{
-				log_notice(_("standby clone: not copying master config file '%s'\n"), master_hba_file);
-			}
-			else
-			{
-				log_info(_("standby clone: master hba file '%s'\n"), master_hba_file);
-				r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
-									  master_hba_file, local_hba_file, false, server_version_num);
-				if (r != 0)
-				{
-					log_warning(_("standby clone: failed copying master hba file '%s'\n"),
-								master_hba_file);
-					retval = ERR_BAD_SSH;
-					goto stop_backup;
-				}
-			}
-		}
-
-		if(strlen(master_ident_file))
-		{
-			if(runtime_options.ignore_external_config_files && ident_file_outside_pgdata)
-			{
-				log_notice(_("standby clone: not copying master config file '%s'\n"), master_ident_file);
-			}
-			else
-			{
-				log_info(_("standby clone: master ident file '%s'\n"), master_ident_file);
-				r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
-									  master_ident_file, local_ident_file, false, server_version_num);
-				if (r != 0)
-				{
-					log_warning(_("standby clone: failed copying master ident file '%s'\n"),
-								master_ident_file);
-					retval = ERR_BAD_SSH;
-					goto stop_backup;
-				}
-			}
-		}
-
-		/*
-		 * When using rsync, copy pg_control file last, emulating the base backup
-		 * protocol.
-		 */
-		if(runtime_options.rsync_only)
-		{
-			maxlen_snprintf(local_control_file, "%s/global", local_data_directory);
-
-			log_info(_("standby clone: local control file '%s'\n"),
-					 local_control_file);
-
-			if (!create_dir(local_control_file))
-			{
-				log_err(_("couldn't create directory %s ...\n"),
-						local_control_file);
-				goto stop_backup;
-			}
-
-			maxlen_snprintf(master_control_file, "%s/global/pg_control",
-							master_data_directory);
-			log_info(_("standby clone: master control file '%s'\n"),
-				 master_control_file);
+			log_info(_("standby clone: master config file '%s'\n"), master_config_file);
 			r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
-								  master_control_file, local_control_file,
-								  false, server_version_num);
+								  master_config_file, local_config_file, false, server_version_num);
 			if (r != 0)
 			{
-				log_warning(_("standby clone: failed copying master control file '%s'\n"),
-							master_control_file);
+				log_err(_("standby clone: failed copying master config file '%s'\n"),
+						master_config_file);
 				retval = ERR_BAD_SSH;
 				goto stop_backup;
 			}
 		}
+
+		if(hba_file_outside_pgdata)
+		{
+			log_info(_("standby clone: master hba file '%s'\n"), master_hba_file);
+			r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
+								  master_hba_file, local_hba_file, false, server_version_num);
+			if (r != 0)
+			{
+				log_err(_("standby clone: failed copying master hba file '%s'\n"),
+						master_hba_file);
+				retval = ERR_BAD_SSH;
+				goto stop_backup;
+			}
+		}
+
+		if(ident_file_outside_pgdata)
+		{
+			log_info(_("standby clone: master ident file '%s'\n"), master_ident_file);
+			r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
+								  master_ident_file, local_ident_file, false, server_version_num);
+			if (r != 0)
+			{
+				log_err(_("standby clone: failed copying master ident file '%s'\n"),
+						master_ident_file);
+				retval = ERR_BAD_SSH;
+				goto stop_backup;
+			}
+		}
+	}
+
+	/*
+	 * When using rsync, copy pg_control file last, emulating the base backup
+	 * protocol.
+	 */
+	if(runtime_options.rsync_only)
+	{
+		maxlen_snprintf(local_control_file, "%s/global", local_data_directory);
+
+		log_info(_("standby clone: local control file '%s'\n"),
+				 local_control_file);
+
+		if (!create_dir(local_control_file))
+		{
+			log_err(_("couldn't create directory %s ...\n"),
+					local_control_file);
+			goto stop_backup;
+		}
+
+		maxlen_snprintf(master_control_file, "%s/global/pg_control",
+						master_data_directory);
+		log_info(_("standby clone: master control file '%s'\n"),
+				 master_control_file);
+		r = copy_remote_files(runtime_options.host, runtime_options.remote_user,
+							  master_control_file, local_control_file,
+							  false, server_version_num);
+		if (r != 0)
+		{
+			log_warning(_("standby clone: failed copying master control file '%s'\n"),
+						master_control_file);
+			retval = ERR_BAD_SSH;
+			goto stop_backup;
+		}
 	}
 
