Another attempt to fix the pgpool recovery scripts.

defaults/main.yml

@@ -19,7 +19,8 @@ psql_db_size_c: 170000000
 psql_listen_on_ext_int: False
 psql_use_alternate_data_dir: False
 # Deb/Ubuntu
-psql_data_dir: '/var/lib/postgresql/{{ psql_version }}/main'
+psql_data_root_dir: '/var/lib/postgresql/{{ psql_version }}'
+psql_data_dir: '{{ psql_data_root_dir }}/main'
 psql_conf_dir: '/etc/postgresql/{{ psql_version }}/main'
 psql_log_dir: /var/log/postgresql
 

templates/pgpool_recovery_stage_1.j2

@@ -5,17 +5,15 @@
 TS=$(date +%Y-%m-%d_%H-%M-%S)
 MASTER_HOST=$(hostname -f)
 # $1 is {{ psql_data_dir }} while  $3 is {{ psql_data_dir }}
-MASTER_DATA=$1
+#MASTER_DATA="{{ psql_data_dir }}"
-MASTER_DATA_BASE_DIR={{ psql_data_dir }}
+MASTER_DATA="${1}"
 RECOVERY_TARGET=${2}
-RECOVERY_DATA=${3}
+#RECOVERY_DATA="{{ psql_data_dir }}"
-RECOVERY_DATA_BASE_DIR={{ psql_data_dir }}
+RECOVERY_DATA="${3}"
 
 logger "pgpool_recovery_1: MASTER_DATA=$MASTER_DATA"
 logger "pgpool_recovery_1: RECOVERY_DATA=$RECOVERY_DATA"
 
-archive_dir={{ psql_wal_archiving_log_dir }}
-
 # Ensure that postgres is shut down on the target node
 ssh -T $RECOVERY_TARGET sudo /etc/init.d/postgresql stop
 
@@ -23,21 +21,20 @@ ssh -T $RECOVERY_TARGET sudo /etc/init.d/postgresql stop
 ssh -T $RECOVERY_TARGET \
     "[ -d $RECOVERY_DATA ] && mv $RECOVERY_DATA $RECOVERY_DATA.$TS"
 
-# If the archive dir is not empty the backup fails
-ssh -T $RECOVERY_TARGET \
-    "[ -d $archive_dir ] && rm $archive_dir/*"
-
 # We only use archived WAL logs during recoveries, so delete all
 # logs from the last recovery to limit the growth.
-rm $archive_dir/*
+rm $MASTER_DATA/archive_log/*
 
 # With this file present, our archive_command will actually
 # archive WAL files.
-touch $archive_dir/backup_in_progress
+touch $MASTER_DATA/archive_log/backup_in_progress
 
-psql -c "select pg_start_backup('pgpool-recovery')" postgres
+# Perform a backup of the database.
-echo "restore_command = 'scp $MASTER_HOST:$archive_dir/%f %p'" > $RECOVERY_DATA/recovery.conf
+ssh -T $RECOVERY_TARGET \
-tar -C $MASTER_DATA_BASE_DIR -zcf pgsql.tar.gz main
+    "pg_basebackup -h $MASTER_HOST -D $RECOVERY_DATA --xlog"
-psql -c 'select pg_stop_backup()' postgres
+
-scp pgsql.tar.gz $RECOVERY_TARGET:$RECOVERY_DATA_BASE_DIR
+# Configure the restore_command to use the archive_log WALs we’ll copy
+# over in 2nd_stage.sh.
+echo "restore_command = 'cp $RECOVERY_DATA/archive_log/%f %p'" | \
+    ssh -T $RECOVERY_TARGET "cat > $RECOVERY_DATA/recovery.conf"
 

templates/pgpool_recovery_stage_2.j2

@@ -1,20 +1,20 @@
 #!/bin/bash
 # Online recovery 2nd stage script
 #
-MASTER_DATA_BASE_DIR={{ psql_data_dir }}
+#MASTER_DATA="{{ psql_data_dir }}"
-MASTER_DATA=${1}
+MASTER_DATA="${1}"
 RECOVERY_TARGET=${2}
-RECOVERY_DATA_BASE_DIR={{ psql_data_dir }}
+#RECOVERY_DATA="{{ psql_data_dir }}"
-RECOVERY_DATA=${3}
+RECOVERY_DATA="${3}"
-port={{ psql_db_port }}        # PostgreSQL port number
 
-archive_dir={{ psql_wal_archiving_log_dir }}
+port="{{ psql_db_port }}"        # PostgreSQL port number
+
+archive_dir="{{ psql_wal_archiving_log_dir }}"
 
 logger "pgpool_recovery_2: MASTER_DATA=$MASTER_DATA"
 logger "pgpool_recovery_2: RECOVERY_DATA=$RECOVERY_DATA"
 logger "pgpool_recovery_2: archive_dir=$archive_dir"
 
-
 # Force to flush current value of sequences to xlog
 psql -p $port -t -c 'SELECT datname FROM pg_database WHERE NOT datistemplate AND datallowconn' template1|
 while read i
@@ -27,11 +27,11 @@ done
 # Flush all transactions to disk. Since pgpool stopped all connections,
 # there cannot be any data that does not reside on disk until the
 # to-be-recovered host is back on line.
-psql -p $port -c "SELECT pgpool_switch_xlog('$archive_dir')" template1
+psql -p $port -c "SELECT pgpool_switch_xlog('$MASTER_DATA/archive_log')" template1
 
 # Copy over all archive logs at once.
-rsync -avx --delete $archive_dir/ \
+rsync -avx --delete $MASTER_DATA/archive_log/ \
-    $RECOVERY_TARGET:$archive_dir/
+    $RECOVERY_TARGET:$RECOVERY_DATA/archive_log/
 
 # Delete the flag file to disable WAL archiving again.
-rm $archive_dir/backup_in_progress
+rm $MASTER_DATA/archive_log/backup_in_progress

templates/pgpool_remote_start.j2

@@ -3,6 +3,6 @@ DEST=$1
 DESTDIR=$2
 
 # Deploy a base backup
-ssh -T $DEST 'cd {{ psql_data_dir }}; tar zxf pgsql.tar.gz' 2>/dev/null 1>/dev/null < /dev/null
+ssh -T $DEST 'cd {{ psql_data_root_dir }}; tar zxf pgsql.tar.gz' 2>/dev/null 1>/dev/null < /dev/null
 # Startup PostgreSQL server
 ssh -T $DEST sudo /etc/init.d/postgresql start 2>/dev/null 1>/dev/null < /dev/null