 stop_backup:
@@ -1562,8 +1537,9 @@ do_standby_promote(void)
 	int			i,
 				promote_check_timeout  = 60,
 				promote_check_interval = 2;
-	bool		promote_sucess = false;
+	bool		promote_success = false;
 	bool        success;
+	PQExpBufferData details;
 
 	/* We need to connect to check configuration */
 	log_info(_("connecting to standby database\n"));
@@ -1637,50 +1613,57 @@ do_standby_promote(void)
 		retval = is_standby(conn);
 		if(!retval)
 		{
-			promote_sucess = true;
+			promote_success = true;
 			break;
 		}
 		sleep(promote_check_interval);
 	}
 
-	if (promote_sucess == false)
+	if (promote_success == false)
 	{
-		PQExpBufferData details;
-		initPQExpBuffer(&details);
-		appendPQExpBuffer(&details,
-						  "Node %i could not be promoted to master",
-						  options.node);
-
-		create_event_record(old_master_conn,
-							&options,
-							options.node,
-							"standby_promote",
-							false,
-							details.data);
-		/* XXX exit with error? */
 		log_err(_(retval == 1 ?
 			  "STANDBY PROMOTE failed, this is still a standby node.\n" :
 				  "connection to node lost!\n"));
+		exit(ERR_FAILOVER_FAIL);
 	}
-	else
+
+
+	/* update node information to reflect new status */
+	if(update_node_record_set_master(conn, options.node) == false)
 	{
-		PQExpBufferData details;
 		initPQExpBuffer(&details);
 		appendPQExpBuffer(&details,
-						  "Node %i was successfully promoted to master",
+						  _("unable to update node record for node %i"),
 						  options.node);
 
-		log_notice(_("STANDBY PROMOTE successful.  You should REINDEX any hash indexes you have.\n"));
-		/* Log the event */
-		create_event_record(conn,
+		log_err("%s\n", details.data);
+
+		create_event_record(NULL,
 							&options,
 							options.node,
-							"standby_promote",
-							true,
+							"repmgrd_failover_promote",
+							false,
 							details.data);
+
+		exit(ERR_DB_QUERY);
 	}
 
-	PQfinish(old_master_conn);
+
+	initPQExpBuffer(&details);
+	appendPQExpBuffer(&details,
+					  "Node %i was successfully promoted to master",
+					  options.node);
+
+	log_notice(_("STANDBY PROMOTE successful.  You should REINDEX any hash indexes you have.\n"));
+
+	/* Log the event */
+	create_event_record(conn,
+						&options,
+						options.node,
+						"standby_promote",
+						true,
+						details.data);
+
 	PQfinish(conn);
 
 	return;
@@ -2232,7 +2215,7 @@ help(const char *progname)
 	printf(_("  -p, --port=PORT                     database server port\n"));
 	printf(_("  -U, --username=USERNAME             database user name to connect as\n"));
 	printf(_("\nConfiguration options:\n"));
-	printf(_("  -b. --pg_bindir=PATH                path to PostgreSQL binaries (optional)\n"));
+	printf(_("  -b, --pg_bindir=PATH                path to PostgreSQL binaries (optional)\n"));
 	printf(_("  -D, --data-dir=DIR                  local directory where the files will be\n" \
 			 "                                      copied to\n"));
 	printf(_("  -l, --local-port=PORT               standby or witness server local port\n"));
@@ -2242,15 +2225,13 @@ help(const char *progname)
 			 "                                      (default: postgres)\n"));
 	printf(_("  -w, --wal-keep-segments=VALUE       minimum value for the GUC\n" \
 			 "                                      wal_keep_segments (default: %s)\n"), DEFAULT_WAL_KEEP_SEGMENTS);
-	printf(_("  -k, --keep-history=VALUE            keeps indicated number of days of\n" \
-			 "                                      history\n"));
-	printf(_("  -F, --force                         force potentially dangerous operations\n" \
-			 "                                      to happen\n"));
+	printf(_("  -k, --keep-history=VALUE            keeps indicated number of days of history\n"));
+	printf(_("  -F, --force                         force potentially dangerous operations to happen\n"));
 	printf(_("  -W, --wait                          wait for a master to appear\n"));
 	printf(_("  -r, --rsync-only                    use only rsync to clone a standby\n"));
+	printf(_("  -c, --fast-checkpoint               force fast checkpoint when cloning a standby\n"));
 	printf(_("  --recovery-min-apply-delay=VALUE    set recovery_min_apply_delay in recovery.conf\n" \
  			 "                                      when cloning a standby (PostgreSQL 9.4 and later)\n"));
-	printf(_("  --fast-checkpoint                   force fast checkpoint when cloning a standby\n"));
 	printf(_("  --ignore-external-config-files      don't copy configuration files located outside \n" \
 			 "                                      the data directory when cloning a standby\n"));
 	printf(_("  --initdb-no-pwprompt                don't require superuser password when running initdb\n"));
@@ -2440,7 +2421,7 @@ copy_remote_files(char *host, char *remote_user, char *remote_path,
 	 */
 	if (is_directory)
 	{
-		/* Files we don't want */
+		/* Files which we don't want */
 		appendPQExpBuffer(&rsync_flags, "%s",
 						  " --exclude=postmaster.pid --exclude=postmaster.opts --exclude=global/pg_control");
 
@@ -2454,11 +2435,11 @@ copy_remote_files(char *host, char *remote_user, char *remote_path,
 							  " --exclude=postgresql.auto.conf.tmp");
 		}
 
-		/* Temporary files we don't want, if they exist */
+		/* Temporary files which we don't want, if they exist */
 		appendPQExpBuffer(&rsync_flags, " --exclude=%s*",
 						  PG_TEMP_FILE_PREFIX);
 
-		/* Directories we don't want */
+		/* Directories which we don't want */
 		appendPQExpBuffer(&rsync_flags, "%s",
 						  " --exclude=pg_xlog/* --exclude=pg_log/* --exclude=pg_stat_tmp/*");
 
@@ -2701,7 +2682,6 @@ create_schema(PGconn *conn)
 	char		sqlquery[QUERY_STR_LEN];
 	PGresult   *res;
 
-
 	/* create schema */
 	sqlquery_snprintf(sqlquery, "CREATE SCHEMA %s", get_repmgr_schema_quoted(conn));
 	log_debug(_("master register: %s\n"), sqlquery);
@@ -3271,6 +3251,92 @@ check_upstream_config(PGconn *conn, int server_version_num, bool exit_on_error)
 }
 
 
+static bool
+update_node_record_set_master(PGconn *conn, int this_node_id)
+{
+	PGresult   *res;
+	char		sqlquery[QUERY_STR_LEN];
+
+	log_debug(_("Setting %i as master and marking existing master as failed\n"), this_node_id);
+
+	res = PQexec(conn, "BEGIN");
+
+	if (PQresultStatus(res) != PGRES_COMMAND_OK)
+	{
+		log_err(_("Unable to begin transaction: %s\n"),
+				PQerrorMessage(conn));
+
+		PQclear(res);
+		return false;
+	}
+
+	PQclear(res);
+
+	sqlquery_snprintf(sqlquery,
+					  "  UPDATE %s.repl_nodes "
+					  "     SET active = FALSE "
+					  "   WHERE cluster = '%s' "
+					  "     AND type = 'master' "
+					  "     AND active IS TRUE ",
+					  get_repmgr_schema_quoted(conn),
+					  options.cluster_name);
+
+	res = PQexec(conn, sqlquery);
+
+	if (PQresultStatus(res) != PGRES_COMMAND_OK)
+	{
+		log_err(_("Unable to set old master node as inactive: %s\n"),
+				PQerrorMessage(conn));
+		PQclear(res);
+
+		PQexec(conn, "ROLLBACK");
+		return false;
+	}
+
+	PQclear(res);
+
+	sqlquery_snprintf(sqlquery,
+					  "  UPDATE %s.repl_nodes "
+					  "     SET type = 'master', "
+					  "         upstream_node_id = NULL "
+					  "   WHERE cluster = '%s' "
+					  "     AND id = %i ",
+					  get_repmgr_schema_quoted(conn),
+					  options.cluster_name,
+					  this_node_id);
+
+	res = PQexec(conn, sqlquery);
+
+	if (PQresultStatus(res) != PGRES_COMMAND_OK)
+	{
+		log_err(_("Unable to set current node %i as active master: %s\n"),
+				this_node_id,
+				PQerrorMessage(conn));
+		PQclear(res);
+
+		PQexec(conn, "ROLLBACK");
+		return false;
+	}
+
+	PQclear(res);
+
+	res = PQexec(conn, "COMMIT");
+
+	if (PQresultStatus(res) != PGRES_COMMAND_OK)
+	{
+		log_err(_("Unable to set commit transaction: %s\n"),
+				PQerrorMessage(conn));
+		PQclear(res);
+
+		return false;
+	}
+
+	PQclear(res);
+
+	return true;
+}
+
+
 static void
 do_check_upstream_config(void)
 {
@@ -3307,7 +3373,6 @@ do_check_upstream_config(void)
 }
 
 
-
 static char *
 make_pg_path(char *file)
 {

repmgr.conf.sample

@@ -7,8 +7,11 @@
 #
 # repmgr and repmgrd require these items to be configured:
 
-# Cluster name
-cluster=test
+# Cluster name - this will be used by repmgr to generate its internal
+# schema (pattern: "repmgr_{cluster}"); while this name will be quoted
+# to preserve case, we recommend using lower case and avoiding whitespace
+# to facilitate easier querying of the repmgr views and tables.
+cluster=example_cluster
 
 # Node ID and name
 # (Note: we recommend to avoid naming nodes after their initial

repmgr.h

@@ -49,6 +49,7 @@
 
 #define MANUAL_FAILOVER		0
 #define AUTOMATIC_FAILOVER	1
+#define NODE_NOT_FOUND      -1
 #define NO_UPSTREAM_NODE	-1
 
 

repmgrd.c

@@ -91,7 +91,6 @@ static void witness_monitor(void);
 static bool check_connection(PGconn *conn, const char *type);
 static bool set_local_node_failed(void);
 
-static bool update_node_record_set_master(PGconn *conn, int this_node_id, int old_master_node_id);
 static bool update_node_record_set_upstream(PGconn *conn, int this_node_id, int new_upstream_node_id);
 
 static void update_shared_memory(char *last_wal_standby_applied);
@@ -284,9 +283,17 @@ main(int argc, char **argv)
 		terminate(ERR_BAD_CONFIG);
 	}
 
-	/* Retrieve record for this node from the database */
+	/* Retrieve record for this node from the local database */
 	node_info = get_node_info(my_local_conn, local_options.cluster_name, local_options.node);
 
+	/* No node record found - exit gracefully */
+	if(node_info.node_id == NODE_NOT_FOUND)
+	{
+		log_err(_("No metadata record found for this node - terminating\n"));
+		log_notice(_("HINT: was this node registered with 'repmgr (master|standby) register'?\n"));
+		terminate(ERR_BAD_CONFIG);
+	}
+
 	log_debug("node id is %i, upstream is %i\n", node_info.node_id, node_info.upstream_node_id);
 
 	/*
@@ -335,9 +342,9 @@ main(int argc, char **argv)
 				log_info(_("starting continuous master connection check\n"));
 
 				/*
-                                 * Check that master is still alive.
-                                 * XXX We should also check that the
-                                 * standby servers are sending info
+				 * Check that master is still alive.
+				 * XXX We should also check that the
+				 * standby servers are sending info
 				 */
 
 				/*
@@ -637,9 +644,9 @@ witness_monitor(void)
 	 */
 	sqlquery_snprintf(sqlquery,
 					  "INSERT INTO %s.repl_monitor "
-					  "           (master_node, standby_node, "
+					  "           (primary_node, standby_node, "
 					  "            last_monitor_time, last_apply_time, "
-					  "            last_wal_master_location, last_wal_standby_location, "
+					  "            last_wal_primary_location, last_wal_standby_location, "
 					  "            replication_lag, apply_lag )"
 					  "      VALUES(%d, %d, "
 					  "             '%s'::TIMESTAMP WITH TIME ZONE, NULL, "
@@ -993,9 +1000,9 @@ standby_monitor(void)
 	 */
 	sqlquery_snprintf(sqlquery,
 					  "INSERT INTO %s.repl_monitor "
-					  "           (master_node, standby_node, "
+					  "           (primary_node, standby_node, "
 					  "            last_monitor_time, last_apply_time, "
-					  "            last_wal_master_location, last_wal_standby_location, "
+					  "            last_wal_primary_location, last_wal_standby_location, "
 					  "            replication_lag, apply_lag ) "
 					  "      VALUES(%d, %d, "
 					  "             '%s'::TIMESTAMP WITH TIME ZONE, '%s'::TIMESTAMP WITH TIME ZONE, "
@@ -1434,25 +1441,6 @@ do_master_failover(void)
 		/* and reconnect to the local database */
 		my_local_conn = establish_db_connection(local_options.conninfo, true);
 
-		/* update node information to reflect new status */
-		if(update_node_record_set_master(my_local_conn, node_info.node_id, failed_master.node_id) == false)
-		{
-			appendPQExpBuffer(&event_details,
-							  _("unable to update node record for node %i (promoted to master following failure of node %i)"),
-							  node_info.node_id,
-							  failed_master.node_id);
-
-			log_err("%s\n", event_details.data);
-
-			create_event_record(NULL,
-								&local_options,
-								node_info.node_id,
-								"repmgrd_failover_promote",
-								false,
-								event_details.data);
-
-			terminate(ERR_DB_QUERY);
-		}
 
 		/* update internal record for this node */
 		node_info = get_node_info(my_local_conn, local_options.cluster_name, local_options.node);
@@ -1749,7 +1737,7 @@ set_local_node_failed(void)
 {
 	PGresult   *res;
 	char		sqlquery[QUERY_STR_LEN];
-	int			active_master_node_id = -1;
+	int			active_master_node_id = NODE_NOT_FOUND;
 	char		master_conninfo[MAXLEN];
 
 	if (!check_connection(master_conn, "master"))
@@ -2089,7 +2077,7 @@ update_registration(void)
 
 		log_err("%s\n", errmsg.data);
 
-		create_event_record(my_local_conn,
+		create_event_record(master_conn,
 							&local_options,
 							local_options.node,
 							"repmgrd_shutdown",
@@ -2234,7 +2222,7 @@ get_node_info(PGconn *conn, char *cluster, int node_id)
 	char		sqlquery[QUERY_STR_LEN];
 	PGresult   *res;
 
-	t_node_info node_info = {-1, NO_UPSTREAM_NODE, "", InvalidXLogRecPtr, UNKNOWN, false, false};
+	t_node_info node_info = { NODE_NOT_FOUND, NO_UPSTREAM_NODE, "", InvalidXLogRecPtr, UNKNOWN, false, false};
 
 	sprintf(sqlquery,
 			"SELECT id, upstream_node_id, conninfo, type, slot_name, active "
@@ -2260,7 +2248,7 @@ get_node_info(PGconn *conn, char *cluster, int node_id)
 
 		log_err("%s\n", errmsg.data);
 
-		create_event_record(my_local_conn,
+		create_event_record(NULL,
 							&local_options,
 							local_options.node,
 							"repmgrd_shutdown",
@@ -2274,7 +2262,7 @@ get_node_info(PGconn *conn, char *cluster, int node_id)
 	if (!PQntuples(res)) {
 		log_warning(_("No record found record for node %i\n"), node_id);
 		PQclear(res);
-		node_info.node_id = -1;
+		node_info.node_id = NODE_NOT_FOUND;
 		return node_info;
 	}
 
@@ -2313,93 +2301,6 @@ parse_node_type(const char *type)
 }
 
 
-static bool
-update_node_record_set_master(PGconn *conn, int this_node_id, int old_master_node_id)
-{
-	PGresult   *res;
-	char		sqlquery[QUERY_STR_LEN];
-
-	log_debug(_("Setting failed node %i inactive; marking node %i as master\n"), old_master_node_id, this_node_id);
-
-	res = PQexec(conn, "BEGIN");
-
-	if (PQresultStatus(res) != PGRES_COMMAND_OK)
-	{
-		log_err(_("Unable to begin transaction: %s\n"),
-				PQerrorMessage(conn));
-
-		PQclear(res);
-		return false;
-	}
-
-	PQclear(res);
-
-	sqlquery_snprintf(sqlquery,
-					  "  UPDATE %s.repl_nodes "
-					  "     SET active = FALSE "
-					  "   WHERE cluster = '%s' "
-					  "     AND id = %i ",
-					  get_repmgr_schema_quoted(conn),
-					  local_options.cluster_name,
-					  old_master_node_id);
-
-	res = PQexec(conn, sqlquery);
-
-	if (PQresultStatus(res) != PGRES_COMMAND_OK)
-	{
-		log_err(_("Unable to set old master node %i as inactive: %s\n"),
-				old_master_node_id,
-				PQerrorMessage(conn));
-		PQclear(res);
-
-		PQexec(conn, "ROLLBACK");
-		return false;
-	}
-
-	PQclear(res);
-
-	sqlquery_snprintf(sqlquery,
-					  "  UPDATE %s.repl_nodes "
-					  "     SET type = 'master', "
-					  "         upstream_node_id = NULL "
-					  "   WHERE cluster = '%s' "
-					  "     AND id = %i ",
-					  get_repmgr_schema_quoted(conn),
-					  local_options.cluster_name,
-					  this_node_id);
-
-	res = PQexec(conn, sqlquery);
-
-	if (PQresultStatus(res) != PGRES_COMMAND_OK)
-	{
-		log_err(_("Unable to set current node %i as active master: %s\n"),
-				this_node_id,
-				PQerrorMessage(conn));
-		PQclear(res);
-
-		PQexec(conn, "ROLLBACK");
-		return false;
-	}
-
-	PQclear(res);
-
-	res = PQexec(conn, "COMMIT");
-
-	if (PQresultStatus(res) != PGRES_COMMAND_OK)
-	{
-		log_err(_("Unable to set commit transaction: %s\n"),
-				PQerrorMessage(conn));
-		PQclear(res);
-
-		return false;
-	}
-
-	PQclear(res);
-
-	return true;
-}
-
-
 static bool
 update_node_record_set_upstream(PGconn *conn, int this_node_id, int new_upstream_node_id)
 {

version.h

@@ -1,6 +1,6 @@
 #ifndef _VERSION_H_
 #define _VERSION_H_
 
-#define REPMGR_VERSION "3.0dev"
+#define REPMGR_VERSION "3.0.1"
 
 #endif